University of Southern California Dissertations and Theses

Suburban Political Behavior In Los Angeles County, California

(USC Thesis Other)

Suburban Political Behavior In Los Angeles County, California

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content This dissertation has been microfilmed exactly as received 66— 11,561 BELL, Charles Gordon, 1929- SUBURBAN POLITICAL BEHAVIOR IN LOS ANGELES COUNTY, CALIFORNIA. U niversity of Southern California, Ph.D., 1966 Political Science, general University Microfilms, Inc., Ann Arbor, Michigan C op yrigh t by CHARLES GORDON B E L L 1966 SUBURBAN POLITICAL BEHAVIOR IN LOS ANGELES COUNTY, CALIFORNIA by Charles Gordon Bell A Dissertation Presented to the FACULTY OF THE GRADUATE SCHOOL UNIVERSITY OF SOUTHERN CALIFORNIA In Partial Fulfillment of the Requirements for the Degree DOCTOR OF PHILOSOPHY (Political Science) June 1966 UNIVERSITY O F SO U T H E R N CALIFORNIA THE GRADUATE SCHOOL UNIVERSITY PARK LOS ANGELES, CALIFORNIA 9 0 0 0 7 This dissertation, written by under the direction of hia.....Dissertation Com mittee, and approved by all its members, has been presented to and accepted by the Graduate School, in partial fulfillment of requirements for the degree of .Gha.r.ls.a.. .G.otr.dojx. JBell. D O C T O R OF P H I L O S O P H Y Dean Date J.Unea..1.9.6.6. DISSERTATION XOM M ITTEE 'hair man TABLE OF CONTENTS Page LIST OF TABLES........................................... vi Chapter I. INTRODUCTION .................................... 1 The Problem and Setting Republican suburbia Is suburbia Republican? Pivotal suburbia The Dimensions Suburbia— a new phenomenon? Sub-urban and sub-culture Suburban growth Metropolitan Government Urbanism The urban schism Metropolitan governments Why resistance? Suburbs and local government Suburban Political Behavior Complex phenomena Chameleon or segregation Structure of the Dissertation The Data Description of data Limits of the data II. THEORY........................................... 66 The Theory of Class ii Chapter Page III. The objective criteria The subjective criteria Culture and class The Theory of Social Mobility Urbanization Residential mobility Occupational mobility Aspiration and motivation Style of life Impact on the individual The Theory of Status Consistency Multiple status dimensions Cross pressures Status inconsistency behavior The Theory of the Individual Within the Group Group defined The primary and secondary groups Reference group theory Patterns of group membership Social reality and norms Conformity and changes Group Political Behavior Primary and secondary reference groups Cross' pressures Attitude stability Some limits Class Political Behavior Trends in class-politics Objective class and political behavior Subjective class and political behavior Conclusions SUBURBIA .................................... Suburbia Defined Previous definitions 176 iii Chapter IV. The Three Definitive Variables Dwelling units owner-occupied. Dwelling units recently built Single unit structures The suburban typology The Relationship Between Variables The county sample Summary for the County Owner-occupancy Age of housing Single unit structures Summary for Suburbia Owner-occupancy Age of housing Single unit structures Summary for Anti-Suburbia Owner-occupancy Age of housing Single unit structures The Effects of Community SUBURBAN POLITICAL BEHAVIOR .... Introduction The Political Variables Partisan voting data Non-partisan elections Ballot propositions Participation The Party Vote Population, Housing, and J'oiitics High family income Professional occupation iv Page 21 8 Chapter Page Politics, Housing, and Population The Political Impact of Three • Selected Variables Median value of property High family income Professional occupation Introduction Is Suburbia Republican? Roots of Suburban Political Behavior Criteria The Two Suburbs Class in Suburbia Property values Occupation and income Residential Mobility Housing and mobility Residential mobility and partisan politics Income Occupation Long-term stability and political behavior Short-term stability and political behavior Summary Needed Research V. CONCLUSIONS AND SUMMARY 266 APPENDICES 302 APPENDIX A APPENDIX B BIBLIOGRAPHY 320 v LIST OF TABLES Table Page 1. Type of Suburb and Region.................... 15 2. Growth Data for Principal Standard Metropolitan Areas ........................ 18 3. Suburban Attitudes Toward the Center City................................. 26 4. Forms of Suburban Government by Forms of Election................................. 35 5. Social and Economic Characteristics of Suburbs by Form of Government............. 35 6. Suburban Party Identification ................ 41 7. Probable Vote, Congress 1954 42 8. Republican Pluralities in Three Suburbs, 1920 and 1952 ............................... 49 9. Political Affiliation of Immigrants to Vestport by Tear of Entry.................. 55 10. Self-Identification by Objective Social Class, 1952 and 1956 70 11. Criteria for C l a s s ........................... 71 12. Father's Income and Student's Occupational Choice...................................... 75 13. Subjective Class by Strata .................. 82 14. Distribution of Occupations by Size of Community................................... 91 15. Inter-Generational Social Mobility by Size of Community-of-Orientation ......... 91 vi Table Page 16. Group Membership by Voting Rate.............. 143 17. Voting Rate by Group Membership ....... 143 18. Party Identification by Class by Region . . .■ 146 19. Perception of Party Best Serving Groups, Party by Groups............................. 149 20. Perception of Party Best Serving Own Group, Party by Groups............................ 149 21. Perception of Party by Preference of Various Groups, Party by Groups ......... 150 22. Perception of Party Preference of Various Groups, Party by Group Members and by Non-Members................................. 1 52 23. The Class Ascribed to Persons Reputed to Believe That "Strikes Should Be Completely Forbidden by Law"........................... 1 54 24. The Class Ascribed to Persons Reputed to Believe That "Government Should Guarantee Employment and Decent Standard of Living to A l l " ............................. 154 25. Party Identification by Occupational Strata by Income........................... 157 26. Conservatism-Radicalism by Occupational Strata...................................... 166 27. Conservatism-Radicalism by Class ............ 166 28. Conservatism-Radicalism by Occupational Strata by Class............................. 167 29. Party Identification by Subjective- Objective Class Patterns.................. 171 30. Presidential Vote by Subjective-Objective Class Patterns............................. 171 31. Issue Attitude by Subjective-Objective Class Patterns............................. 172 vii Table 32. 33. 34. 35. 36. 37. 38. 39. 40. 41 . 42. Preliminary Analysis of Four Variables to Be Used to Define the Suburban Areas of Los Angeles County ...................... Correlations Between the Four Preliminary Defining Variables ........................ Median Values for Fourteen Selected Variables by the High and Low Decile of Dwelling Units Occupied (A)........................... Median Values for Fourteen Selected Variables by the Low and High Decile of Housing Units Recently Built (B—1) ................ Median Values for Fourteen Selected Variables by the High and Low Decile of Single Unit Structures (C) ......................... Median Values for Twelve Selected Variables by the Suburban and Anti-Suburban Typology .................. ................ Correlation Between Owner Occupancy (A) and Fifteen Selected Variables by Three Community Typologies ................ Correlation Between Housing Built 1950- 1960 (B—1) and Fifteen Selected Variables by Three Community Typologies ................................. Correlation Between Housing Built 1940- 1950 (B-2) and Fifteen Selected Variables by Three Community Typologies ................................. Correlation Between Housing Built Before 1940 (B-3) and Fifteen Selected Variables by Three Community Typologies ................................. Correlation Between Single Unit Structures (C) and Fifteen Selected Variables by Three Community Typologies ................ Page 181 ! 1 8 2 186 188 190 192 i 194 195 196 197 198 viii Table Page 43. Correlation Between Median Value of Property (D) and Fifteen Selected Variables by Three Community Typologies ................ 199 44.. Correlation Between Residency Same As in 1955 (E-1) and Fifteen Selected Variables by Three Community Typologies................................. 200 45. Correlation Between Moved Into House, 1958- 1960 (E-2) and Fifteen Selected Variables by Three Community Typologies ...... 201 46. Correlation Between Residence Elsewhere in 1955, Los Angeles (E-3) and Fifteen Selected Variables by Three Community Typologies . .................. 202 47. Correlation Between Residence Elsewhere in 1955, California (E-4) and Fifteen Selected Variables by Three Community Typologies ...................... 203 48. Correlation Between Family Income, $3,000- $10,000 (F-1) and Fifteen Selected Variables by Three Community Typologies................................. 204 49. Correlation Between Family Income, $10,000 Plus (F-2) and Fifteen Selected Variables by Three Community Typologies ............ 205 50. Correlation Between Occupation, Professional (G-1) and Fifteen Selected Variables by Three Community Typologies ................ 206 51. Correlation Between Occupation, Managerial (G-2) and Fifteen Selected Variables by Three Community Typologies ............... 207 52. Correlation Between Race, Vhite (H—1) and Fifteen Selected Variables by Three Community Typologies ...................... 208 53. Correlation Between Race, Negro (H—2) and Fifteen Selected Variables by Three Community Typologies ...................... 209 ix Table Page 54. The Relationship Between Variables by Three Community Typologies ................. 214 55. Changes in Sign for Statistically Significant Correlations by Three Community Typologies . 216 56. Median Values for Selected Political Variables...................................... 226 57. Correlation Between Owner-Occupancy (A) and Fifteen Selected Political Variables by Three Community Typologies .................. 229 58. Correlation Between Housing Built 1950-1960 (B—1) and Fifteen Selected Political Variables by Three Community Typologies...................................... 230 59. Correlation Between Single Unit Structures (C) and Fifteen Selected Political Variables by Three Community Typologies ............. 231 60. Correlation Between Median Value of Property (D) and Fifteen Selected Political Variables by Three Community Typologies...................................... 232 61. Correlation Between Residency Same in 1955 (E-1) and Fifteen Selected Political Variables by Three Community Typologies...................................... 233 62. Correlation Between Moved Into House, 1958- 1960 (E-2) and Fifteen Selected Political Variables by Three Community Typologies . . 234 63. Correlation Between Residence Elsewhere in 1955, Los Angeles (E-3) and Fifteen Selected Political Variables by Three Community Typologies .................. 235 64. Correlation Between Family Income, $3,000- $10,000 (F-1) and Fifteen Selected Political Variables by Three Community Typologies ........................ 236 x Table Page 65. Correlation Between Family Income, $10,000 Plus (F-2) and Fifteen Selected Political Variables by Three Community Typologies . ...................... 237 66. Correlation Between Occupation, Professional (G-1) and Fifteen Selected Political Variables by Three Community Typologies.................................... 238 67. Correlation Between Occupation, Managerial (G-2) and Fifteen Selected Political Variables by Three Community Typologies.................................... 239 68. Changes in the Correlation Between Housing Built 1950-1960 and Certain Political Variables by Three Community Typologies.................................... 241 69. Correlation Between Democratic Registration, September 1958, and Selected Political, Housing, and Population Variables by Three Community Typologies .................. 243 70. Correlation Between Democratic Registration, September 1960, and Selected Political, Housing, and Population Variables by Three Community Typologies .................. 244 71. Correlation Between Democratic Registration, September 1962, and Selected Political, Housing, and Population Variables by Three Community Typologies .................. 245 72. Correlation Between Brown Vote, 1958, and Selected Political, Housing, and Popu lation Variables by Three Community Typologies.................................... 246 73. Correlation Between Betts Vote, 1958, and Selected Political, Housing, and Popu lation Variables by Three Community Typologies.................................... 247 xi Table Page 74. Correlation Between Kennedy Vote, 1960, and Selected Political, Housing, and Popula tion Variables by Three Community Typologies................................. 248 75. Correlation Between Brown Vote, 1962, and Selected Political, Housing, and Popula tion Variables by Three Community Typologies........................ 249 76. Correlation Between Rose Vote, 1962, and Selected Political, Housing, and Population Variables by Three Community Typologies ...................... 250 77. Correlation Between Gibson Vote, 1962, and Selected Political, Housing, and Popula tion Variables by Three Community Typologies................................. 251 78. Correlation Between "Yes" Vote on Subversives Control (Proposition No. 24) and Selected Political, Housing, and Population Variables by Three Community Typologies................................. 252 79. Correlation Between "Yes" Vote on State Water Bonds (Proposition No. 1) and Selected Political, Housing, and Population Variables by Three Community Typologies................................. 253 80. Correlation Between Democratic Turn-out, 1960, and Selected Political, Housing, and Popula tion Variables by Three Community Typologies................................. 254 81. Correlation Between Republican Turn-out, 1960, and Selected Political, Housing, and Popula tion Variables by Three Community Typologies................................. 255 82. Correlation Between Democratic Turn-out, 1962, and Selected Political, Housing, and Popula tion Variables by Three Community Typologies................................. 256 xii Table Page 83. Correlation Between Republican Turn-out, 1962, and Selected Political, Housing, and Popula tion Variables by Three Community Typologies...................................... 257 84. Distribution of Value of Property for the Suburban and Anti-Suburban Tracts ......... 258 85. A Comparison of Suburban and Anti-Suburban Political Behavior with Value of Property Held Constant ...................... 260 86. Distribution of High Family Income for the Suburban and Anti-Suburban Tracts ..... 261 87. Distribution of Professional Occupation for the Suburban and Anti-Suburban Tracts . . . 262 88. A Comparison of Suburban and Anti-Suburban Political Behavior with High Income Held Constant..................................263 89. A Comparison of Suburban and Anti-Suburban Political Behavior with Professional Occupation Held Constant .................... 264 90. Suburban Party Identifiers .................... 267 91. Democratic Party Registration for Los Angeles County and Its Suburban Tracts................269 4 92. Democratic Vote in Los Angeles County and Its Suburban Tracts ............................. 270 93. Correlation Between Suburban Criteria and Selected Political Variables in Los Angeles County............................. 276 94. Correlation Between Suburban Criteria and Selected Political Variables in Los Angeles Suburbs........................... 276 95. Correlation Between Age of Housing and Property Value by County and Suburbs .... 277 96. Correlation Between Family Income and Age t of Housing by County and Suburbs..............278 xiii Table Page 97. Correlation Between Value of Property and Family Income by County and Suburbs .... 279 98. Distribution of Three Class Variables within Suburban Census Tracts .................... 281 99. Partisan Differences Between Suburbs and Anti-Suburbs ............................... 281 100. Partisan Differences Between Suburbs and Anti-Suburbs Holding Property Value Constant.................................... 282 101. Partisan Differences Between Suburbs and Anti-Suburbs Holding Occupation and Family Income Constant .................... 283 102. A Comparison of the Partisan Differences Between Suburbia and Anti-Suburbia and the Differences Between the T w o ........... 285 103. Residential Mobility and Partisan Behavior in Los Angeles County...................... 288 104. Residential Mobility and Partisan Behavior in Los Angeles Suburbs.................... 290 105. Residential Mobility by Suburb and Anti-Suburb............................ 290 106. Residential Mobility by Owner-Occupancy . . . 291 107. Residential Mobility by Single Unit Structures 291 108. Residential Mobility by Recently Built Housing 291 109. Correlation Between Residential Mobility and Family Income, $3,000-$10,000 by Three Community Typologies ...................... 293 110. Correlation Between Residential Mobility and Family Income, $10,000 Plus by Three Community Typologies ...................... 293 111. Correlation Between Residential Mobility and Professional Occupation by Three Community Typologies ...................... 294 xiv Table Page 112. Correlation Between Residential Mobility and Managerial Occupation by Three Community Typologies.................................... 295 113. Relation of Reported Changes in Own Occupa tional Status to Changes in Self-Identif ication i ith Parties, 1956 .................. 299 114. Pearsonian Correlation Between Pour State-Vide Democratic Candidates 1958, Democratic Regis tration 1958, All Los Angeles County Democratic Assembly Candidates 1958, Stevenson 1956, and Kennedy 1960. (Based on All Thirty-One Assembly Districts in Los Angeles County). . 304 115. Pearsonian Correlation Between Four State-Vide Democratic Candidates 1958, Democratic Regis tration 1958, and Fourteen Los Angeles County Assembly Candidates 1958, Stevenson 1956, and Kennedy 1960 by Assembly Districts. (Based on Fourteen Los Angeles County Assembly Districts with Democratic Incumbents Seeking Re-election) ......................... 305 116. Pearsonian Correlation Between Four State-Vide Democratic Candidates 1958, Democratic Regis tration 1958, Eleven Los Angeles County Demo cratic Assembly Candidates 1958, Stevenson 1956, and Kennedy 1960. (Based on the Eleven Los Angeles County Assembly Districts in Which There was a Democratic Incumbent Seeking Re-Election Vho Had Not Successfully Cross- Filed in the Primary)......................... 306 117. Pearsonian Correlation Between Four State-Vide Democratic Candidates 1958, Democratic Regis tration 1958, Fifteen Los Angeles County Demo cratic Assembly Candidates 1958, Stevenson 1956, and Kennedy 1960. (Based on Fifteen Los Angeles County Assembly Districts with Republican Incumbents Seeking Re-Election) . 307 118. Comparison of City and Suburban Presidential Vote in Thirteen Selected Areas, 1952-1960 . 308 119. Relation of Type of Community to Party Identification ................................ 310 xv Table Page 120. Suburban Vote 1948 by Area . . ................ 311 121. Suburban Vote 1952 by Area........................ 311 122. Southern Voting Patterns in Central City and Suburbs, 1948-1956 ...................... 312 123. Voter Party Preference, 1956 Presidential Election......................................... 313 124. Voter Candidate Preference, 1956 Presidential Election......................................... 314 125. Occupation by Community of Orientation .... 315 126. Percentage of Time Spent in Present Occupational Category......................316 127. Percentage of Time Spent in Other Than Present Occupational Category ............. 317 xvi CHAPTER I INTRODUCTION The Problem and Setting In both professional literature and popular media changing spatial and functional patterns of metropolitan living are alleged to have had a direct influence upon the political system and political beliefs. Republican suburbia Republican. After the 1952 Presidential Election, Jake Arvey of Chicago is reported to have remarked that the Democrats were beaten in the suburbs. Senator Robert Taft made the prediction that until Democrats were able to win in the suburbs the Republicans would be unbeatable. Lou Harris cited election trends in three New York City suburbs as demonstrating the increasing suburban Republican vote.^ An A.F. of L.-C.I.O. "Political Memo" (circa January 1961) reported that while John Kennedy won thirteen Popular political myth suggests that suburbs are 1 Louis Harris, Is There (New York: Harper and Brothers, 1 metropolitan areas in 1960, Richard Nixon won seven of the adjacent suburban areas. Another study of the 1960 election, by the Republican National Committee*s Research Division, attributed Nixon's defeat to vote losses in suburban areas. (It should be noted, however, that the 1956—1960 decline in the Republican presidential suburban vote was just about the same as their 1956-1960 loss in the city vote). Two years later, according to a subsequent report, the Republican suburban congressional vote had increased two and one-half percentage points over 3 comparable figures for 1958. Edward C. Banfield, in a recent article, attributes the difficulties of establishing metropolitan governments to the fact that suburbs are Republican while central 4 cities are Democratic. G. Edward Janosik cites the fact that out of twenty-four congressional districts wholly or largely within twenty suburbs, seventeen (71 per cent) were captured by Republicans in 1952, while two years later sixteen (67 per cent) were again won by the 2 Research Division,.Republican National Committee. "The 1960 Election" (Washington, D.C.: mimeographed, 1961), p. 20. 3 Research Division, Republican National Committee, "The 1962 Elections" (Washington, D.C.; Mimeographed, 1963), pp. 28-29. 4 Edward C. Banfield, "The Politics of Metropolitan Area Organization," Midwest Journal of Political Science. I, 1 (May, 1957), p. 86. 5 Republicans. V. 0. Key, after examining suburban vote patterns, was unwilling to assign any fixed political behavior to suburbia— pointing out that there are many £ kinds of suburban communities. In Wisconsin, Leon Epstein reports that between 1948 and 1954 (four gubernatorial elections), an appreciable dif ference in vote patterns among cities seemed to correlate to 7 city size. Writing in 1958, Fred I. Greenstein and Raymond E. Wolfinger concluded that there is a greater Republican vote in suburbs than in central cities (holding education, 8 income, and occupation constant), Janosik, moreover, points out that prior to the recent great influx of popula- 9 tion (post World War II), many suburbs were Democratic. If Epstein, Greenstein, Wolfinger, and Janosik are correct, then there would seem to be some substantial support for the position that the suburbs, as such, have given rise to a new center of Republican party strength, ^G. Edward Janosik, "The New Suburbia," Current History, XXXI, 180 (August, 1956), p. 92. ^V. 0. Key, Politics. Parties, and Pressure Groups (4th ed.; New York: Thomas Y. Crowell Company, 1960), pp. 275-78. 7 Leon D. Epstein, "Size of Place and the Division of the Two-Party Vote in Wisconsin," Western Political Quarterly. IX, 1 (March, 1956), p. 143. 8 Fred I. Greenstein and Raymond E. Wolfinger, "The Suburbs and Shifting Party Loyalties," Public Opinion Quarterly. XXII, 4 (Winter, 1958-59), p. 477. 9 Janosik, op. cit., p. 92 4 However, Bennett M. Berger in his analysis of a recently formed working class suburb found that workers continued to vote Democratic as they had when they were central city residents.^ And Thomas R. Pye, in his doctoral dissertation, found that suburbia was not homo geneous, but socially, economically, and politically diverse. Samuel Lubell, in a pre-election syndicated series, reported in 1962 that occupation and family back- 1 2 ground continue to divide suburban political loyalties. Granting that there are Democrats residing in suburbia, is it possible that the suburban environment may be a factor in itself, producing a greater than "expected” Republican vote? Is suburbia Republican? Robert C. Vood, in his general consideration of suburban life, discusses several factors which may account for the apparent suburban Republican voting behavior. He offers both "conversion" (also known as "chameleon effect"), ^Bennett M. Berger, Working—Class Suburb (Berkeley and Los Angeles: University of California Press, 1960), p. 110. 11 Thomas Roy Dye, "Certain Political Correlates of the Areal Differentiation and Stratification of Sub populations in a Metropolitan Area," (unpublished Ph.D. dissertation, Department of Political Science, Pennsylvania University, 1961), Chapter II, passim (microfilm). 1 2 Samuel Lubell, "Class, Party, Divide Suburban Voters," Los Angeles Times. October 24, 1962. and "transfer" (most of those who move to the suburbs were already Republicans) which suggests that the move to sub- 1 3 urbia functions as a screen. Wood's most cogent observa tion in regard to suburban voting behavior is that the . . . lack of detailed information is especially evident as far as the patterns of suburban political behavior are concerned. . . .14 Most often, the theories of observers rest upon either some implied but unstated assumptions or only vague generalizations as to what constitutes suburbia and the identity of "Republican vote." Greenstein and Wolfinger, typically, define suburbia as a dormitory community lying on the edge of a metropolitan area from which a large 1 5 number of residents commute to work in the central city. Suburbia is quite often defined in such economic terms as would normally preclude a Democratic vote of any appreciable size regardless of location. Thus, John R. Seeley, R. Alexander Sim, and Elizabeth W. Loosley describe the suburban area which they studied as an elite dormitory 1 6 separated' from inferior housing and industry. Yet, it is 1 3 Robert C. Wood, Suburbia, Its People and Their Politics (Boston: Houghton Mifflin Company, 1959), pT 140. Ibid., p. vi. 1 5 Greenstein and Wolfinger, loc. cit. 1 6 John R. Seeley, R. Alexander Sim, and Elizabeth W. Loosley, Crestwood Heights (New York: Basic Books, Inc., 1956), p. 38. 6 quite possible to define suburbia in terras which do not carry partisan connotations. For example, on the basis of four factors: 1. percentage of home ownership; 2. percentage of dwellings recently constructed; 3. recent rate of population increase; and 4. percentage of population in family units; five "suburban" areas were located in the Los Angeles area 1 7 which were Democratic by political registration. Definitions of the "Republican vote" are most often translated in terms of the most recent vote for President— a dubious equation. For example, Berger uses the Presi dential votes of 1948, 1952, and 1956 to test the political 1 8 behavior of his respondents. However, a correlation of votes cast for President, Governor and other statewide offices, United States Senate, State Assembly, and political registration in Los Angeles County, indicates that the greatest caution should be exercised in equating any particular vote with "party strength." For example, the Pearsonian correlation between the 1960 Democratic Presidential vote and 1958 Democratic registration by Assembly districts in Los Angeles County was .927, but the correlation between the same Presidential vote and the 1958 Democratic State assembly vote by Assembly districts in 17 Downey, Lakewood, Lawndale-Wiseburn, Montebello- Pico, and Norwalk-Artesia. 1 8 Berger, op. cit.. p. 110. 1 9 Los Angeles was .298. Similarly, the Pearsonian for the 1958 Democratic Gubernatorial vote and the 1958 Democratic registration was .991 but for the Gubernatorial and Democratic Assembly vote was .295. Pivotal suburbia The national suburban population (as defined by the Bureau of the Census) is the fastest growing in the United States (compared to the rural and central city populations). Comprising almost one-third of the total national popula tion, and largely located in those states having the greatest number of electoral votes, suburban residents may well be of crucial political importance in the future. Thus, in its 1960 Report, the Republican National Committee saw the suburbs as Nixon's stumbling block, reporting that in the nineteen suburbs examined, the Kennedy vote increased over the Stevenson vote in every one; while the Nixon vote compared to the Eisenhower vote decreased in nine, increased in one, and gained less than Kennedy's gain in nine 20 (absolute figures). The Republican Party is not alone in its concern and interest in the suburban vote. A paper delivered in 1 9 Pearsonian correlations calculated for the thirty- one Assembly districts in Los Angeles County by use of Library Program LECORR, University of Southern California's Computer Sciences Center. See Appendix A, Tables 114-117. 20 "The 1960 Election," op. cit.. pp. 18-19. 1961 by George M. Belknap (on the staff of the Democratic National Committee at the time) examined the political impact of suburban social and residential mobility. The argument that upward economic mobility during prosperous times has removed the economic class attitudes of the 1930’s overlooks one important fact; class feelings are relative. . . . Fortunately, for Democratic fortunes in 1960 . . . we did not assume that the person whose improved economic condition allowed him to move to the suburbs had automatically become a Republican.21 It seems safe to assume that both parties will continue to weigh the suburban vote in their considerations and that serious research into the complex dimensions of suburban political behavior is well warranted. The Dimensions Suburbia— a new phenomenon? The growing awareness and concern with suburban political behavior has not occurred apart from a greater interest in metropolitan politics, governmental organiza- 22 tions, economics, and social structures. In recent 21 George M. Belknap, "Motivational Research in the 1960 Presidential Campaign," a paper delivered at the 1961 Convention of the American Political Science Association. The Democratic National Committee found that out of thirteen suburbs, Kennedy won six and increased the Demo cratic share of the vote in the other seven. See: Research Division, Democratic National Committee, "The 1960 Election Report" (Washington, D.C.: Mimeo, 1962), Appendix A-Y. 22 See, for example. Scott Greer's The Emerging City (Glencoe: Free Press, 1962) and his Governing the Metropolis (New York: John Wiley and Sons, 1962T"; also Charles R. years there has been a considerable amount of work done in the area of "community” power structures, politics, and sociology, largely in response to the limits of the 23 more formal traditional areal political analysis. In contrast are the more narrowly conceived studies of the politics of each of many of the major American cities, for example, those edited by Edward C. Banfield and published by the Joint Center for Urban Studies. In addition, there are the recent papers presented at Adrian’s Governing Urban America (New York: McGraw-Hill, 1961); or the early classic of ¥. Lloyd Varner’s Yankee City series (New Haven: Yale University Press, 1945-1959). Jean Gottman's Megalopolis (New York: Twentieth Century Fund, 1961) stands by itself as a study of the regional metropolitan area from Boston to Charleston, while Lewis Mumford’s The City in History (New York: Harcourt, Brace and Vorld, Inc., 1961 ) is even more expansive in both time and area. For an extensive examination of urban bibliography see: R. T. Daland, "Political Science and the Study of Urbanism." American Political Science Review. LI, 2 (June, 1957), pp. 491-509. 23 See, for example, Floyd Hunter’s Community Power Structures (Chapel Hill: University of North Carolina Press, 1953); the collection of studies edited by Morris Janowitz, Community Political Systems (Glencoe, 111.: Free Press*J 1 96T7~; Robert A. Dahl’s Vho Governs? (New Haven Yale University Press, 1961); Edward C. Banfield’s Political Influence (Glencoe, 111.: Free Press, 1961); N. V. Polsby, Community Power and Political Theory (New Haven and London:Yale University Press, 1963); and M. K. Jennings, Community Influentials: The Elite of Atlanta (New York! Free Press of Glencoe, 1964). On the limits of areal analysis see; Otis D. Duncan, Ray P. Cuzzort, and Beverly Duncan, Statistical Geography. Problems in Analyzing Areal Data (Glencoe, 111. : Free Press, 1961). 10 24 various political science conventions. While concern with suburban political patterns is of fairly recent origin, the suburb itself is not. Lewis Mumford, in his The City in History, points out that the suburb is not a phenomenon of modern technological city life, for, as he says: . . . the fact is that the suburb became visible almost as early as the city itself . . .25 Mumford cites the Greece of Epicurus, early Rome, and the Greater Ur as examples. He goes on, describing other early suburbs: By the time maps and airviews of late medieval cities were made, we find detailed evidence of little huts, cottages, and villas, with ample gardens, springing up outside the cityts walls. By the sixteenth century the land so used served for more than summer residences and recreation. As early as the thirteenth century, indeed, Villani [Giovanni Villani] reported that the land for a circle of three miles around Florence was occupied by rich estates with costly mansions; . . . From the beginning, the privileges and delights of suburbanism were reserved largely for the upper class; . . . a derivative of the relaxed^ playful, goods-consuming aristocratic life Five at the 1962 and four at the 1963 Annual Convention of the American Political Science Association. For example: Robert Warren*s "A Municipal Services Market Model of Metropolitan Organization," or, as an alternative to Martin*s laissez faire model of metropolitan government see Mathew Holden*s "The Governance of the Metropolis as a Problem in Diplomacy." Closer to this dissertation is Thomas R. Dye*s paper: "Leadership and Constituency in Fifteen Suburban Communities." 25 Mumford, on. cit„. p. 483. 26Ibid.. p. 484. 11 Mumford contends that the goal of these early suburbanites was not only to escape the foul air of the city but to create a self-centered unique life— in Mumford*s words: tl • to withdraw like a monk and live like a prince. ,,27 Jean Gottman, in examining the metropolitan sprawl of the northeast United States, points out: From the 1760fs on, such suburban houses, either principal or secondary residences, proliferated around Boston, New York, and Philadelphia and in Rhode Island,28 He cites as concrete examples: Cambridge, near Boston; Greenwich Village, near Manhattan; and Schuylkill Hills, near Philadelphia, It is abundantly clear, then, that the suburb is not a recent manifestation of urbanization but has moved apace with the city, as we know it. What is new is the more recent transition of the United States (and several other "western" nations) from a rural—agricultural economy to an urban—industrial economy featuring a high 29 technology with a bewildering division of labor. 27Ibid. 28 Gottman, op. cit., p, 211. 29 In Western Europe, according to Gottman, cities are growing by leaps and bounds, even in the stagnant economies of Portugal and Spain— and this growth is most vigorous on the periphery. He cites the growth of Lisbon along the estuary of the Tagus River, while another sub urban area stretches along the electric railway to Sintra; see Jean Gottman, Economics. Esthetics, and Ethnics in Modern Urbanization (New York: Twentieth Century Fund, 1962), pp. 11-12. 12 The historian Paul Mantoux has observed that urban ization and the evolution, and increasing division of labor 30 are closely linked. Gottman reminds us that in the 18th and 19th centuries industrial—technological development concentrated large masses of vorkers on production sites, but that in the 20th century industrialization and technol ogy have modified this pattern— with more and more of the labor force turning to "non-productive" functions. He points out that the shift to white collar, non—industrial employment has added much to the variety of the labor 31 force. In fact, the two phenomena, urbanization and technological development, are so closely interrelated that Hope T. Eldridge flatly states that technology is the sine 32 qua non of urbanization. As a result, Luther H. Gulick points out, The urban regions are now the focus of our industrial economy, the nodes of our transportation and circula tion system, and the centers of management and adminis tration as well as of population concentration.33 30 Paul Mantoux, The Industrial Revolution in England in the Eighteenth Century (London; Johnathan Cape, 1928). 31 Gottman, Megalopolis. op. cit.. pp. 52—53. In 1953, according to the Federal Reserve Board, within the non—agricultural economy, non—industrial employment sur passed industrial employment; see "Recovery in the Labor Market," Federal Reserve Bulletin (Washington, D.C.: Federal Reserve Board, 1959), May 1959, pp. 471-476. 32 Hope Tinsdale Eldridge, "The Process of Urbaniza tion," Demographic Analysis, ed. by Joseph J. Spengler and Otis Dudley Duncan (Glencoe: Free Press, 1956), p. 342. 33 Luther H. Gulick, The Metropolitan Problem and American Ideas (New York: Alfred A. Knopf, 1962), p. 6. 13 The metropolitan area is not only the source of production and management, but as Duncan and Reiss remind us, also the 3 4 major consumer vithin the system. So it is that the political scientist must study the city if he is to study the politics of a majority of the population. And the rationale, here, is that the suburb is an integral part of the city— that it is one of the many 35 specialized subareas vithin the functional vhole. Sub—urban and sub-culture Beyond the fact of economic dependence the rela tionship of the suburb to central city is not clear. Just vhat the pattern or flov of population shifts is vithin the metropolitan area has never been satisfactorily determined. The general hypothesis is that population flovs outvard. As Bernard Lazarvitz states, It is assumed that a metropolitan community grovs by extending its central and subcentral business districts into the surrounding residential areas and by building never, more expensive residential areas in the outer or suburban belt. 34 Otis Dudley Duncan and Albert J. Reiss, Jr., Social Characteristics of Urban and Rural Communities (Nev Yorks John Viley and Sons, Inc., 1956), pp. 2-3. 35 For a more extensive discussion of this, see Albert J. Reiss, Jr.fs article, "Research Problems in Metropolitan Redistribution," American Sociological Reviev. XXI, 5 (October, 1956), pp. 571-577. 36 Bernard Lazarvitz, "Metropolitan Community Resi dential Belts," American Sociological Reviev. XXV, 2 (April, 1960), p. ,245. 14 This harks back to the idealized schema posited by E. W; Burgess that the city is essentially a series of concentric 37 areas. Another theory, most familiar to Los Angeles residents, is that population grows out from the center city along transportation routes. What is of concern here is the result of two major population shifts: first, from the farm to the city; and second, from the city to the suburb. Unfortunately, for standardization, cities do not exhibit these phenomena all at the same time— so that the population shifts with which we are concerned have already occurred, are occurring, or will occur in the future depending upon which city one analyzes. For this hypothesis there is some data. Leo F. Schnore, in examining 403 suburbs tyy region, size, and age for their degree of industrialization found differences for 38 each of the three variables. The results indicate that older, larger suburbs (mostly eastern) are industrial while younger, smaller suburbs (primarily southern and western) are residential. There is a s.trong implication in the data that as suburbs grow in size and age they tend to become more industrial. This information lends great weight to Gottman*s statement that: 37 R. E. Park, E. W. Burgess, and R. D. McKenzie, The City (Chicago: University of Chicago Press, 1925). 38 Leo F. Schnore, "The Functions of Metropolitan Suburbs," The American Journal of Sociolocv. LXI. 5 (March. 1956), p. 7551------------ -------------- 15 . . • suburban sprawl was formerly and still is determined largely by the scattering of manufacturing establishments, away from the areas already too crowded by such plants. . . TABLE 1 TYPE OF SUBURB AND REGION Characteristics Per Cent of Suburbs Industrial N Region N. E. 65 173 N. Central 44 119 South 31 32 Vest 29 79 Size (1950) 50,000+ 77 57 25-50,000 65 86 10-25,000 38 260 Age (years) -30 20 40 30-40 20 35 40-50 35 65 50+ 61 263 Queen and Carpenter put it that most of the communities 40 within our metropolitan areas are, or have been, suburbs. Or, more simply, today's slums were probably yesterday's suburbs• 39 Gottman, Megalopolis. op. cit.. p. 210. 40 Stuart Queen and David B. Carpenter, The American City (New York: McGraw-Hill Book Company, Inc., 1953), p. 119. 16 This suggests that much of the discussion about suburban political behavior fails to distinguish between suburban types. Broadly speaking, it is generally assumed that suburbs, contra central cities, are higher in socio economic status. Recent research by Leo Schnore and 41 David Varley substantiate this. However, Lazerwitz found that in the southern area of the nation the pattern was reversed, with the central city having a higher socio- 42 economic status than the suburb. It seems clear, at least, that the metropolitan areas are divided into mosaics of subareas or, perhaps, subcommunities. Urban ecologists have often described these sub- 43 areas as being significant in urban life. Eshref Shevky and Wendell Bell have found that urban residents with similar status attributes and life style tend to cluster 44 together in specific residential areas. Scott Greer, 41 Leo P. Schnore and David V. Varley, "Some Con comitants of Metropolitan Size," American Sociological Review. XX, 4 (August, 1955), pp. 410-14. 42 Bernard Lazerwitz, "Some Characteristics of Residential Belts in the Metropolitan Community" (unpub lished Ph.D. dissertation, University of Michigan, 1958), pp. 53-66 (microfilm). 43 Amos A. Hawley, Human Ecology; A Theory of Com munity Structure (New York! Roland Press, 1950); Leslie Kish, "Differentiation in Metropolitan Areas," American Sociological Review. XIX, 4 (August, 1954); and Otis and Beverly Duncan, "Residential Distribution and Occupational Stratification," The American Journal of Sociology. LX, 5 (March, 1955). 44 Eshref Shevky and Vendell Bell, Social Area Analvsis (Stanford: Stanford University Press, 1955)• 17 using the Shevky-Bell Index, found that neighborhood patterns within a metropolitan area still survive, at least 45 to a limited degree. Milton M. Gordon, in analyzing the concept of sub-culture, suggests that it is a subdivision of the national culture composed of various social situa tions, e.g., class, status, ethnic background, religion, 46 etc. Sub-cultures probably have an areal or ecological dimension too. Valter C. Kaufman and Scott Greer, examining political behavior for population typologies based on the Shevky-Bell Index, found significant 47 differences in voting behavior. This suggests that suburbs may well exhibit political behavior patterns related to their sub-urban sub-cultural nature. Suburban growth All of this acquires political significance when the suburban population growth rates and suburban propor tion of total population is taken into account. An examination of national population trends 1900-1950 shows that the suburban areas have not only been the fastest 45 Scott Greer, "Urbanism Reconsidered: A Compara tive Study of Local Areas in a Metropolis," American Sociological Review, XXI, 1 (February, 1956), p. 19. 46 Milton M. Gordon, "The Concept of the Sub-Culture and Its Application," Social Forces. XXVI, 1 (October, 1947), p. 40. 47 Valter C. Kaufman and Scott Greer, "Voting in a Metropolitan Community: An Application of Social Area Analysis," Social Forces. XXXVIII, 3 (March, 1960), p. 203. 18 growing areas but that they will soon be, if they have not 48 already become, the major population areas, TABLE 2 GROWTH DATA FOR PRINCIPAL STANDARD METROPOLITAN AREAS Per Cent of U.S, Population Rate of Growth Previous Decade Per Cent of Total U.S0 Growth Previous Decade xear ---------- Central Cities Rings Central Cities Rings Central Cities Rings 1950 32.3 23.8 13.7 34.8 30.7 48.6 1940 31.6 19.5 5.1 13.8 22.8 34.9 1930 31.8 18.0 23.3 34.2 43.3 32.9 1920 28.9 14.8 26.7 22.4 46.8 20.8 1910 25.0 12.7 35.3 27.6 37.4 15.7 1900 21 .0 10.7 — ------- — — ---- —— — Census data for 1960 shows a central city population of 58 million and a ring population of 55 million, with growth rates for the decade of 11 per cent and 49 per' cent 49 repectively. The Department of Commerce states in a 48 Donald J, Bogue, "Urbanism and Metropolitanism," The Suburban Community. ed. by William M, Dobriner (New York: G, P. Putnam^ Sons, 1958), p. 24. 49 U.S. Bureau of the Census, U.S. Census of Popula tion: 1960, Number of Inhabitants. United States Summary. Final Report PC (1)-1A. U.S. Government Printing Office, Washington, D.C., 1961, p. 1—106, Not every suburb grew. In thirty-three SMA*s there were population declines in the suburbs. This emphasizes 19 1961 publication: For the first time, the population in the areas outside central cities (within the SMA's) almost equaled that inside central cities. It is a reasonable assumption that the 1960 census is the last in which central cities will be more populous than the suburbs . *0 Various problems exist in accurately comparing growth rates of suburbs, central cities, and rural areas due to: (1) changing definitions by the Bureau of the Census, and (2) the fact that functional suburbs can and do exist within city limits of the central city. Nevertheless, the suburbs are, or soon will be, larger than the central cities in terms of population and their population growth rates show no signs of declining, at the point that it is well to ask which suburb one is examining. For,a more extensive discussion of differing suburbs, see: Business and Defense Services Administration, Changing Metropolitan Markets 1950—1960, U,S. Government Printing Office, Washington, D.C., 1961, ^Business and Defense Services Administration, op. cit.. p. 20. Figures for "ring" and "central city" growth are complicated by central city annexations which took place in the 1950Ts, One hundred and sixty of the 212 SMA's had boundary changes in the decade0 Unadjusted data shows that the rings (or suburbs— the Bureau of the Census interchanges them somewhat casually) grew about four and one—half times as fast as the central cities. Adjusted data shows the rate to have been closer to forty times as fast, accounting for almost all (about 97 per cent) of the population growth in the SMAts. Leo F. Schnore, "Municipal Annexations and the Growth of Metropolitan Suburbs, 1950-1960," The American Journal of Sociology. LXVII, 4 (January, 1962), pp. 407-09. 20 least in the south and west.^ Philip M. Hauser predicts that over 80 per cent of the expected population growth of 64 million between 1960 and 1980 will be in the 52 suburbs. Vithin this national pattern of suburbanization (replacing urbanization which is largely accomplished) the Pacific southwest region and particularly Los Angeles stand out as the prototypes. Raymond P. Cuzzort found that the most suburbanized economic region of the United States was the Pacific southwest, with 52 per cent of its population 53 located in the suburbs. While Los Angeles is such an extreme case of decentralized population that Edwin A. Cottrell and 51 See^Otis D. Duncan and Albert J. Reiss, Jr., o^*_cit., p. 6. Otis Duncan casts doubt on the meaning of data comparing central cities and rings (or suburbs) due to differences in population density, city boundaries, and incorporated vs. unincorporated suburbs. In short, he maintains, the aggregates subsumed under the two rubricks are too disparate within the two to make comparisons between the two meaningful. Otis D. Duncan, "Research on Metropolitan Populations: Evaluation of Data," Journal of the American Statistical Association, LI, 276 (December, 1956), p. 592. 52 Philip M. Hauser, Population Perspectives (New Brunswick, N.J.: Rutgers University Press, 1960), pp. 101-06. 53 Raymond P. Cuzzort, "Suburbanization of Service Industries within Standard Metropolitan Areas," Studies in Population No. 10 (Oxford, Ohio: Scripps Foundation and Miami University, 1955), p. 50. 21 Helen L, Jones have stated that in one sense the Los Angeles Metropolitan area is not yet a community at all, but rather a collection of cities, towns, districts, neighborhoods, and semirural areas, in the process of becoming a community . '. . , Richard V. Redick found that Los Angeles has the highest rate of decentralization'of any of the metropolitan areas he studied. However, all the other areas, excepting Boston, exhibited some decentralization too.^ Metropolitan Government Urbanism Urbanism has resulted in much more than a redistri bution of people and a change in economic organization. The result has been the proliferation of problems, largely unsolved, which give a completely new dimension to govern ment and politics. Luther Gulick, writing in 1962, takes the position that 54 Edwin A. Cottrell and Helen L. Jones, "I. Charac teristics of the Metropolis," Metropolitan Los Angeles (Los Angeles: The Haynes Foundation, 1962), p . 50. 55 Richard ¥. Redick, "Population Growth and Distri bution in Central Cities, 1940-1950," American Sociological Review. XXI, 1 (February, 1956), p. 43. One interesting side effect of this pattern of sub urbanization is that while large metropolitan newspapers have been declining in numbers over the past few years, suburban neighborhood newspapers have been flourishing. The suburban newspaper market has reached the point that there is now a colorgravure Sunday supplement especially published for the suburban areas. "Suburbia Today" ("The Magazine of Pleasant Places") claims distribution in 1,001 "selected suburbs" through 229 newspapers. In the Los Angeles area these include twelve community newspapers from the Burbank Review to the San Marino Tribune. 22 . . . the major unsolved problems of our society and government arise primarily from the nev rapid urbani zation of the c o u n t r y . 56 And, since it is in the urban areas that most Americans now live, it seems safe to predict that urban areas will continue to be the major source of governmental-political problems. Edward C. Banfield and James Q. Wilson describe the major dimensions of these problems as cleavages within the urban areas: Vithin the cities and metropolitan areas the most important cleavages are those between (1) haves and have-nots, (2) suburbanites and the central city, (3) ethnic and racial groups, and (4) political p a r t i e s . 57 The suburban schism The dimension to be examined— suburbanites and the central city— has engendered some very sharp conflicts not to be lightly dismissed. For example, William Zeckendorf, writing on the conflict in The Atlantic Monthly oVer four teen years ago, said: Satellite towns (suburbs), which are the product of decentralization are parasites. The high cost of main tenance of the central core that supports the whole metropolitan area is borne by the city, but the revenues and benefits go to the towns at the peri phery. . . .58 ■^Gulick, on. cit.. p. 9. 57 Edward C. Banfield and James Q. Wilson, City Politics (Cambridge: Harvard University Press and Massachu setts Institute of Technology Press, 1963), p. 35. William Zeckendorf, "Cities Versus Suburbs," The Atlantic Monthly. (July 1952), p. 24. At that time Zecken dorf was President of Webb and Knapp, real estate devel opers; he advocated unilateral annexation. 23 Levis Mumford is of the opinion that spreading suburban areas undermine and erode old city centers without forming coherent systems able to assume the burdens of the center, and that ". . • left to themselves, as Los Angeles already demonstrates, these forces (of decentralization) 59 will automatically destroy the city." Eventually, of course, the broad cleavages are expressed in concrete substantive or procedural issues. It is these issues which we find at the "core" of metro politan government movements. Each of these issues is part and parcel of the suburbanization of urbanization. A recent publication of the American University lists thirteen of these issues: (1) transportation, (2) traffic controls, (3) highways, (4) water supply, (5) sewage disposal, (6) sanitation, (7) housing, (8) zoning and land use, (9) taxation, (10) local financing, (11) assessments, (12) schools and public education, and (13)' hospitals and public schools.^ Not surprisingly, Gulick adds to the standard list of service failures in metropolitan areas two other "failures" which underlie the service failures: the lack of planning and lack of region-wide democratic 59 Mumford, op. cit., p. 503, italics added. ^"The Challenge of Metropolitan Government," (Washington, D.C.: The American University, 1958), p. 16 (mimeo). Also, see Dan R. GrantTs "Urbanism and Suburban Nashville: A Case Study in Metropolitanism," Journal of Politics, XVII, 1 (February, 1955), pp. 82-99, for an examination of the issues in a concrete case. 24 methods for reaching consensus (the latter, of course, another way of stating the existence of systematic functional and areal cleavages within the region) Robert Wood examines the cost features of this cleavage and lists three major negative aspects of suburban growth in terms of increased costs: (1) costs to central city and suburbs due to separation of place of work and residence, (2) costs to the central city due to serving a daytime population between 30 and 50 per cent in excess of its resident (taxpaying) population, and (3) costs to the suburbs due to service duplications and loss 6 2 of the economics of scale. Wood also suggests another major schism: As the suburban exodus goes on, we are told that the fringe and the central city are drifting farther and farther apart, the former increasingly Republican, the latter Democratic, so that the metropolitan areas are likely to be split into two warring camps.63 Banfield sees these two warring camps as a major impediment 64 to metropolitan government. However, there is good reason to believe that a strict assignment of Republicans 61 Gulick, op. cit.. pp. 120-24. ^ 2 Robert C. Wood, "Metropolitan Government, 1957: An Extrapolation of Trends," American Political Science Review. LII, 1 (March, 1958), pp. 109-110. ^ Ibid. . p. 110. 64 Edward C. Banfield, "The Politics of Metropolitan Organization," op. cit.. pp. 86. 25 to suburbia and Democrats to central city considerably overstates the case. Frederick Wirt found that one-third of the suburban areas which he examined were heavily Democratic on the Ranney-Kendall typology.^ Banfield and Wilson suggest (aside from the partisan split between suburb and central city) that center city government now finds itself facing an alliance in many state legislatures 66 between suburbs and rural hinterland. This is not too surprising, since the suburb quite often impinges directly on rural areas. Indeed, the suburban resident soon finds early chickens and late smudge pots more than he bargained for, while the farmer finds run-off patterns altered, tax f \ 7 assessments increased, and sewage disposal complicated. Frederick M. Wirt, "Suburban Patterns in American Politics," a paper delivered at the 1960 American Political Science Association Convention (mimeo), pp. 10-11. See also, pp. 5 and 9 supra. ^Banfield and Wilson, op. cit.. p. 37. An excellent example of this alliance can be found in an article by David D. Gladfelter, "Water for Wauwatosa," Cases in State and'Local Government, ed. by Richard T. Frost (Englewood Cliffs: Prentice-Hal1, Inc., 1962), pp. 280-291. In the case cited, Wauwatosa, through Wiscon sin^ Public Service Commission, was able to force Milwaukee to supply it with water. Wauwatosa without water would have had to become annexed to Milwaukee (or so Milwaukee hoped); with water the suburb could continue to exist as a separate entity. f\T For an extensive discussion of the rural-suburban problem and suggested solutions, see: G. J. Hein, "The Stake of Rural People in Metropolitan Government," Farm Economics Division, United States Department of Agriculture, Miscellaneous Publication No. 869 (Washington, D.C.: United States Government Printing Office, 1961), p. 21. TABLE 3 SUBURBAN ATTITUDES TOWARD THE CENTER CITY Advantages and Disadvantages Former Residents of Central City Other Residents Total Advantages Good public transportation 10.0% 20.3% 18.0% Good public utilities 18.6 20.7 20.3 Accessible and convenient 18,6 13.7 14.8 No advantages 25.7 28.7 28.0 Other and no answer 27.1 16.6 18.9 Total 100.0 100.0 100.0 Disadvantages Noisy, dirty, unsafe 18.6 21 .9 21.2 Lack of space and privacy 41 .4 46.5 45.3 Higher cost of living 8.6 9.9 9.6 No disadvantages 4.3 10.4 9.0 Other and no answer 27.1 11.2 14.8 *Source: Hawley and Zimmer, op. cit., p. 164. ro O' 27 Both the suburbanite and the rural resident are united in opposition to the central city and resort to the legislature to protect their interests, particularly in resisting * political expansion of central city. Reason for this resistance is not far to find. The implications of political boundaries provide one index. As Gulick points out: Every established local governmental boundary, . . . soon acquires a fiscal significance. . . . Once bonds are issued— for twenty or forty years— it becomes extremely difficult to abolish the district or join it to another district. There is nothing so hard to shift geographically as debts. . . .68 The existence of fixed and ' ’immortal" boundary lines soon creates and sustains a governmental institution, a political system, a political organization, and a structure of political power. These are not easily altered nor do 69 those individuals involved take kindly to change. There is no doubt that resistance to metropolitan government has been both combative and highly successful. (While over stating the case, perhaps, the Jules Feiffer cartoon TO illustrates the schism.) 68 Gulick, op. cit.. p. 50. 69 For an extensive discussion of the many obstacles to the solving of metropolitan problems, see Stanley Scott (ed.), Metropolitan Area Problems (Berkeley: Bureau of Public Administration and University Extension, 1960). A good example of the resistance of "institutions" to city merger can be found in“Philip Moneypenny and Gilbert Y. Steiner's "Merger? The Illinois Consolidation Case," Cases in State and Local Government, op. cit., pp. 267-79. 70 Jules Feiffer. New Republic. February 12, 1962, p. 25; reproduced with the permission of the publishers. f o r n e w t h o s e o f u s who me touv im rue vimharos o f s u b u r b a n c m D e fe n se me B een c o n c e r n ? w ith ru e P to tie M o f how to m a iu ta iw l a w am p o r OCR fOLLOWiU6 A nOCLCNH ASSAULT. 28 THE 816 a m WOULD, OF COURSE. BE AWMIHILAT6P r m & W O IM P L lF n m T H & R C M C D C F e n s e p r o b l e m s . i H H e t e u R t e m . h o w e v e r , FOR THOM OF US IW 6 U6 U R W A t h e r e A * e e o w w o t o b e COMPUCATIOW9. UK WOOlP 0£ SUBJECT TO MASS OWSUAUS0TS OF R E fo e e e s f r o m th e c m . WHILE COR. H E A R T S ,A S Au MARK must, oo our TO THESE VICTIMS TH W 0 0 p o s e a t h r e a t to OUR C AREFU U n PLAUWFP PEOSRAM . how c a w ow e re c i. a r a d i o - a c t i v e M o e t h a t Them w o w .o m t e e h a w IWOUR T O W )? WO, w e CAW OWW PRESERVE OUR I tiA U O f L ife m 0ARRICADIW6 o w e STREETS AWP R£-i?lieeCT||)6 ALL MlORAWr t r a f f i c r o Tfie p y g ti c w o n - 0JAUSI A \ o m THEM PERHAPS, WITH (MPROICP DIRECTIONAL 6i6W s a m p f r e e ro ad m a p s . a 0l)T WHEl) MAW'S SURVIVAL IS AT st a k e ire m a h w e u s o p e e w p e R TO T he B A S E R M STltiCTS. OUR 3A R R lC A 0eS M1 6 HT HAVE TO 66 O B F e w e o m f o r c e o f a r m s . 6UT J l^ T AS w e ARE W IU W 6 To 60 TO WAR - TO 0EFEUD OUR fr e e d o m s o w e s h a u , ee W ILW J6 t o g e fe w p w h a t's L 6FT O f IT 0H MAUUIW6 TUP S U 6 D R 0 A M B A R R IC A D E S f im su b u r b a io c iv ic c e f e w s e o n e m o t t o i s : IF HOP CAM'T 6 6 T H O U RSevf A RUSSIAN, s e m e so p - a w AM6RICAW- l t f - 1 29 Metropolitan governments In the President’s Report on Recent Social Trends (1934), R. D. McKenzie describes the emergence of the metropolitan region and "recent" efforts to create metro politan governments. In the years since, dozens of efforts 71 have been made to create metropolitan governments. According to Scott Greer, not until 1957 was there a single 72 case of success. Gulick’s report is of the same tenor: Except for the three referendums in Miami and Dade County, Florida, in 1956, 1958 [sic] and 1961, and somewhat less sweeping decisions made in Baton Rouge in 1947 and in Atlanta in 1950, no metropolitan area in the United States has by the vote of its people indicated a desire to extend . . . local government to cover its "metropolitan r e g i o n . "73 Conversely, time and time again, various metro plans have been rejected by the voters. When metro does succeed, it may be in reality only a glorified special district as in the case of King County’s (Washington) metropolitan government, which after a firs; failure was 71 These efforts have not been casual or offhanded by any means. According to one source, nearly one hundred metropolitan studies and surveys have been made in the last few years. Government Affairs Foundation, Metropolitan Surveys: A Digest (Chicago: Public Administration Service, 1958). 72 Scott Greer, The Emerging City, op. cit., p. 181. Greer cites R. D. McKenzie's "The Rise of Metropolitan Communities," Recent Social Trends in the United States. Report of the President’s Research Committee on Social Trends (New York: McGraw-Hill Book Company, Inc., 1933). 73 Gulick, op. cit., p. 98 30 reduced to a sever and storm drainage district and then accepted (in September 1958) by both center city and 74 suburbs. Vhile there is quite obvious resistance to unification,' annexation continues apace. In fact, accord ing to Gulick, annexation has been "unusually active" with at least 2,400 square miles containing over two million persons being annexed to over 1,000 cities between 1950 75 and 1957. The average, typical annexation involved 2.4 square miles containing 1,666 people— hardly a grandiose process and certainly not an answer to metropolitan area problems. ¥hy resistanc-e? Several theories have been offered to explain suburban reluctance to merger, annexation, or to metro governments. These include: (1) greater political efficacy, (2) escape from heavy, center city taxes, (3) ignorance regarding what metro government can do for suburban areas, (4) the differing socioeconomic levels of various suburbs and the central city, (5) governmental "game preserves," (6) industrial tax dodges, (7) zoning variance, and (8) sin enclaves. Examples of points six, seven, and eight can be found in Los Angeles County in ^Don Becker, "The Fight for Seattle Metro," Cases in State and Local Government, op. cit., pp. 292-301. 75 Gulick. op. cit.. pp. 51-2. and see fn. 50. p. 19 supra. ----- Vernon, Dairy Valley, and Gardena respectively. Whether or not suburban residents have a better chance to be politically effective is doubtful. Amos H. 2 Hawley and Basil G. Zimmer, examining the Flint, Michigan, area, found that while suburban residents felt they were closer to their government, they were actually less knowledgeable about their local government than were central city residents. Examining the second theory of suburban resistance to merger— taxation— Hawley and Zimmer found that suburban residents favored tax increases to provide more public 77 services. By holding age, occupation, education, take- home pay, and family composition constant, the apparent resistance to higher taxes for better public services disappeared. Resistance to paying taxes cannot be con sidered, however, apart from consideration of the expenditure of tax money. It is significant that suburbs, in contrast to center cities, require greater capital expenditures and less welfare expenditures. There is a difference in the needs of the suburbanite and the central T A Amos H. Hawley and Basil G. Zimmer, "Resistance to Unification in a Metropolitan Area," Community Political Systems. ed. Morris Janowitz (Glencoe: Free Press, 1961), pp. 176-78. 77 Ibid.. pp. 162-69 and 173-74. Their respondents were aware of the lower level of public services available in suburbia. 32 78 city residents Quite often one goal of suburban residents is to keep out, through zoning and other restrictive practices, those segments of the population requiring greater welfare expenditure, thereby shifting 79 the load to central city. According to Harvey Brazer, there is a direct relation between the proportion of a metropolitan area’s population living outside the central 80 city and the per capita expenditures of the central city. Granting the great difference in fiscal requirements and desires, an.d the equally wide variance in tax base, the point must be made that proposed solutions to the tax— service problem of local governments may be rejected because local residents place a higher value on their subjective political efficacy than on their objective 81 fiscal efficiency. Hawley and Zimmer, after examining suburban residents* political knowledge, desire for more public service, and choice of alternative solutions, came to the tentative conclusion that resistance to merger is 78 For a discussion of the different fiscal require ments and tax resources of suburban and central city areas, see: Robert C. ¥ood*s 1400 Governments (Cambridge: Harvard University Press, 196171 Wood’s Metropolis Against Itself (New York: Committee for Economic Development, 1959); and Harvey E. Brazer*s Some Fiscal Implications of Metropoli tan! sm (Washington, D.C.: The Brookings Institution, 1962) Reprint Series No. 61. 79 Brazer, op. cit.. p. 77. 80Ibid.. p. 76. 81 Hawley and Zimmer, op. cit.. passim. 33 rooted largely in ignorance of how much metro government O p could do for them and how little local government can do. 83 Vood attributes resistance to metro government to apathy. There are, moreover, local forces willing and able to vigor ously protect their own governmental "game preserves" by playing on these fears, lack of knowledge, and apathy. Ironically, these "forces" include those very political leaders over whom the suburbanite assumes a sense of direct control! Suburbs and local government Finally, three recent studies of suburban govern mental organization and its relation to social and economic differences found a relation which suggests: (1) that local governmental boundaries often coincide with area speciali zation, (2) that different governmental forms appeal to different demographic groupings, and (3) that cooperation between local governments is related to similarity of the 84 populations involved. 82Ibid., p. 182. 83 Vood, 1400 Grovernments. op. cit.. Chapter 5. 84 Oliver P. Villiams, Thomas R. Dye, Harold Herman, and Charles S. Liebman, "Urban Differentiation and Political Integration," a paper delivered at the 1962 American Poli tical Science Association Convention (mimeo). Leo F. Schnore and Robert R. Alford, "Forms of Government and Socioeconomic Characteristics of Suburbs," Administrative Science Quarterly. VIII, 1 (June, 1963), pp. 1-17. Charles S. Liebman, "Functional Differentiation and Political Characteristics of Suburbs," American Journal of Sociology. LXVI, 5 (March, 1961). 34 Oliver Williams found that not only are dormitory suburbs often incorporated, but that there are, in addition, . . . industrial enclaves, recreation resorts, commer cial centers, intellectual retreats, racial and ethnic ghettoes, company towns, and religious colonies which correspond roughly with local political units. He found that adjacent communities with similar socio economic status tended to enter into cooperative govern mental activities more often than did adjacent communities with divergent socioeconomic status. Schnore and Alford found that manufacturing suburbs favored either the commission or mayor-council forms of local government (71 and 62 per cent, respectively), while dormitory suburbs favored the council-manager form (67 per 86 cent). They also found differences in the distribution of partisan and non-partisan governments. (See Table 3.) Some of the differences found by Schnore and Alford may be due to the age of the suburb; older suburbs ibay tend to the commission form since it was popular when they were founded. (See Table 4.) There also seemed to be a relationship between 85 Williams et al.. op. cit.. p. 1. 86 Schnore and Alford, op. cit., p. 10. The authors examined 300 suburbs in the United States, including twenty- four in the Los Angeles area. An earlier study by Charles S. Liebman (one of the authors of the article by Williams et al.) found that residential suburbs favored the city manager form while manufacturing suburbs favored other forms. 87 Schnore and Alford, op. cit.. p. 11. 35 TABLE 4 FORMS OF SUBURBAN GOVERNMENT BT FORMS OF ELECTION Form Percentage of Partisan N Commission 29 41 Mayor-Council 57 146 Council-Manager 16 113 Total 38 300 TABLE 5 SOCIAL AND ECONOMIC CHARACTERISTICS OF SUBURBS BT FORM OF GOVERNMENT Fo rm Per Cent Per Cent Finished Vhite Collar High School Commission Mayor-Council Council Manager 47 48 56 42 46 56 36 88 social characteristics and form of government. In terms of metropolitan government these differ ences offer real obstacles. Suburban Political Behavior Complex phenomena Oscar Handlin, in his introduction to Albert Gordon1s Jews in Suburbia, asserts that The move to the suburbs has been one of the most striking phenomena of recent American life. Although there are precedents for the spread of urban population away from city centers, even in the nineteenth century [see Mumford, supral. the movement of the past twenty years has been of such a magnitude as to demand serious examination. Surely this trend reflects important changes in the character of society and in the system of values in the United S t a t e s . ^9 An examination of this population movement implies three basic effects, according to the authors of The American 90 Voter. First, shifts in political-population aggregates without a change in political affiliations; second, 89 Albert I. Gordon, Jews in Suburbia (Boston: Beacon Press, 1959), p. ix. Gordon looks at eighty-nine suburbs and while concerned with the impact of suburbia on Jews his work has broader impact. 90 Angus Campbell, Philip E. Converse, Varren E. Miller, and Donald E. Stokes, The American Voter (New York: John Viley and Sons, Inc., 1960), pp. 442-43. This book is largely a study of voting in presidential elections and, as the authors state, they do not ". . . imply that voting behavior in presidential elections and other national elections (congressional) is essentially the same." Ibid.. p. 7. The work is based on data collected by the Survey Research Center for the elections of 1948, 1952, and 1956. 37 population residence changes combined with political affiliation changes (social mobility inducing both changes), and third, residential change injecting individuals into a conflicting community political environment. However, suburbanization, unlike religion or race, affects the individuals political behavior for only a 91 limited time in his political life cycle. This is the period of time that Seymour Lipset and his colleagues call 92 the political generation. Following: (1) socialization and adolescence, and (2) the first vote, they suggest that there is a third period in the individual's political life cycle which they call (3) the political generation which is, in turn, followed by (4) maturity, and (5) old age. Election research indicates that an individual's greatest political activity and involvement usually occurs in this 91 This assumes that the individual moves to suburbia as ap adult rather than being raised there from birth. The latter case does of.course exist; but as the children of suburbia mature and their parents age, perhaps moving back to center city after the children have left home (for center city too?) the character of suburbia will certainly change. Indeed, if suburbia is an ecological expression of a parti cular point on the life cycle, then it may no longer exist when the children have grown and when the first generation adults have aged. 92 Seymour M. Lipset, Paul F. Lazarsfeld, Allen H. Barton, and Juan Linz, "The Psychology of Voting: An Analysis of Political Behavior," Handbook of Social Psychology, ed. by Gardner Lindzey, Vol. II (Cambridge: Addison-Wesley Publishing Company, Inc., 1954), pp. 1114-1150. 38 93 "political generation" during his thirties and forties. Significantly, it is at the beginning, or just prior to this "political generation," that the move to suburbia is most often made; a move which appears to be a significant aspect of upward social mobility, status inconsistency, change in life style and consumption patterns, and the assumption of new social roles. This conjunction may well have considerable significance for an individualTs political "set" and subsequent behavior. j The major popular and professional concern has been whether or not suburbanism has— in some modern political aspect of the American Horatio Alger social mystique— 94 produced a greater Republican vote. It is generally held that the traditional big city, Democratic vote is no longer as dominant because of the growing suburban Republican 95 vote. According to Angus Campbell and Homer C. Cooper, ". . . the metropolitan suburbs are the only areas in which 93 See, for example: Bernard R. Berelson, Paul F. Lazarsfeld, and William N. McPhee*s Voting (Chicago: University of Chicago Press, 1954), or Campbell et al., The American Voter, op. cit. 94 As we shall see, "produced," "greater," "Republican," and "vote" are almost useless terms unless rigorously defined (which they seldom are). 95 Philip M. Hauser suggests that the urban environ ment itself tends to break down traditional political affiliation just as it tends to break down all traditions. Philip M. Hauser, "Some Political Influences of Urbaniza tion," Reader in Sociology, eds. Paul K. Hatt and Albert J. Reiss, Jr., (Glencoe: Free Press, 1951), p. 466. 39 96 Strong Republicans outnumber Strong Democrats." While, as we saw earlier, Greenstein and Wolfinger went one step farther and maintained that there was an identifiable suburban factor which, after controlling for the effects 97 of education, income, and occupation, did not disappear. While Greenstein and Wolfinger found a suburban Republican identification. Thomas R. Lye rejected registration as a meaningful indication of suburban political behavior because: The propensity of suburban residents to register Repub lican regardless of their voting habits made meaningless any classification based upon registration a l o n e . 98 Conversely, Lazerwitz found that in the period 1950-1956 suburbs were increasing their Democratic registration and congressional vote while continuing to cast a "Republican" 99 vote for president. Whether or not suburban residents are Republicans appears to depend upon which election criterion is used— congressional or presidential— and whether or not one is speaking of identification or registration. In addition, too, the regional location of the suburb may determine the voting behavior. 96 Angus Campbell and Homer C. Cooper, Group Differ ences in Attitudes and Votes (Ann Arbor: University of Michigan Press, 1956), pT 46. For the actual figures see Appendix A-VI. 97 Greenstein and Wolfinger, op. cit.. p. 477. 98 Dye, op. cit., p. 58. 99 Lazerwitz, dissertation, op. cit.. p. 81. 40 Frederick M. Wirt suggests there may be a difference in the political behavior of suburbanites depending upon the differing economic function of the suburb. Using Victor Jones1 economic classifications he examined suburbs for four factors: (1) manufacturing level, (2) rental costs, (3) housing growth, and (4) employment-residence ratio. Wirt found a correlation between the Republican vote and each factor except housing growth.A year later, Charles Liebman published his analysis of Republican senatorial voting in tyenty-one suburbs as a function of economic 102 organization. Classifying the suburbs as: (1) wholly residential, (2) mostly residential with some manufacturing, (3) mostly manufacturing with some residential, and (4) wholly manufacturing, he found that there appeared to be a greater Republican vote in the residential suburbs than in the manufacturing suburbs. But, in examining suburban party identification, Campbell and Cooper found that Democratic identifiers still outnumber Republican identifiers.^^ (See Table 5.) 100... . • , . c Wirt, op. cit.. p. 15. ^ ^ Ibid., pp. 15-16. See Victor Jones* "Economic Classification of Cities and Metropolitan Areas," Municipal Yearbook. 1954 (Chicago: International City Managers Asso- ciation, 1955), pp. 62-70. 1 02 Liebman, op. cit.. p. 488. 103 While the differences were striking, they were not statistically significant, due largely to the small N. 104 Campbell and Cooper, op. cit.. p. 46. See Appendix A TABLE 6 SUBURBAN PARTY IDENTIFICATION Characteristic Per Cent Per Cent Totals Strong Democratic identifiers 1 5 Weak Democratic identifiers 32 Independent Democratic identifiers 9 56 Strong Republican identifiers 20 Weak Republican identifiers 9 Independent Republican identifiers 9 38 Independents, no identification, etc. 6 Eliminating the Independent Democrats and Independent Republicans gives the Democrats little better than a three- to-two split in suburbia. Note that while Dye found that suburbanites register Republican but do not necessarily vote Republican (supra), Campbell and Cooper found that they identify as Democrats but vote Republican and did so even in the 1954 Congressional elections. A further exami nation of the data indicates that this Republican edge in vote was not due to any appreciable crossing of party lines by Democrats but, rather, that a larger proportion of Democrats did not vote.^^ 1Q5Ibid.. p. 47. 42 TABLE 7 PROBABLE VOTE, CONGRESS 1954106 Vote Suburban Respondents I I Per Cent N Democratic 41 33 Republican 59 47 Alfred de Grazia's comparison of the 1948 and 1952 107 presidential vote elaborates on this point. De Grazia’s data shows that while Stevenson's share of suburban elec toral support (outside the south) was equal to or even slightly better than Truman's, Eisenhower picked up a large proportion of the electors who did not vote at all 108 in 1948. Thus, while the Truman-Stevenson suburban vote was fairly consistent, the Dewey (1948)-Eisenhower vote increased markedly. 106 Ibid., p. 25. Dye's data based on one metropoli tan area only while Campbell and Cooper used Survey Research Center national data. The figures are extracted from Table 111-6, Campbell and Cooper, op. cit. 107 ! Alfred de Grazia, The Western Public (Stanford: Stanford University Press, 1954), pi 165. See Appendix A-VII and VIII for non-voter data for metro centers and suburbs and change 1948-1952. 108 The complexities of the "suburban vote" are well illustrated by the somewhat different patterns in the south. In the first place, southern voter participation is consid erably lower in presidential elections than in the rest of 43 Chameleon or segregation? Totton J. Anderson, writing in Western Politics, raises the central question regarding the complexities of suburban political behavior when he suggests that . . . the hypothesis that Democrats moving from metro politan areas join the resident Republican majority in suburbia . . , (is) . . . well worth investigation.1^9 Do Democrats moving to "Republican" suburbia change political complexion to match that of their neighbors— earlier referred to as the "chameleon effect"? Or, is suburbia merely an ecological screen sifting out Republi cans, or potential Republicans, from the center city aggregates?^® An examination of gross election statistics by Edward C. Banfield suggests that there is a chameleon effect, since the net Republican suburban vote has not been the nation (and much lower in congressional elections). Thus, comparisons of Democratic and Republican vote usually account for less than half of the electorate. Nevertheless, there is a slightly higher turn-out in southern suburbs than in southern central cities for both congressional and presi dential elections. In 1952 southern suburbs voted for Stevenson three to two, while the central cities split evenly. But in 1956 the Republicans succeeded in taking the central cities two to one and the suburbs three to two. Data from Lazerwitz, dissertation, op. cit.. pp. 68-71. See Appendix A, IX for data. 109 Totton J. Anderson, "California: Enigma of National Politics," Western Politics, ed. Frank H. Jonas (Salt Lake City: University of Utah Press, 1961), p. 77. 110 Apparently, the screening effect or selective migration thesis was first proposed by Harlan Douglas in his The Suburban Trend (New York: The Century Company, 1925), p. 34. 44 diluted by the population movement of Democrats from central 111 city. As indicated above, Greenstein and Wolfinger find evidence t*f the chameleon effect in their study. Helen Hill; Miller, writing in The Atlantic Monthly* , recently observed that The suburbs are the growing areas. The central cores of the big cities, where the Democratic machines of the last generation flourished, are now being depopulated. As the incomes of their former inhabitants rise, these people move to the suburbs and lose their sense of Democratic togetherness.112 A journalist in the New York Times comments that the majority of new arrivals to suburbia vote Republican in 113 about the same ratio as the earlier settlers. Robert C. Wood's recent study of the New York metropolitan area casts some doubt on these professional and popular opinions, however. Wood attributes the Republican gains of 1948-1952 in suburbia to the effects of a general national trend, pointing out that For 114 metropolitan areas showing increasing Repub lican strength in 1952 over 1948, the gain in eight central cities was greater than in their s u b u r b s .114 111 Banfield, "The Politics of Metropolitan Area Organization," op. cit.. pp. 84-5. Banfield makes the assumption that the aggregate number of Democrats who have moved from central city to suburbia is greater than the Democratic vote in suburbia. ^^Helen Hill Miller, "The City Vote and the Rural Monopoly," The Atlantic Monthly. October 1962, p. 65. 113 "Both Parties Court the Suburbs in Bid for New Resident's Votes," New York Times. May 31, 1956, pp. 1; 15. 114 Vood, 1400 Governments. op. cit.. p. 113, fn. 11. 45 Vood points out that in 1956 Republican gains continued in the central cities with only slight Democratic gains in Los Angeles and San Francisco. In contrast, in spot congressional elections, 1946-1954, Democrats scored 115 gains in the suburbs of St. Louis and Boston. In short, the data is certainly inconclusive and often contrary to the proffered chameleon effect. Certainly, one can make a good case that much of the so-called Republican gain in suburbia was due to the Eisenhower candidacy rather than an increase in support for the Republican Party. While the pattern is visible in 1952, it stands out quite clearly in 1956 when Eisenhower far out-paced his congressional ticket. As David Riesman said of the suburbanite: If he is political at all— rather than parochially civic-minded, . . . he is apt to be an Eisenhower Republican, seldom informed, rather angry, and only spasmodically p a r t i s a n . 1 1 6 This may mean that the Republican Party is not converting Democrats as they move to suburbia. Democrats who move to suburbia will probably continue to vote Democratic, particularly as the segregating effect of suburbia diminishes. As the middle-class suburban hegemony decreases so will the Republican vote. As early as 1954, 115Ibid. ^^David Riesman, "The Suburban Dislocation," The Annals of The American Academy of Political and Social Sciences, CCCXIV (November, 1957), p^ 142. 46 de Grazia suggested that Republican vote in suburbia was attributable to the socioeconomic status differences between 117 suburbia and center city. Berger's study indicates that * 118 blue collar suburbs stay Democratic. A recent study of St. Xouis Catholics, a traditionally Democratic group, who have moved to the suburbs in large numbers, substantiates 119 the segregation hypothesis. This study, based on a sample of 1,800 St. Louis residents, concludes that The most striking finding . . . is, simply, the continuing strength of the tie between the Democratic Party and the Catholic electorate in the St. Louis metropolitan area. Suburbanization has had an effect, and yet we find little difference in party preference between city and suburban C a t h o l i c s .120 Moreover, it is interesting to note that Greer's data, like de Grazia's, and Campbell and Cooper's, shows a smaller "Didn't Vote" figure for suburbanites than for central city residents. J However, St. Louis' suburban Catholics did prefer Eisenhower more than did the central city Catholics. This difference became marked as education increased, and 117 De Grazia, op. cit.. p. 122. 11 8^ . , Berger, op. cit. 119 Scott Greer, "Catholic Voters and the Democratic Party," Public Opinion Quarterly, 25, 4 (Vinter, 1961). 1 20 Ibid., p. 623. See Appendix A-X. 121 Ibid.. pp. 616, 618-619, and 620-621 . 47 1 22 particularly for second generation Catholics. Similarity in party support remained, however, when generation and 1 23 education are controlled. This seems to demonstrate the difference between voting for an attractive candidate and 1 24 abandoning one's party. This would seem to be partic ularly true for the younger voter— the typical suburbanite— who has yet to "fix" his political behavior firmly with his political identification. It appears, at this point, though only tentatively because of deficiencies in data and definition, that the so-called suburban chameleon effect was largely an Eisen hower phenomenon and that the major suburban impact on politics is probably related to the screening or segregation effect and manifests itself more in the areas of metro politan government problems and the larger political process. The Political Process A new dimension Writing in the fourth edition of Politics, Parties. and Pressure Groups, V. 0. Key points out: 1 22 Ibid., pp. 620-21. For the data on the secondary analysis of Greer's data, see Appendix A, XI. 123Ibid., pp. 618-19. 1 24 In another earlier study of a similarly strong Democratic group, Jews, Lawrence Fuchs found that Eisenhower 48 The significance of urbanization for the party system does not come solely from the fact that the Democratic Party developed urban bastions. Rather, the process of urbanization created the raw materials for party cleavages more nearly along class lines than the politics of sectionalism.125 Key also pointed out that Democratic dependence on the urban vote increased before that party's dominance in the cities in the 1930's. Data gathered by the Michigan Survey Research Center on the 1952 election indicates that urban class cleavages may be replacing sectionalism as a major dimension in American politics. Excluding the south, there appeared to be little sectional difference in recent party identifica- 1 26 tion, according to the authors of The Voter Decides. j There appeared, however, to be a suburban-center city i urban sectionalism replacing the older regional section alism. Evidence of the relative newness of this sectionalism is indicated by the fact that Key did not discuss suburban politics (or metropolitan political made considerable inroads into the Jewish Democratic vote in suburbs such as Vestchester, New York; Newton, Massa chusetts; and Vest Hartford, Connecticut. See Lawrence H. Fuchs, "American Jews and the Presidential Vote," American Political Science Review. IL, 2 (June, 1955), p. 387. ^2^V. 0. Key, op. cit.. p. 274. 1 26 Angus Campbell, Gerald Gurin, and Varren E. Miller, The Voter Decides (Evanston, 111.: Row, Peterson and Company, 1954), pT 93. This may have been a transient phenomenon of the Eisenhower candidacy. 49 behavior) in the first edition of his Politics. Parties and Pressure Groups, nor did Edward McChesney Sait in his 1 27 earlier w<?rk, American Parties and Elections. The significance of this sectionalism is found in the rapid suburban growth which has already been cited. As Key said, The real suburban GOP gain came from the far more rapid growth of the total vote in the suburbs than in the central cities.128 Similarly, Samuel Lubell contrasted the per cent of Repub lican pluralities cast by selected suburbs in 1920 and TABLE 8 REPUBLICAN PLURALITIES IN THREE SUBURBS, 1920 AND 1952 Suburb 1920 1952 Philadelphia 8?6 52% New York 8 44 Chicago 7 40 From an almost insignificant (7 to 8 per cent) proportion 1 27 Key, op. cit.. and Edward McChesney Sait, Ameri can Parties and Elections (New York: D. Appleton—Century Company, Inc., 1939). 1 28 Key, op. cit.. p. 276. 1 29 Samuel Lubell, Revolt of the Moderates (New York: Harper and Brothers, 1956), p. 112. 50 in 1920, Republicans by 1952 could (and did) look to the suburbs for a substantial proportion (40 to 52 per cent) of their support. Republicans gained in two ways— suburbs supply a substantial proportion of Republican votes while the Democratic central city Vote is apparently declining in strength, absolutely as well as relatively. Richard Scammon pointed out that the total vote in the fourteen major Standard Metropolitan Areas rose from 20,811,000 in 1952 to 21,218,000 in 1956, but that in every 1 30 one, the total vote in the central city declined. What appears here is not only a new sectionalism but a new population aggregate with a fundamental social change. As Donald Bogue described America in the fifties: It was a decade marked by extraordinary changes in the class structure, particularly in the growth of the white collar class and the spread of suburbia; . . .131 There are gathered in. our metropolitan areas a growing number of people of professional, technical, and white collar skills, while at the same time there is a decrease 1 32 in the number of unskilled laborers. Looking at these 1 30 Richard M. Scammon, "Voting for President in the Larger Metropolitan Areas, 1952-1956," Midwest Journal of Political Science, I, 3-4 (November, 1 9 p^ 330. 1 31 Donald J. Bogue, "Urbanism in the United States, 1950," American Journal of Sociology, LX, 5 (March, 1955), p. 1 3. 132 In addition, of course, the rural farm popula tion is also declining. 51 changing population patterns, Charles Nixon predicted that: . . . it is clear that voting power continues to go to those who experience the problems of urban and metro politan life. It is not going to the older city organi zation, but rather to the surrounding suburban a r e a s . 133 At the present time, as Lubell has stated, the spectacular population shift to the suburbs has produced Republican votes which closely counterbalance Democratic * 1 A central city votes (he cited 1952 figures). Regardless of the Eisenhower phenomenon, this may well mean that the metropolitan, new middle class is well on its way to becom- 135 ing the most powerful single factor in national politics. The political significance of this new sectionalism of class was not lost on either presidential candidate in 1960. According to Theodore H. White, Richard Nixon considered seven states as crucial (California, New York, Illinois, Ohio, Pennsylvania, Texas, and Michigan), while John Kennedy had nine states on his "crucial" list— Nixon’s 1 36 seven plus New Jersey and Massachusetts. Each of these 1 33 Charles R. Nixon, "The Coming Electorate: 1965- 1970." Western Political Quarterly, XIII, 3 (September, 1960), p. 625. 1 34 Lubell, op. cit., p. 112. 1 35 See Arthur N. Holcomb's The Middle Class in Amer ican Politics (Cambridge: Harvard University Press, 1940) for a general discussion of American middle class voting behavior before the post-World War II suburban explosion. 1 36 Theodore H. White, The Making of the President. 1960 (New York: Atheneum Publishers, 1961), pp. 265-66. - 52 states has large metropolitan and suburban populations. In eight of these states the final margin of victory was less 1 37 than four percentage points. The ninth state, Massachu setts, was "crucial" for Kennedy the way Kansas was "crucial" for Alf Landon. Nixon's original campaign was geared to hit the suburbs hard, but it was not until October that Kennedy altered his strategy and turned to the great suburbs too, 1 38 particularly those in the populous northeastern states. The fact that Nixon’s candidacy did not receive the support in suburbia that had been given to Eisenhower probably cost 1 39 him the election as much as any other single factor. With eight more years of observation and experience, Lou Harris changed his evaluation of suburban politics (see his comments circa 1954, p. 1, supra), and in 1962 observed that The world used to be full of small-town people who grew up and came to the city and became Democrats. After 1 37 Kennedy took six of the eight: Illinois (50.0 per cent), New Jersey (50.0 per cent), Michigan (50.9 per cent), Pennsylvania (51.1 per cent), and New York (52.5 per cent). Nixon took two: California by 50.1 per cent and Ohio with the relatively comfortable 53.3 per cent. Kennedy's Massa chusetts vote was 60.2 per cent. Figures from: U.S. Bureau of the Census, Statistical Abstract of the United States; 1961 (82d ed.; Washington, D.C., 1961), p^ 349. 1 38 White, op. cit.. pp. 264, 321, and 353. 1 39 In both Illinois and New Jersey, which Kennedy took with a flat 50.0 per cent, Nixon's suburban losses, compared to Eisenhower's 1956 vote, accounted for more than the margin of his defeat. See: "The 1960 Election," op. cit.. and "The 1960 Election Report," op. cit. 53 World War II city people went to the suburbs and became Republicans. Now city people are going to the suburbs but staying Democrats. If I were advising the Republi cans I'd set great store by what happens in the suburbs.140 Some others have also given serious second thought to Republican suburbia. After the 1952 elections Time Magazine reported: One night last week, Democratic National Chairman Stephen Mitchell quietly slipped into enemy territory and outlined a plan for psychological warfare. Before 300 well-groomed members of the Democratic Club of Evanston, one of Chicago's richest suburbs, Mitchell admitted that the party's in trouble in suburbia.141 Five years later, the New York Times reported that at a recent meeting, Democratic women were told: . . . their party was steadily gaining strength in the traditionally Republican suburbs as the "Country Club Set" was nudged aside by a "Supermarket Set."142 If the suburbs are becoming, in fact, more Demo cratic, the new sectionalism may be hoist to its own petard, a creature and victim of population movement. Aside from ^^Lou Harris, Life, May 11 , 1962, p. 89. 1 4 1 "Psychology in Suburbia," Time, LXII, 17 (October, 1953), p. 25. A precursor, perhaps, of increasing Democratic elec toral strength in the suburbs has been the recent Democratic club movement. Both Francis Carney and Bernard Crick have observed that the membership of the California Democratic Council's clubs is largely suburban middle class and resi dent in Republican areas. See Francis Carney, "The Rise of the Democratic Clubs in California," Case Studies in Practi cal Politics (New York: Henry Holt and Company, Inc., 1958), p. 16; and Bernard Crick, "California's Democratic Clubs: A Revolt in the Suburbs," The Reporter, May 31, 1956, pp. 35-39. 1 42 "Democratic Gains in Suburbia Cited," New York Times, June 8, 1957, p. 21. Note how quickly a "tradition" is established! 54 the drop in Republican presidential vote in 1960, there are other signs that suburbia may be becoming less Republican. Frederick M. Wirt found that while there was a considerable increase in the suburban Republican presidential vote 1948- 1952-1956 with some little increase in the Republican con gressional vote, there was also a lower Republican mayoralty 143 vote. Jerome G. Manis and Leo C. Stine were of the opinion that the heavy Eisenhower vote in suburbia was a reflection of national trends and not a phenomenon unique to 1 44 suburbia. V. 0. Key offered some proof of the latter contention, citing the drop in Democratic vote in central cities from 1948 to 1952 of four and one-tenth percentage points, compared to a drop of four and three-tenths in the 145 suburbs. Suburban "Republicanism" may be a product of other transient phenomena besides the Eisenhower candidacy. David Wallace, studying the changes in the political affiliation of immigrants to Westport, Connecticut (a New York suburb), found that the earlier immigrants were more Republican than 1 43 Wirt, on. cit.. pp. 6-7. 144 Jerome G. Manis and Leo C. Stine, "Suburban Residence and Political Behavior," Public Opinion Quarterly, XXII, 4 (Winter, 1958), p. 489. They precede Wood in this opinion by three years, see pp. 43-44 supra. 1 45 Key, op. cit.. p. 276. 55 146 the more recent immigrants. The newer immigrants were more Democratic than Republican and judging from the per centage of unaffiliated, were more likely to have a polit ical affiliation than the earlier immigrants. TABLE 9 POLITICAL AFFILIATION OF IMMIGRANTS TO WESTPORT BY YEAR OF ENTRY Before 1945 1945-51 1952-55 1956 and Later Republicans (268) 40.2% 30.2 % 28.5% 31 .5% Unaffiliated (225) 33.6 29.1 25.7 20.2 Democrats (331) 26.2 40.7 45.8 48.3 N: 824 214 179 253 178 Indeed, the Republican Party ] had better heed Lou Harris' advice and look at, rather than to, suburbia. G. Edward Janosik's comment on the new sectionalism suggests another modification of the old sectionalism: . . . political loyalties existing in suburban areas prior to the influx of new residents followed no particular pattern. Surrounding such metropolitan regions as Baltimore, Washington, D.C., New Orleans, Houston [sic] and Los Angeles there prevailed an allegiance to the Democratic Party [all of these are southern metropolitan areas except Los Angeles which has a large southern immigrant population], while 146 David Wallace, "The Sociology of Stability and Change in One Suburb's Voting," (unpublished Ph.D. disserta tion. Department of Political Science, Columbia University, 1962), p. 67 (microfilm). 56 New York, Philadelphia, and Chicago were encircled by predominantly Republican areas. 147 It appears, then, that when one speaks of the Suburban Vote he may be paying unwitting homage to the place theory of politics— another form of sectionalism— and ignoring the more fundamental factors underlying the political behavior of the "place." The political salience of the City Vote or the Farm Vote comes not from the place but the groups which reside therein— the fact that one person lives in the city and another on the farm is not the relevant fact. Just as the rural-urban dichotomy is one based on economic-social-cultural differences, so too is the suburban-center city dichotomy based on economic-social— cultural differences. As suburban growth continues to the point that suburbs are more populous than the center city it should not be too surprising if the later immigrants modify the dichotomy, producing a suburbia more nearly like the central city.^® Another new dimension Explaining the growing liberal-conservative split within the Texas Democratic Party, Willie Morris stated that "throughout the ’fifties a large managerial and 1 47 Janosik, op. cit., p. 92. 1 48 With one exception: the visible minorities, par ticularly Negroes, do not seem to be moving into the suburbs in any appreciable number. 57 technical class grew up in the state's expanding metro- 1 49 politan suburbs." It is Morris' opinion that while this new suburban class has retained its inherited party label it has acquired a different political ideology. If the metropolitan areas have been the major source of domestic political issues since the 1930's it may well be that the suburbs will, in turn, be the new source of issues. Just as the politics of the '30's were domestic and class-based, so the issues of the 160's appear to be shaped largely by international events. Campbell and Cooper found that metropolitan suburban respondents were more concerned about world affairs than were central city respondents and that, conversely, they appeared less concerned with existing 1 50 domestic social problems. Dye, applying a McClosky-type C-R five-item battery to his suburban respondents, concluded that there was a consistent areal pattern of resistance to political change or innovation which was directly related to the social 151 characteristics of the area. Several other authors have advanced a suburban psychology. Lundberg and his associates found that suburbanites differed psychologically from 1 49 Willie Morris, "Texas Politics in Turmoil," Harper's. September 1962, p. 82. 1 50 Campbell and Cooper, op. cit., pp. 119, 139, and 143 (Tables B-10, B-56, and B-667T 1 51 Dye, op. cit., pp. 164-65. 58 1 52 central city residents. Later, Sylvia Fava, examining suburbia, found different patterns in the way of life 1 53 between suburbs and central city. Daniel Bell attributed these differences to what he called "status politics and new 1 54 anxieties." Bell's position was that groups advancing in wealth and social position are often anxious and politically feverish— as anxious and feverish as groups that have become declasse. Peter Blau, in his study of social mobility, found a striking similarity in behavior patterns between 155 those moving upward and those moving downward. The idealized suburban polity, according to Scott Greer, would be a homogeneous community, fairly small in size, with a personalized political system in which every 156 one participates. Actually, according to Greer, suburban residents are less interested in local elections, have a lower turnout than their central city brethren, and are less 1 52 George A. Lundberg, Mirra Komarovsky, and Mary Alice Mclnerny, Leisure; A Surburban Study (New York: Columbia University Press, 1934), p"! 42. 153 Sylvia Fava, "Suburbanism As a Way of Life," American Sociological Review, XXI, 1 (February, 1956), passim. ^^Daniel Bell, The End of Ideology (Glencoe: The Free Press, 1960), p. 101. 1 55 Peter M. Blau, "Social Mobility and Interpersonal Relations," American Sociological Review. XXI, 3 (June, 1956). 1 56 Greer, Governing the Metropolis, op. cit.. p. 87. r 59 1 57 less well informed about local community politics. In another publication, Greer has described suburban politics as "trivial" and suburban government as of "limited liability." ^ Thus, the great domestic political problems which are settled in the polling places of the central city: public housing, aid to the handicapped, slum clearance, improvement in race relations, and aid to dependent children are problems which the suburban middle class hopes to escape by moving to suburbia. The problems remain, however, and the only escape managed by suburbanites is from any ability as citizens to influence the decisions. Not only have they removed themselves as political civic leaders but their middle-level cadres have followed them to suburbia; thus those suburbanites who would lead, or at least participate in central city government, find few willing to follow and (or) participate with. It is not surprising that the proposed solutions to central city's problems are not welcomed by the suburbanite. Having abandoned the central city for the trivial politics of suburbia, the only meaningful political arena 157Ibid., pp. 88-89. 1 58 Greer, Emerging City, op. cit.. pp. 142-50. 60 left is the state and national— and to this the suburbanite 1 59 may- turn with a vengeance. Structure of the Dissertation This first chapter, "The Problem and Setting," has been devoted to an examination of the nature of suburbia, its history, and place within contemporary America. After looking at some of the opinions of political scientists and journalists as to the political behavior of suburbia, Chapter I delineated the history of the suburb and its relation to center city. The two causal hypotheses of suburban political behavior which are most often advanced— chameleon and segregation— were then examined in some detail. Finally, the question as to whether or not the suburbs offered a new political dimension was considered. The last section of the chapter discussed the data used, and their limits, in the dissertation. Chapter II, "Theory," presents the theoretical framework underlying the study. Specifically, the theories of class, social mobility, status, and of the individual within the group are discussed. The second half of the chapter attempts to discuss the narrower aspects of these theories as they relate to political behavior. 1 59 It has already been shown that it appears that at the national level suburbanites have a higher voter partic ipation rate than do central city residents and that there are fewer who do not profess a political affiliation. 61 Chapter III, "Suburbia," considers various possible definitions of suburbs and attempts to define them in terms of variables with little or no political bias. For purposes of comparison, two other typologies have been constructed. The first, "Anti-suburbia," is the antithesis of suburbia. The other is a sample of the whole county. The second part of Chapter III describes the population and housing configurations for the three communities. Chapter IV, "Suburban Political Behavior," describes the political behavior of the three communities and the correlations between the political, population, and housing characteristics within each. Chapter Y, "Conclusions and Summary," attempts to go beyond the correlations of Chapter IV and to discuss the causal relations involved in suburban political behavior. The findings in chapters III and IV are considered in relation to the theory of Chapter II and the hypotheses of Chapter I. The Data Description of data Two kinds of data will be employed in this disser tation: (1) election and registration data, and (2) population and housing data. The election data, originally tabulated by precincts for Los Angeles County, cover the years 1958-1962 inclusively. The population and housing 62 data are from the 1960 Los Angeles County census. The fact that political data and census data are not collected or tabulated from comparable areal units presented a problem of major porportions. However, generous support by the University of Southern' California enabled research into the problems of such data collection and collating. Later use of a file prepared by Dr. Dwain Marvick (with some assis— 1 60 tance by the author) solved the problem. While the Marvick file contained almost all of the Los Angeles County election and registration data for the years 1958, 1960, and 1962; and all of the Bureau of the Census data for Los Angeles County for 1960, only some of the data were extracted. The census (population and housing) data extracted included: 1. Census tract number; 2. Percentage of dwelling units owner-occupied; 3. Percentage of dwelling units built since 1950; 4. Percentage of dwelling units built 1940—1950; 5. Percentage of single unit structures; 6. Percentage of structures having two—four units; 7. Percentage of housing units moved into 1958-1960; 8. Percentage of housing units moved into 1954-1957; 9. Median value of property; 10. Percentage of white population; 1 60 Dwain Marvick, "Los Angeles Political Statis tics," a tape record for 1960 Los Angeles census data to which has been added election and registration data for 1958-1962. 63 11. Percentage of Negro population; 12. Percentage of Spanish surname, U.S. born; 13. Percentage living in same house as in 1955; 14. Percentage living in another house in Los Angeles, 1955; 15. Percentage living in another house in California, 1955; 16. Percentage of families with incomes between $3,000 and $1 0, 0 0 0; 17. Percentage of families with incomes over $10,000; 18. Percentage of occupied males in Professional employment; and 19. Percentage of occupied males in Managerial employment. The election and registration data included: E L 1. Percentage of Democratic registration, September 1958; * n 2. Percentage of Democratic registration, September 1960; Q 3. Percentage of Democratic registration, December 1960; E l 4. Percentage of Democratic registration, September 1962; EL 5. Percentage of Democratic registration, December 1962; 6. Percentage of vote, Brown for Governor, 1958;^ 7. Percentage of vote, Engle for U.S. Senate, 1958;^ 8. Percentage of vote, Richards for State Senate, 1958; •L 9. Percentage of vote, Betts for State Treasurer, 1958; I. 10. Percentage of vote, Kennedy for President, 1960; 11. Percentage of vote, Brown for Governor, 1962;^ 12. Percentage of vote, Richards for U.S. Senate, 1962;^ 13. Percentage of vote, Rees for State Senate, 1962;^ 14. Percentage of vote, Rose for Secretary of State, 1962;^ 15. Percentage of vote, Betts for State Treasurer, 1962; 16. Percentage of vote, Richardson for Superintendent of Public Instruction, 1962;c 17. Percentage of vote, Gibson for County Tax Assessor, 1962;c EL Percentages based on two-party totals. bPe rcentages based on vote for two major party candidates. cPercentages based on vote for two major candidates. 64 18. Percentage of vote, "Yes" on Subversive Controls (Proposition No. 24), 1962; 19. Percentage of vote; "Yes" on Water Bonds (Measure No. 1); 20. Percentage of Democrats voting, 1960; 21. Percentage of Republicans voting, 1960; 22. Percentage of Democrats voting, 1962; and 23. Percentage of Republicans voting, 1962. Limits of the data All data used are in the form of area aggregates (by census tract). Certain specific limits exist in the use and interpretation of such data. First, the behavior of aggregates cannot always be 161 imputed to individuals. For example, if a census tract is 75 per cent Negro and 75 per cent Democratic, the pro portion of Negroes who are also Democrats may range from two-thirds to all— this assumes that the age and class distribution for white and Negro is the same, if not the range could well be greater. However, use of aggregate data to describe-the behavior of the aggregates is valid. Second, use of the Pearsonian correlation assumes certain characteristics about the data: (1) that it is bivariate, and (2) that it exhibits a normal distribution pattern. While the political behavior data meet these requirements, some of the population and housing data ■■■■■■ ■■■■ ■! I 4 ^^W. S. Robinson, "Ecological Correlations and Behavior of Individuals," American Sociological Review. XV, 3 (June, 1950), p. 3. 65 clearly do not— most notably the mobility and occupation data . Third, the period of time studied, 1958-19.62, is too short to support valid trend analysis or extrapolation. The short period of time, including only one presidential election and two gubernatorial elections, may well not be typical of California. Moreover, the impact of a resurgent Democratic Party and the end of cross-filing suggest that California politics are in transition and that there will be no "typical" election for some time. Fourth, the area examined, Los Angeles County, is probably more suburban than urban— in this it is not typical of most metropolitan areas. Once described as "seven suburbs in search of a soul," Los Angeles suffers from an overabundance of suburbs. The diversity of suburbs found in Los Angeles County could be untypical of the rest of the nation. However, the work of Schnore and Alford, and Williams, et al., support the position that there are many kinds of suburbs serving many different functions. Thus Los Angeles offers not only the typical differences in suburbs but by use of the single county, political and cultural history are held more constant for the different types than might otherwise be the case. CHAPTER II THEORY The Theory of Class One of the major social concepts identified with suburbia is class. It is impossible to read of or discuss suburban life without direct or implied reliance on the concept of class. Yet, there is no general agreement among sociologists as to what factors, or combination of factors, define a social class. Beyond the general agreement that the concept denotes the horizontal stratification of a popu lation, there is general disagreement on whether class is defined by: economic power, occupation, subjective assign ment, culture, life style, role, education, or (perhaps) motivation. The literature is extensive; for some of the broader considerations of class definition see; Bernard Barber, Social Stratification (New York: Harcourt, Brace, and Company, 1957); Hi Hi Gerth and C. ¥. Mills, From Max Veber: Essays in Sociology (New York: Oxford University Press, 1958}; Joseph A. Kahl, The American Class Structure (New York: Rinehart and Company^ Inc., 1957); Valter Gold- schmidt, "Social Class in America——A Critical Review," American Anthropologist. LII, 4 (October-December, 1950), pp. 483—98; Milton M. Gordon, "Social Class in American Sociology." The American Journal of Sociology. LV, 3 (Novem ber, 1949), pp. 262-68; and Kurt Mayer, "The Theory of Social Class," Transactions of the Second World Congress of Sociology. Vol. II (London: International Sociological Asso- ciation, 1954), pp. 321-335. 66 67 There is considerable contention, not only over the definition of class, -but as to whether or not classes exist in contemporary America as meaningful discrete groups. One sociologist, Oliver Cox, insisted that there is no such thing as an objective class amenable to physical identifi cation in America, that social class is merely a heuristic 2 concept significant only to the person using it. His is the position that the social class gradient is a continuum. Kurt Mayer has pointed out: Modern social classes are no longer visibly marked off from each other or demarcated by tangible boundaries. They have no legal standing; nor are they organized groups. There are no official, rigid criteria of class position. All the upper and middle class symbols are theoretically accessible to anyone with the necessary wealth to purchase them. . . while Bernard Barber has suggested that • . . such evidence as there is seems to show that [for the general population] there is a considerable amount of either ignorance or vague knowledge about the system of stratification.4 Social classes seem to be less well defined in America today than they were fifty years ago; this is particularly true in metropolitan areas where there is a less visible elite and the symbols of status may be 2 Oliver Cox, Caste. Class, and Race: A Study in Social Dynamics (New York: Doubleday and Company, 1948), p. 305. 3 Mayer, op. cit.. p. 322. 4 Barber, op. cit.. p. 197. 68 displayed in cultural anonymity.^ Gerhard Lenski has offered some data to substantiate the ambiguity of "class." He reported that a panel of twenty-four long-time residents (seventeen years-to—life) of Danielson, Connecticut, were asked to rank 150 families (out of a total population of % 6,000) as to their relative standings. The number of cate gories established and used by each panel member ranged from three to seven, and all had difficulty in defining the exact limits for each category used. Sociologists, explaining classes, as opposed to describing (or denying) their existence usually have taken a functional point of view. Kingsley Davis and W. E. Moore regarded class as a functional necessity for all complex societies, serving to motivate and place people within the 7 system. It has also been suggested that a class structure serves to integrate society, order relationships, and 5 • ' Gideon Sjoberg, "Are Social Classes in America Becoming More Rigid?" American Sociological Review. XVI, 6 (December, 1951), p. 783; and see also, William H. Form and Gregory P. Stone, "Urbanism, Anonymity, and Status Symbolism," The American Journal of Sociology. LVII, 5 (March, 195771 pp. 504—14. A corollary of this elite anonymity might well be a power anonymity. ^Gerhard E. Lenski, "American Social Classes: Statistical Strata or Social Groups?" American Journal of Sociology. LVIII, 2 (September, 1952), pp. 141-43. 7 Kingsley Davis and W. E. Moore, "Some Principles of Stratification," American Sociological Review. X, 2 (April, 1945), pp. 242-249. 69 g provide a context for stability. Aside from validation, sociologists have also been attempting*to define class. There are two major criteria, the objective and the subjective. Objective definitions of class include: income, occupation, residence, religion, education, and family origin. Generally speaking, the objective point of view maintains that classes, regardless of criteria, are determined outside the consciousness of the individual. The alternative has been perhaps best expressed by Robert M. Maclver and Charles H. Page: No matter what objective criterion we use, we do not have a social class unless class consciousness is presentT^ Richard Centers maintained that class, as distin guished from stratum, is a psychological phenomenon; that a man's class is part of his ego, a feeling on his part of belonging to or identification with something.^ The two criteria do not conflict in a majority of cases. Heinz Eulau compared objective and subjective classifications and found that a substantial majority of subjective class g Talcott Parsons, "Social Classes and Class Conflict in the Light of Recent Sociological Theory," American Economic Review. XXXIX, 3 (May, 1949), pp. 21-22, 9 Robert M. Maclver and Charles H. Page, Society (New York: Rinehart and Company, 1949), p. 350. ^Richard Centers, The Psychology of Social Class (Princeton: Princeton University Press7 1949), p^ 27. Stratum, for Centers, is objective, determined by occupation. identifications agreed with objective classification.^ TABLE 10 SELF-IDENTIFICATION BY OBJECTIVE SOCIAL CLASS, 1952 AND 1956 Objectively Classified Self-Identification Middle Class Working Class 1952 1956 1952 1956 Middle class 6 4 % 7 0 % 2496 2 2 % Working class 36 30 76 78 N: 366 402 766 856 Concern with the identification of class appears to be somewhat more than heuristic, as Cox charged. Harold W. Pfautz, in summarizing the results of many studies, found definite behavior differences which appeared to coincide 1 2 with class differences. These class-related differences Heinz Eulau, Class and Party in the Eisenhower Years (New York: The Free Press of Glencoe, 1962), pi 54. EulauTs study is based on national data collected by the University of Michigan*s Survey Research Center (hereafter referred to as "SRC." Ivan D. Steiner, in a much more limited survey of Dearborn, Michigan, high school students, found that about one-third subjectively identified with a class to which they had not been assigned on the basis of objective criteria. Ivan D. Steiner, "Some Social Values Associated with Objectively and Subjectively Defined Social Class Memberships," Social Forces, XXXI, 4 (May, 1953), pp. 328-29. 1 2 Harold W. Pfautz, "The Current Literature on Social Stratification: Critique and Bibliography," The American Journal of Sociology. LVIII, 4 (January, 1953), pp. 402-404. Covers period from 1945 to 1952 plus a few earlier works recently translated, but excludes Ph.D. dissertations. See also Jerome K. Myers and Bertram H. 71 included: stability of marriage, reaction to crisis, adult- parent adjustment, veteran*s adjustment, differences in child training, concepts of parental role, concepts of marital roles, leisure activities, drinking mores, prefer ence in personality traits, and sexual behavior. The objective criteria The fact that sociologists cannot agree on the hallmarks of class is not too surprising, since those whom the sociologists would classify cannot agree either. In a national NORC poll taken in September 1949 a sample was asked, In deciding whether a person belongs to your class or not, which of these things do you consider most important to know? . , .13 TABLE 11 CRITERIA FOR CLASS Criteria Per Cent by Response Sort of work 23 Family 7 Amount of money 9 Education 8 Beliefs and feelings 38 Other 4 Don't know 11 Roberts, Family and Class Dynamics in Mental Illness (New York: John Wiley and Sons,Inc., 1959); and August B. Hollingshead and Frederick C. Redlich, Social Class and Mental Illness (New York: John Viley and Sons, Inc., 1958). 1 3 Mildred Strunk, "The Quarter*s Polls," Public Opinion Quarterly, XIII, 4 (Vinter, 1949-50), p. 711. 72 While the single largest response was subjective, still a majority placed first reliance on objective 1 4 criteria, and of these, occupation came firsto Occupation is not just a job. The occupation one follows fills most of one's waking time. Occupation has much to do with determining the location and kind of residence of the family, the children's playmates and 1 5 schooling, social contacts, and leisure time activities. Occupation, as the criteria for class, has become more and more complex since the Industrial Revolution. Much of the ambiguity regarding classes (compared to a hundred years ago) can be attributed to the impact of industrialization on all social structures. Industrialization means that all social institutions which interfere with the functioning of the economy are minimized— family and kinship, religion, and rigid social stratification.^ As Talcott Parsons says: 1 4 Similar results are reported by Centers, in his Psychology of Social Class, op. cit., published in the same year as the NORC poll (1949)• He reports that respondents ranked beliefs and attitudes first as criteria for deter mining class. See Centers, op. cit.. p. 91. 1 5 Dewey Anderson and Percy E. Davidson, Ballots and the Democratic Class Struggle (Palo Alto: Stanford Univer- sity Press, 1943), p^ 82. ^Talcott Parsons, "Social Strains in America," The New American Right. Daniel Bell, ed., (New York: Criterion Books, 1955), p. 121. 73 The United States of course has a class structure, but it is one which has its primary roots in the system of occupational roles. . . . In America, . . . it is clearly the occupational system rather than kinship continuity that prevails.17 Sociologists have found occupation a reliable guide to class and prestige in other industrial nations. They have also found a high level of agreement between the populations of these industrialized nations as to the rankings of various occupations, which suggests that it is industrialization, per se. which gives rise to the class- prestige rankings of occupation and not national culture, 1 8 or some other national variable. Sociologists have found, in particular, that occupation, rather than income, is the better guide to socioeconomic rankings in the United 1 9 States. The individualTs status in the community, 17Ibid. 1 8 Alex Inkeles and Peter H. Rossi, "National Com parisons of Occupational Prestige," American Journal of Sociology, LXI, 4 (January, 1956), pp. 329-39. The six nations studied and found to have similar rankings of occu pation were: Germany, Great Britain, Japan, New Zealand, Russia, and the United States. France, although an indus trialized nation, is measured by French socialogists at the Institut National d*Etudes Demographiques by three criteria: Standard of life (ownership of property, ownership of an automobile, use of domestic help, and ownership of a tele phone); education; and the average age of the individuals profession (the higher the age the higher the prestige). Marcel Bresard, "La Mobilite Sociale: Le Choix d*une echelle sociologique," Transactions of the Second World Congress of Sociology, Vol. II (London: International Sociological Asso- ciation, 1954), pp. 399-401. 1 9 See V. Lloyd Warner, Social Class in America (Chicago: Science Research Associates, 1949); J..A. Kahl and 74 according to Morris Rosenberg, depends largely on the work 20 he does and how well he does it. How others rank him, and how he ranks himself will depend on his occupation and, as a result, the individual plays the role of his occupa tion. The doctor or the teacher has a certain role to play, and the role, in turn, commits the individual to certain 21 patterns of thought and behavior. In many occupations there is an ideological content as well as a technical 22 skill required. How one "chooses" his occupation depends, in turn, on his values, attitudes, and personality needs as well as the opportunity for choice. There also appears to be some relationship between father's income and student's occupational choice. A similar relationship exists between student's I class and occupational choice. Students from the upper class tend to choose business or free professions more 23 often than do students in the middle class. J. A. Davis, "A Comparison of Indexes of Socio-economic Status," American Sociological Review, XX, 3 (June, 1955), pp. 317-25; or 0. D. Duncan and Beverly Duncan, "Residential Distribution of Occupational Stratification," American Journal of Sociology. LX, 5 (March, 1955), pp. 493-503. 20 Morris Rosenberg, Occupations and Values (Glencoe: Free Press, 1957), p. 2. 21 Ibid. 22Ibid.. p. 81. 2^Ibid.. p. 55. 75 TABLE 12 FATHER*S INCOME AND STUDENT*S OCCUPATIONAL CHOICE24 Father*s Income Occupational Choice Business and Salaried Free Profession Profession Other Per Cent N -$7,500 38$ 45 17 100 1 ,367 $7,500-$10,000 42 $ 32 26 100 324 $10,000-$20,000 56$ 26 18 100 336 $20,000-$30,000 68$ 21 11 100 135 $30,000+ 71$ 16 13 100 137 One of the major occupational dimensions which seem to divide people is the "heads or hands" or white collar- blue collar dichotomy. White collar workers, apparently for status reasons, fear to identify with or associate with the 25 dirty-handed blue collar workers. Robert R„ Alford has suggested that the move across the manual-non-manual "line" has substantial social significance— more than shifts within the two categories.2^ Another objective measure of class— income— appears not to be as good an index as occupation except at the extremes. Within the broad middle range of income, how one 24Ibid.. p. 56. 25 Daniel Bell, "The Capitalism of the Proletariat," The End of Ideology, Daniel Bell, ed. (Glencoe: Free Press, T960), p. 216.- ™ 26 Robert A. Alford, "A Suggested Index for the Asso ciation of Social Class and Voting," Public Opinion Quarterly, XXVI, 3 (Fall, 1962), pp. 419-20. obtained the income and what one spends it for appears to 27 be more significant than the amount obtained. Similarly, residence appears not to be too closely related to class. It has been suggested that there may well be a time lag between achievement of a given class level and the purchase 28 of a comparable residence. Barber, however, has taken the position that residence may be a common measure of class for all societies— since so much of the individual’s 29 life is centered about his home. Barber also found religion to correlate with class, not only as between the major denominations of Protestantism, Catholicism, and Judaism, but also within the Protestant denominations.3^ For example, Episcopalians, Congregationalists, and Presbyterians have a larger proportion of upper class affiliates and a smaller proportion of lower class affiliates than do Baptists, Lutherans, and Christian 31 churches. Catholic church membership, "organized" by parish lines, is not quite as reliable a guide to class. 27 Kahl and Davis, op. cit.. p. 322. p f t Ibid.. p. 321. Kahl and Davis thought of the dif ference as a time lag, but with contemporary methods of home finance one may acquire a residence with a status above one’s own status at time of purchase— the lag is reversed! 29 Barber, op. cit., p. 144. 30Ibid., pp. 155-57. 31 Ibid., p. 157. Barber’s source was the Federal Council of Churches* Information Service. XXVII, May 15, 1948. 77 Another category, which is neither objective nor subjective but a combination of both— life style— has been found to correlate to class in a number of community 32 studies. Gerth and Mills have pointed out that status is normally expressed by a specific style of life which is 33 expected of all those who belong to a given class. Seymour Lipset and Hans Zetterberg have suggested that different style of life choices are available— that at the same occupational-income level individuals will vary to some extent in style of life— and that these may be referred 34 to as consumption classes. But, as Barber has pointed out, the anomalous display of social class symbols appears 35 to have modified style of life as a key to class. The subjective criteria Richard Centers, in The Psychology of Social Classes, took the position that social classes are inter nally cohesive and functional social—psychological 36 groupings. Centers* position was that "class" has 32 See A. B. Hollingshead, "Selected Characteristics of Classes in a Middle Vestern Community," American Sociol ogical Review. XXI, 4 (August, 1947), pp. 385-95. 33 Gerth and Mills, op. cit., p. 187. 34 Seymour Martin Lipset and Hans L. Zetterberg, "A Theory of Social Mol lity," Transactions of the Third World Congress of Sociology. Vol. Ill(London:International Sociological Association, 1956), p. 157. 35 Barber, op. cit., p. 162. See also Sjoberg, op. cit. ■^Centers, op. cit. 78 subjective content, and that there is a class conflict 37 thereby. Class, for Centers, carries the burden of shared values, interests, objectives, and needs which are 38 rooted in common economical or political aspects. Thus, Class . . . can well be regarded as [a] psychologi cal phenomenon . . . a man's class is part of his ego, a feeling on his part of belongingness to something; an identification with something larger than himself.*9 The use of subjective identification has its limits. National surveys run by Gallup, Fortune, and Cantril in 1939, 1940, and 1941, respectively, found that on the average over eighty per cent of respondents identified with 40 the middle class. However, when the term "working class" was added to researchers' lists, it was found that a majority of respondents then identified with the working class rather than the middle class.^ Centers, as an early exponent of subjective class 37Ibid.. p. 23. - 3^Ibid., pp. 26-27. 40 See Ibid.. pp. 30-31, for a summary of these studies. 41 See Angus Campbell, Philip E. Converse, Warren E. Miller, and Donald E. Stokes, The American Voter (New York: John Wiley and Sons, Inc., 1960), p. 158. So few respon dents identify with either the upper or lower class that some researchers have dropped these alternatives from their studies. See, for example, Centers, op. cit.; Kahl, op. cit.; and Campbell, et__al., op. cit. 79 identification, has also been criticized for method and 42 concept. Such criticism basically denies the validity of assuming that there is a relationship between self- identification and a class consciousness which produces class behavior or class attitudes. Criticism has also been levied against Centers since he offered respondents their choice of limited classifications— the method (a checklist) 4 3 assumes the result. Centers’ data appears to have pro vided a new insight into the dimensions of class political behavior. The problem in using subjective classifications is that cause and effect are not at all clear. Kahl, in examining Centers’ data, found a slightly higher predicta bility of ideology when occupation was varied than when class identification was varied (as did this writer).^ 42 See Ely Chinoy, "Research in Class Structure," Canadian Journal of Economic and Political Science. XVI, 2 (May, 1950) , pp. 255-63; H"I J\ Eysenck, "Social Attitude and Social Class," British Journal of Sociology. I, 1 (March, 1950), pp. 55-66; Goldschmidt, op. cit.. pp. 483-98; and Llewellyn Gross, "The Use of Class Concepts in Sociological Research," American Journal of Sociology, LIV, 3 (March, 1949), pp. 409-21. 43 Arthur Kornhauser, "Public Opinion and Social Class," The American Journal of Sociology, LV, 4 (January, 1950), p. 339. Joseph Kahl found that about one-third of those who self-identified as "middle class" in an open ques tionnaire put themselves in the "working class" on a closed questionnaire; most of those who did so were skilled or semi-skilled workers. Professionals, businessmen, salesmen, and clerks put themselves in the "middle class" on both open and closed questionnaires while manual workers consistently replied "working class." See Kahl, op. cit., pp. 169-70. 44 Kahl, op. cit.. p. 165. (See correlations, this dissertation, Chapter II, p. 165.) 80 Kahl made a "cautious interpretation of the data," suggest ing that perhaps both class consciousness and ideology are 4 5 consequences of occupational position. Some sociologists have suggested that social role is 46 an index of social class. Heinz Eulau has suggested "role" as a basic ;.unit of analysis in the linkage between institutions (e.g., class) and person in the political system* The basic unit of institutional analysis is position. The basic unit of behavioral analysis is act. What links act and position is the role a person takes, or is expected to take, in the performance of those acts that are relevant to the position he occupies in an institutional o r d e r .47 Eulau emphasized expectations because role is thought to be 48 determined by what others expect. Eulau discussed the individual as playing a "citizen role" within the political 49 system and a "class role" within the class system. He 45Ibid., p. 166. 46 T. H. Marshall, Citizenship and Social Class (Cambridge: Harvard University Press, 1950), pp. 92-3; and Kornhauser, op. cit.. p. 338. 47 Eulau, op. cit.. p. 23. Eulau is taking up Ralph Linton’s "dynamic aspect o$ status" and Theodore M. Newcomb’s "ways of behaving which are expected." See: Ralph Linton, The Cultural Background of Personality (New York: D. Appleton-Century, Inc., 1945), p. 50; and Theodore M. Newcomb, Social Psychology (New York: Dryden Press, 1950), p. 280. 48 Eulau, op. cit.. p. 24. 49 Ibid., p. 28. There are some problems here. Since role involves interaction with another there must be 8T examined the problem of role conflict (subjective vs, objective) and found that one role or the other tends to 50 dominate. Obviously, unless the researcher identifies this dominant role, he may well categorize the respondent by the minor role category, attributing behavior or opinion to the minor role which would then lead to false conclusions. KahlTs data suggested that there is a fairly sizable minority who may well be in a role conflict situation (supra, fn. 43). The majority appeared to be without role conflict. Ivan D. Steiner examined the value systems of those who had class role conflict and those who 51 did not. Steiner found no pattern or values which suggested that those with class role conflict tended to define the dominant role as either subjective or objective; those with conflicting class roles did not resolve the another to interact with; in political systems this is no problem, but in defining the role of a child there are obvious variations with and without a parent. Similarly, the role of the middle class seems to be at best poorly defined— there are more likely many middle class roles, since there seem to be several middle classes— "middle class" is too broad a concept to connote a single class role. Of greater significance is the question of whose expectations do the middle class strive to fulfill? The concepts of role, reference group, identification, and "other directed" all appear to overlap. 50Ibid., p. 173. 51 Steiner, op. cit. 82 52 conflict in any consistent manner. Finally, one should note that there is a rather high correlation between objective and subjective class criteria. Centers reported that his respondents, by 5 strata, were fairly consistent in picking their own social class. As can be seen in the following table, -there was a clear-cut majority in each strata who picked either "middle 53 class" or "working class." The correlation between class 54 and occupation for this data is +.541. TABLE 13 SUBJECTIVE CLASS BT STRATA55 Occupation Class Strata Upper Middle Wo rking Lower N Large business 1 3 % 78 7 54 Professional 4 % 81 10 — 73 Small business 3# 70 24 — 131 White collar 2fo 61 34 — 172 Skilled-manual 2 % 26 71 1 163 Semi-skilled 1 io 14 83 1 174 Unskilled — 18 75 7 77 52Ibid., p. 330. 53 Centers, op. cit., p. 86. 54 See p. 146, fn. 252, for discussion of this correlation. 55 Percentages do not add up to 100 due to elimina tion of the "Don*t know" and the "Dontt believe in class" categories. 83 Culture and Class Bernard Barber has pointed out that we do not know much about the personality correlates of social roles in general or of social class positions. Another fruitful approach to class is the cultural concept. The cultural approach to class is based on two assumptions: (1) that classes are "little worlds" and (2) that the process of growing up in a class is reflected in the individual’s 57 personality structure. The concept of life style alluded 5 8 to above implies that social classes are also cultures. Talcott Parsons has suggested that: There seem to be inherent tendencies for those who are structurally placed at notably different points in a differentiated social structure to develop different "cultures."59 Jerome K. Myers and Bertram H. Roberts made the point that a person’s behavior, thoughts, sentiments, and values are molded largely by his culture, citing the "^Barber, op. cit., p. 302. See the Pfautz listing and other citations in fn. 12, p. 70. 57 Milton M. Gordon, "Kitty Foyle and the Concept of Class as Culture," American Journal of Sociology, LIII, 3 (November, 1957), p. 210. 58 Pfautz, op. cit.. p. 403. 59 Parsons, "Social Class and Class Conflict," op. cit.. p. 23. Ann Roe points out that different occupa tional groups have different customs and even different standards of conduct. See Ann Roe, The Psychology of Occupations (New York: John Wiley and Sons, Inc., 1956), p. 298. 84 findings of Margaret Mead that adolescents in Samoa suffered none of the "storm and stress" common to the American ado1e s c ent.^ M. Brewster Smith and his colleagues took the position that each culture creates special "plights" to which the individual in that culture must adjust. The common problems of childbirth, food, death, etc., are handled differently. Different ways of life create differ ent needs, different responses, and lead to different 61 personality traits. Thus, the sociologist, psychologist, and anthropologist often approach class not by its demo graphic characteristics, but by descriptions of the needs, experiences, and personalities of the class members. August B. Hollingshead and Frederick C. Redlich found that, "a definite association exists between class 6 p position and being a psychiatric patient." They found that the lower the class the greater the proportion of 6 3 patients. They also found a significant relationship ^°Myers and Robert's, op. cit.. p. 10. They cited Margaret Mead*s Coming of Age in Samoa (New York: W. Morrow and Company, 1921T) and Ruth Benedict*s Patterns of Culture (Boston: Houghton and Mifflin, 1934). 61 M. Brewster Smith, Jerome S. Bruner, Robert W. White, Opinions and Personality (New York: John Wiley and Sons, Inc., 1956). 6 2 Hollingshead and Redlich, op. cit., p. 216. 63Ibid., p. 210. 8 5 between class position and type of disorder, those in-the higher class positions being more neurotic while those in 64 the lower class positions were more psychotic. There appears, then, to be significant differences in personality disorders by class position. Another study of personality disorders by occupation, published at the same time, found 65 a relationship between suicide rates and occupation types. Powell found a lower suicide rate for blue collar than for white collar workers; the pattern of suicide rates suggested to him the existence of different success ideologies oper ating as a basis for the anomic behavior.^ Thus, the existence of differing suicide rates by class and the relationship of class position with types of personality disorders tends to substantiate the class-as-culture theory. Another attribute of the class-as-culture theory is the relationship of motivation and aspiration to social class. One sociologist defined middle class in terms of aspirations: A member of the middle class . . . is an individualist who may be anywhere in the social scale from the bottom to the top. The fundamental characteristic which marks him out as a member of the middle class is not his position in the social hierarchy [to which he refers 64 Ibid., p. 223. This finding may be a function of definition since most psychologists are not lower class. 6 5 Elwin H. Powell, "Occupation, Status, and Suicide," American Sociological Review. XXIII, 2 (April, 1958), data based on Tulsa, Oklahoma, from 1937 to 1956. 66Ibid., p. 137-39. 8<? as "from the bottom to the top"], but rather his atti tude toward the society in which he lives. If he is determined to move into a preferred position in the social scale, he is middle class . . . When he ceases to seek a higher standing and concentrates upon main taining the position he now has, then he is no longer a member of the middle class.67 However, as Barber suggested, lower levels of aspiration for the working class may well be the result of a realistic appraisal of life-opportunity rather than a 68 personality variable. The Theory of Social Mobility Like class, social mobility is also closely linked to suburbia. In fact, suburbia has been viewed as a resi dential expression of upward social mobility. Social mobility exists when class lines are not rigid— the greater the mobility the less charply defined are these class lines. One indication of a relatively high rate of social mobility is the relative ambiguity of class labels. The ultimate in social mobility would be the complete lack of class— at least in logic. Actually, the concept and virtue of social mobility depends upon the existence of a socially recognized class structure. To be socially mobile one must be able to fi 7 Henry Grayson, The Crisis of the Middle Class (New York: Rinehart and Company^ Inc., 1955), p. xi. This is not to be confused with another publication, by Lewis Corey, The Crisis of the Middle Class (New York: Covici-Friede, 1935T* Grayson*s simple definition has a fatal logical flaw— he uses a dichotomous variable within a trichotomous framework, to which class does the non-aspirer belong, upper or lower? 68 * Barber, op. cit., p. 309. distinguish between a class of origin and a class of aspira tion or attainment.' The term "social mobility" refers to the process by which individuals move from one position in 69 the social hierarchy to another. However, as Charles F. Vestoff, et al., pointed out, the term "move" and the con cepts of social hierarchy are too ambiguous to make them 70 operational tools of research. Social mobility implies "getting ahead"— but of what, or of whom? Further, it appears that social mobility is oriented to different goals for different strata of society. For the manual worker or salesman, upward mobility means owning one's own business; for the white collar class, mobility usually occurs within 71 the bureaucracy of government or large scale industry. For this study, social mobility has meaning, for as Lipset and Bendix have said, A person who moves up in the social hierarchy will tend to change his friends, join new organizations, move to 69 See Seymour Martin Lipset and Reinhard Bendix, Social Mobility in Industrial Society (Berkeley and Los Angeles: University of California Press, 1959), pp. 1-2; also see Morris Rosenberg, Occupations and Values, op. cit. Natalie Rogoff, Recent Trends in Occupational Mobility (Glencoe: The Free Press, 1953); Peter H. Rossi. Why Families Move (Glencoe: The Free Press, 1955); Pitirim Soro kin, Social Mobility (New lork: Harpers, 1927); and Charles F. Vestoff, Marvin Bressler, and Philip C. Sagi, "The Con cept of Social Mobility: An Empirical Inquiry," American Sociological Review. XXV, 3 (June, 1960), pp. 375-85. 70 Vestoff, et al., op. cit.. p. 376. 71 Lipset and Bendix, op. cit.. pp. 172-79. 88 a new neighborhood: perhaps he will even change his religious affiliations; in some cases he will change his name; often he will alter his political attitudes.»2 Broadly speaking, one may define social mobility in one of two ways: inter-generational or intra-generational, that is, by the reference points of movement. Inter- generational measures of social mobility usually compare father to son, while intra-generational measures of social mobility compare the same individual at different times in his life. Social mobility has been related in various studies to such factors as education, intelligence, and 73 country of birth. Vestoff, et al., made an attempt to evaluate the interchangeability of indices of social mobility (occupation, education, and income) and found that they are not interchangeable, concluding that social 74 mobility is a multi-dimensional concept. Urbanization Social mobility seems also to be a function of the rates of change within society: ^ Ibid. , p. 6. 73 For education, see Richard Centers, "Education and Occupational Mobility," American Sociological Review, XIV, 1 (February, 1949), pp. 143-44; for intelligence, see Elbridge Sibley, "Some Demographic Clues to Stratification," American Sociological Review, VII, 3 (June, 1942), pp. 322-30; and for country of birth, see Percy E. Davidson and H. Dewey Anderson, Occupational Mobility in an American Community (Stanford: Stanford University Press, 1937). ^Vestoff, et al., op. cit. . pp. 384-85. 89 . . . societies living under new or changing conditions are usually characterized by a wealth of achievable statuses . . .75 The organization of society has a direct impact on the accessibility of achievable status vis a vis ascribed status. It appears that industrial-urban societies have higher rates of social mobility than do rural-agricultural 7 fs societies. Lipset and Bendix reported studies of various urban centers which indicated that there is a relationship 77 between urbanization and social mobility. This is, apparently, not restricted to the "western” cultures, since high social mobility has been found in Tokyo, Japan and 78 Poona, India. One apparent reason why social mobility, particu larly upward social mobility, is higher in urban areas is 7*5 Ralph Linton, The Study of Man (New York: Apple- ton-Century-Crofts , Inc.~J 1 936) , pp. 1 29-1 30. 7 f i Lipset and Bendix, op. cit., passim. 77 They listed studies of Helsinki, Finland; Stock holm, Sweden; Sao Paulo, Brazil; Kansas City, Missouri; Aarhus, Denmark; and Indianapolis, Indiana. Both social mobility and urbanization may be effects of an expanding economy— correlation does not establish causation. See Lipset and Bendix, op. cit.. pp. 28-33. 78 Lipset and Bendix cite: A. G. Ibi, "Occupational Stratification and Mobility in a Large Urban Community: A Report of Research on Social Stratification and Mobility in Tokyo, II," Japanese Sociological Review, IV (1954); and N. V. Sovani and Kusum Pradhan, "Occupational Mobility in Poona City Between Three Generations," The Indian Economic Review, II (1955). Lipset and Bendix, op. cit., p. 29. Their citations are incomplete. 90 the expansion of occupation opportunities and the concen tration of bureaucratic (typical white collar) enterprise in metropolitan areas. Thus, Lipset and Bendix found higher rates of social mobility in urban areas than in 79 rural areas. The root of this appears to be in the more complex division of labor in urban areas combined with the disproportionally greater growth of urban area populations. The larger the city the higher the percentage of nonmanual occupations. Similarly, the status of a man's occupation is related to the size of his community-of-orientation. A smaller proportion of farm children grow up to occupy white 80 collar jobs than do metro children. Lipset and Bendix, in examining the occupations of fathers and sons by community size, found a consistently greater proportion of sons in the nonmanual category, despite father’s occupation, when their community-of-orientation population was greater than 81 % 250,000. According to Lipset and Bendix, it is clear from the data above, that working class youth growing up in large cities are much more likely to reach a higher occupational level than those coming from smaller communities.82 79 Lipset and Bendix, op. cit.. p. 213. 80 Ibid., p. 205. See Appendix A, Table 122, for complete data. Also see: Seymour Martin Lipset, "Social Mobility and Urbanization." Rural Sociology, XX, 3-4 (September-December, 1955), pp. 220-28. 81 Ibid., p. 207. 82Ibid., p. 206. TABLE 14 O'! DISTRIBUTION OF OCCUPATIONS BY SIZE OF COMMUNITY Size of Community Urban Rural Non-Farm Rural Farm Occupation 250,000+ 25,000- 250,000 2,500- -2,500 25,000 Nonmanual Manual Farm Not reported 40.8/ 57.8 0.4 0.9 37.4% 60.7 0.9 0.9 34.6/ 26.2/ 61.2 62.9 3.2 9.5 1.1 1.4 5.3/ 18.2 75.4 1.1 TABLE 15 INTER-GENERATIONAL SOCIAL MOBILITY BY SIZE OF COMMUNITY-OF-ORIENTATION84 Son1 s Occupation Fatherfs Occupational Group Professional and Business White Collar , Semi- and and Sales OKiiiea Unskilled Nonmanual N: " Nonmanual N: Communities 31?6 11 5 Communities 28/ 106 Over 250, 37/° 30 Under 250 21/ 52 000 Population 65/ 97 ,000 Population 43/ 106 54/ 55 50/ 70 d Si ■ 83 Ibid., p. 218., Lipset and Bendix use 1950 Census 84. While the data is consistent in direction, only the differences in the "Skilled" category are statistically significant at the .05 level by chi square. 92 Examination of education and social mobility appears to modify the supposed impact of community size, indicating that much of the social mobility advantages of growing up in 85 a large city is due to better educational opportunities. One other factor examined by Lipset and Bendix in the dif fering mobility rates between small towns and large cities was religion. First generation differences do appear, but 86 they disappear by the third generation. Residential mobility Residential mobility is the sine qua non of suburbia as a geographical expression of social mobility. Lipset and Bendix state that . . . the successful•individual need only change his residential neighborhood to bring his economic and his social status into line.^7 There is a question as to which comes first, social or residential mobility (see fn. 28). Richard Scudder and C. Arnold Anderson found that sons who left a small Kentucky town were more likely to rise above their parents’ 83Ibid.. p. 213. 86 Ibid., pp. 49-50. Religion does not appear to affect aspiration levels either, according to Lipset and Bendix, ibid., p. 52. However, E. Digby Baltzell, in his Philadelphia Gentlemen: The Making of a National Upper Class (Glencoe; Free Press, 1958), suggests that the upwardly mobile change churches, which implies that aspira tion may affect religious affiliation. 87 Lipset and Bendix, op. cit., pp. 66-67. 93 88 occupational status than sons who remained. As Scudder and Anderson point out, It is to be expected that individuals who leave their home communities become detached from stable status relationships and manifest unusual m o b i l i t y .89 Lipset and Bendix describe the migration-mobility cycle as one in which lower class migrants to large cities occupy the lower class occupations and residences while native lower class urbanites move upward (and probably "outward" to suburbia).^ The result of this combination of social mobility and residential mobility may be that old friends and old norms are replaced by new friends and new 91 norms. Occupational mobility The most often used measure of social mobility is 92 occupational mobility. However, as Lipset and Bendix 88 Richard Scudder and C. Arnold Anderson, "Migration and Vertical Occupational Mobility," American Sociological Review. IXX, 3 (June, 1954), p. 330. 89Ibid., p. 329. 90 Lipset and Bendix, op. cit.. p. 216, make the com ment that urbanites seldom move to small cities and that their residential mobility occurs within the large city or between large cities. 91 See Norman Kaplan, Reference Group Theory and Voting Behavior (unpublished Ph.D. dissertation. Department of Political Science, Columbia University, 1955), passim (microfilm); and Anderson and Davidson, op. cit.. p. 106. 92 There is, in fact, a tendency in many studies of social mobility to treat occupation as an adequate single 94 point out, it is often useful to separate occupational from economic status, since amount of income is not necessarily the best indicator of occupation status, or style of life, 93 although it is obvious that there is some relation. * Studies of social mobility which rely on occupational mobility make the assumption that there is some popular con sensus on the relative status of the different occupations— 94 and, in fact, there is. However, there is also a close relationship between occupation and education, occupation and income, occupation and age, and occupation and community-of-orientation (as we have seen),^ In looking at inter-generational mobility, Rogoff found that from 60 to 75 per cent of the population are 96 engaged in occupations other than those of their fathers. Given the large number of occupations which differ more in index of social mobility and to use the terms almost inter changeably. See Vestoff, et al.. op. cit.. p. 378. 93 Lipset and Bendix, op. cit.. pp. 272-73. 94 National Opinion Research Center, Opinion News. IX (September 1, 1947), pp. 3-13, reprinted in Class. Status and Power. Reinhard Bendix and Seymour Martin Lipset (eds.) (Glencoe: Free Press, 1953), pp. 411-26. See also, M. E. Deeg and D. G. Paterson, "Changes in Social Status of Occu pations," Occupations. XXV, 4 (January, 1947),*pp. 237-41; and Inkeles and Rossi, op. cit.. pp. 329-39. 95 For education, see: Richard Centers, "Education and Occupational Mobility," op. cit.; for income and age, see Rogoff, op. cit., p. 19. 9 6 Rogoff, op. cit.. pp. 23-4. 95 name than in status, this is not surprising. Further, as Rogoff found, occupational mobility is most likely to take 97 place between "adjacent" occupations. Another major factor in inter-generational occupational mobility is the changing proportion of occupations from one generation to the next. Thus, the increasing supply of vacant nonmanual occupations has not only allowed but has encouraged upward mobility. As Elbridge Sibley pointed out, . . . changes in the national economy between 1870 and 1930 produced a very marked upward shift of the center of occupational gravity. . . . Some 9,000,000 persons who were white-collar workers in 1930 would have been engaged in manual labor if the occupational distribution of 1870 had persisted.98 Compared to inter-generational mobility, intra- generational occupational mobility is less fluid. Rogoff is of the opinion that no more than 50 per cent of the population moves out of the occupation class in which it 99 starts. Lipset and Bendix have pointed out that there are 97 Ibid. For other studies of inter-generational occupational mobility see: Davidson and Anderson, op. cit.; Morris Ginsberg, Studies in Sociology (London: Methuen and Company, 1932), pp. 170-74; and Richard Centers, "Occupa tional Mobility of Urban Occupational Strata," American Sociological Review. XIII, 2 (April, 1948), pp. 197-203. 98 - Sibley, op. cit.. p. 323. See also; Gerhard E. Lenski, "Trends in Inter-generational Occupational Mobility in the United States," American Sociological Review, XXIII, 5 (October, 1958), pp. 514-23. 99 Rogoff, op. cit.. pp. 25-27. She summarizes earlier studies of intra-generational occupational mobility: Davidson and Anderson, op. cit.; F. V. Taussig and C. S. Joslyn, American Business Leaders (New York: The Macmillan Company, 1932); William H. Form and Delbert C. Miller, 96 many shifts from one occupation to another, particularly in the lower strata, but the shifts occur between occupations of similar status. They found that there was little cross ing of the manual-nonmanual line: . . . all those who work with their hands have spent 80 per cent of their working lives in manual occupa tions; all in nonmanual employment have spent 75 per cent of their careers in such p o s i t i o n s . 100 As Ely Chinoy points out: Two ladders of advancement seem to have emerged in industry. One, open to workers, is short, with few rungs, usually ending with foremanship. The other, open to those whose education and training enable them to begin as technicians or white collar workers, is longer, and may eventually lead to the top levels of industry. What this seems to imply is that education is quite often 102 the key to upward occupational mobility. "Occupational Career Patterns as a Sociological Instrument," American Journal of Sociology. IL, 4 (January, 1944), pp. 317-29; and Louis I. Dublin and Robert J. Vane, Jr., "Shifting of Occupations Among Wage Earners As Determined by Occupational History of Industrial Policyholders," Monthly Labor Review, XVIII, 4 (April, 1924), pp. 732-40. ^^Lipset and Bendix, op. cit., p. 165. For data see Appendix A, tables 123 and 124. ^^Ely Chinoy, Automobile Workers and the American Dream (New York: Doubleday and Company, Inc., 1955), p^ 67 102 See Centers, "Education and Occupational Mobility," op. cit.; C. Arnold Anderson tries to refute the Centers thesis that there is a relationship between occupa tional mobility and education but fails to make a good case. See C. Arnold Anderson, "A Skeptical Note on the Relation of Vertical Mobility to Education," American Journal of Sociology. LXVI, 6 (May, 1961), pp. 560-70. 97 Aspiration and motivation In addition to the well known factors of education and training which tend to keep the lower class individual from attaining a higher class status there may be some psychological factors, a unique lower class value system • which does not motivate toward the higher class status. In fact, this value system does seem to exist and it 103 places a low value on education and training. Herbert H. Hyman cited the results of a National Opinion Research Center poll which found that adult in different classes placed a differential emphasis on a college education as 104 essential to success. The higher the class, occupational rank, or education achieved the more emphasis adults placed 10 5 on the necessity of a college education. Similarly, youths from higher class families placed more emphasis on th© need for a college education than did youths from lower class families (see data, p. 74, this chapter, and 106 fn. 23). Hyman also found a reduced striving for success 103 Herbert H. Hyman, "The Value Systems of Different Classes: A Social Psychological Contribution to the Analysis of Stratification," Class, Status and Power, op. cit., pp. 426-42. Also, see: Bernard C. Rosen, "The Achievement Syndrome: A Psychological Dimension of Social Stratifica tion," American Sociological Review. XXI, 2 (April, 1956), pp. 203-211. 104u _ .. Hyman, op. cit. 105Ibid., p. 430. 106Ibid., p. 432. 98 and a greater awareness of lack of opportunity among the 107 lower class youth. Chinoy also reported that there was a general lack of interest, hope, and desire among automo bile workers, which he attributed to an awareness of limited opportunities, the uncertainty of criteria for advancement, 108 and the nature of advancement (often to an "out-group"). Style of life One way of "arriving," a way which is becoming more and more available, is to assume the style of life of the status one aspires to. Thus, as William Form and Gregory Stone pointed out, in seeking status the urbanite may fre- 109 quently rely on appearance rather than reputation. Such reliance on appearance is facilitated by the anonymity of urban life. With the distribution of income becoming more even in the United States the style of consumption would 110 seem to be of increasing importance for stratification. 107 Ibid.. p. 438. Hyman reports that there is some reason to believe that the lower class individual has some "deviant" occupational goals— undertaker, night club singer, county judge, or professional boxer— goals with low status but possible high financial rewards and status within the lower class. 1 08 Chinoy, op. cit.. p. 49. Barber, op. cit., p. 309, also found a lower level of aspiration among lower class respondents. 109 Form and Stone, op. cit., pp. 504-14. 110 Lipset and Bendix, op. cit., p. 273; the authors also found the same trends toward income equalization in Sweden, Germany, the Netherlands, and Great Britain. 99 Lipset and Bendix have suggested that within each income bracket it is the cultural background or the social aspirations that influence the way income is spent. The authors cite an unpublished survey of San Francisco longshoremen's style of life aspirations and political * 11 o ideology. Another study by Form and Stone found that the social status accorded an individual may depend on 113 one's dress. Such subjective criteria as status symbols and style of life certainly reduce the reliability of the standard objective status symbols (e.g., occupation and education). Impact on the individual Lipset and Bendix suggested that social mobility is, likely to disrupt primary group structures. E. E. Lemasters, in his study of the impact of mobility on the family unit, 114 found that most families were disrupted. As he said: 1111bid.. p. 275. 112 Ibid., p. 273; longshoremen who left the docks area were found to be more conservative politically than those who resided in the docks area (income held constant). Authors cite unpublished study by Joseph Aymes, a graduate psychology student at the University of California, Berkeley. 113 Villiam H. Form and Gregory P. Stone, The Social Significance of Clothing in Occupational Life, Technical Bulletin No. 247 (East Lansing: Michigan State College, 1955), cited in Lipset and Bendix, op. cit.. p. 273. 114 E. E. Lemasters, "Social Class Mobility and Family Integration," Marriage and Family Living, XVI, 3 (August, 1954), pp. 226-32. 100 . . . differential social mobility within the family breaks up the close ties between the members . . . they do not have a common sub-culture; they do not even speak the same language, in a sense; they do not inhabit the same community; . . . they live in different social worlds; they marry into different social class groups; they very often have quite different value systems. 11 j Those moving upward or downward are torn between the diverse cultures of their initial orientation and their later posi tions (and aspirations). And, as Vestoff, et al.t pointed out, there have been an increasing number of studies con cerned with the relationship between social mobility and different cultural patterns, value commitments, and per- 1 1 A sonality patterns. The Theory of Status Consistency Multiple status dimensions Not until recently has there been much thought about or investigation of the impact of socil mobility on the individual as well as society in terms of multiple status hierarchies. Aristotle, Marx, and Varner represent the majority of social philosophers and social scientists who have viewed the status structure as vertical and uni dimensional— as a single hierarchy upon which each individual, occupies a single position. Their critics, on the other hand, view the structure as a combination of 113Ibid.. p. 229. ^^Vestoff, et al. t op. cit. . pp. 376-77. 101 several parallel vertical hierarchies which are not perfectly correlated and upon which an individual may 117 occupy several different "levels” at the same time. Max Weber suggested that social stratification appears to 118 be three-fold: economic, social honor, and power. Kurt Mayer suggested that there are as many status hierarchies as there are distinguishable patterns of interpersonal 119 relations. Gerhard Lenski, examining the impact of these status inconsistencies (low status crystallization as it related to political attitudes), used four hierarchies: 1 20 income, education, occupation, and ethnic. And, as was earlier shown, Vestoff, et al., found that such indices are not interchangeable. As Mayer pointed out, many differences in authority and power which operate in contemporary American society are not directly and, or, visibly connected with economic difference. As a result, the status of an individual at any given time is not necessarily the equivalent of his economic 117 Gerhard E. Lenski, "Status Crystallization: A Non-Vertical Dimension of Social Status," American Sociolo gical Review. XIX, 4 (August, 1954), p. 405. Durkheim's concept of anomi can be said to have been based on an under standing of the multi-status dimensions of society. 118 Max Weber, "Class, Status, Party," From Max Weber: Essays in Sociology. Garth and Mills, eds.; op. cit., pp. 180-95. 11^Mayer, op. cit.. pp. 321-35. 1 20 Lenski, "Status Crystallization," op. cit., pp. 406-07. ----- 102 1 21 or class position. This is well illustrated by the rise in the standard of living of wide sectors of the working class which has minimized or eliminated the differences between working class and middle class income without equalizing prestige. The complexity of the dynamics involved are well illustrated by recent research of Leo F. Schnore, in which he pointed out that some bases of status are unchangeable as ascribed (sex, race, kinship), while other status indicators are changeable and can be achieved 1 2 2 (religion, education, place of residence). Cross pressures Elton F. Jackson suggested that the individual in status inconsistency suffers two major consequences: frus- 1 23 tration and uncertainty. The situation is frustrating in that the expectations of others are contradictory and cannot, therefore, be satisfied; and gives rise to uncer tainty in that the individual does not know what to expect 1 21 Mayer, op. cit.. pp. 322-23. For noneconomic power basis in other nations, see: Dwaine Marvick (ed.) Political Decision Makers (Glencoe; The Free Press, 1961). 1 22 Leo F. Schnore, "Social Mobility in Demographic Perspective," American Sociological Review, XXVI, 3 (June, 1961), p . 413. 1 23 Elton F. Jackson, "Status Consistency and Symptoms of Stress," American Sociological Review, XXVII, 4 (August, 1962), pp. 469-80. 103 1 24 from others. Bernard Berelson tells us that in the matter of political choice the individual under cross 1 25 pressures quite often withdraws. However, in the nature of status inconsistency, there is little opportunity to withdraw from ascribed status and great reluctance to with draw from achieved status. Where there is a possibility of choice, as Tamotsu Shibutani stated, the individual makes a choice which is essentially one between norms and perspectives. Status inconsistency behavior In a society which holds social mobility to be a central value and in which it is a fairly common occurrence, the experience should not be damaging to an individualTs personality. There are situations, however, when damage does result. For example, when there is inconsistency between ascribed status and achieved status, when the move covers a wide range, when there has been little opportunity to learn the norms of the status achieved; or when there is a basic antagonism between the norms of origin and the 1 24 Ibid., p. 470. 1 25 Bernard R. Berelson, Paul F. Lazarsfeld, and William N. McPhee, Voting (Chicago: University of Chicago Press, 1954), pp. 19-20. 1 26 Tamotsu Shibutani, "Reference Groups and Social Control," Human Behavior and Social Processes. Arnold Rose (ed) (Boston: Houghton Mifflin Company, 1962), p. 140. 104 1 27 norms of achievement. C. Wright Mills, in his classic, White Collar, discussed what he calls the "status panic": . . . given occupational levels, however caught in status ambivalence, do enjoy typical levels of pres tige. . . . however, . . . the enjoyment of prestige is often disturbed and uneasy, . . . the basis of prestige, the expression of prestige claims, and the ways these claims are honored, are now subject to great strain, a strain which often puts men and women in a virtual status p a n i c .128 The intense status concern of politics is shared by two - different types of individuals who "arrive'1 at this concern from opposite directions. One type concerned with status is the downwardly mobile old-family, Anglo-Saxon, Protestant; the other type is the immigrant ethnic who is 1 29 upwardly mobile. Peter Blau found that the upwardly mobile and the downwardly mobile were more likely to have out-group prejudices or to be preoccupied with their health than either those who were nonmobile lower class or 1 30 nonmobile upper class. And, as Robert Lane pointed out, 1 27 Elizabeth Douvan and Joseph Adelson, "The Psycho dynamics of Social Mobility in Adolescent Boys," The Journal of Abnormal and Social Psychology. LVI, 1 (January^ 1958), p. 32. 128C. Wright Mills, White Collar (New York: Oxford University Press, 1956), p. 240. 1 29 . Richard Hofstadter, "The .Pseudo-Conservative Revolt," The New American Right. Daniel Bell, ed. (New York: "Criterion Books, 1955), p"I 44, As Mayer points out, the confusion of status and economic class tripped up Marx and his followers who today fail to recognize the middle class status motivation, op. cit.» p. 323. ^■^Peter Blau, "Social Mobility and Interpersonal Relations," American Sociological Review. XXI, 3 (June, 1956), pp. 290-95. 105 it is not from the lower classes that expressions of dis satisfaction arise, but from those who are caught between 1 3i their achievements and their aspirations. Jackson's study of the relationship between status inconsistency and behavior found significant differences between individuals with high consistency and low con- 1 32 sistency. Further, the patterns of response were quite different according to the roots of status inconsistency. Individuals with a low achieved status, or a high ascribed status compared to their achieved status, responded somatically. On the other hand, respondents with high achieved status, or an ascribed status inferior to their 1 33 achieved status, responded politically. Another study, by Lenski, found a significant dif ference between low and high "crystallizers" in their 1 34 political preference, stand on issues, and "liberalism.'.' 1 31 Robert Lane, Political Life (Glencoe: The Free Press, 1959), pp. 130 and 156. 1 32 Jackson, op. cit.. 133Ibid., p. 479. 1 34 Lenski, "Status Crystallization," op. cit. Low crystallizers are those with status inconsistencies and high crystallizers are those without status inconsistencies. Lenski used a ranking method for each of four status measurements (income, occupation, education, and ethnic) and calculated each individual's deviation from the mean.. Each individual's crystallization was calculated by: C = 100 - J H d-,2 + d22 + d 32 + d4 2 ) * ‘ Use of squared deviation emphasizes the larger deviations 106 Low crystallizers, those with greater status inconsisten cies, supported Democratic candidates to a greater extent 135 than did high status crystallizers. Lenski* s data also indicated that for low crystallizers there are differences in partisan affiliation depending on the nature of the status inconsistencies. Those with high achieved status and low ascribed status were much more likely to be "strongly Democratic" (over 50 per cent) than those with high ascribed status and low achieved status (about 30 1 36 per cent). Lenski also found that those with high achieved status and low ascribed status were more likely to be liberals than those with low achieved and high 1 37 ascribed status. If suburbia is made up of the socially mobile, as may be the case, the status inconsistency (or status crystallization) findings warn us that any broad assumption of a direct correlation between upward social mobility and voting Republican is indeed suspect. compared to the smaller. Liberalism, to Lenski, is support of government health insurance, price controls, and a general extension of governmental controls. In light of Herbert McClosky, Angus Campbell, and others, Lenskifs liberalism is rather narrow in scope. 1 35 Ibid.« p. 410. Lenski uses the 1950 Michigan gubernatorial and the 1948 and 1952 presidential elections. 136Ibid., p. 411. 137Ibid. 107 The Theory of the Individual Within the Group While the concept of group life and the group as a major influence in individual life can be ^traced back to Aristotle, it can be fairly said that not until Arthur F. Bentlyls The Process of Government was there a systematic 1 38 attempt to come to grips with the concept. Since then, much has been done, not only in political science but in sociology, psychology, social psychology, economics, and others of the social sciences, with David B. Truman 1 39 presiding at the revival for political scientists. Group defined As generally defined by Bently, a group means: . . . a certain portion of the men of society, taken, however, not as a physical mass cut off from other 1 38 Arthur F. Bently, The Process of Government (Chicago: University of Chicago Press, 1908). Aristotle*s statement that man is a social animal certainly implies the relationship of man to the group. John Dewey*s statement that: "Associated activity needs no explanation; things are made that way. . . ." disposes of the problem of "early citationi" See: John Dewey, The Public and Its Problems (New York: Henry Holt Company, Inc., 1927). James Madisonrs classic comment on the mischiefs of faction are an example of a more narrow consideration of the relation of groups to government. 1 39 David B. Truman, The Governmental Process (New York: Alfred A. Knopf, 1951). This revival had occurred earlier among some of the other disciplines, e.g., Muzafer Sherif*s The Psychology of Social Norms (New York: Harper and Brothers, 1936) and Linton*s The Cultural Background of Personality, op. cit. To be sure, somewhat in line with Madison*s concern we find Pendleton Herringfs Group Repre sentation Before Congress (Baltimore: The John Hopkins Press, 1929). 108 masses of men, but as mass activity, which does not preclude the men who participated in it from partici pating likewise in many other group activities.140 And, There is no group without its interest. An interest, as the term will be used in this work, is the equivalent of a group. . . . The group and the interest are not separate.141 Thus, Bently defined group in terms of activity and interest. To these minimum criteria, Truman added communi- 142 cation or interaction. The primary and secondary groups Beyond the minimum, however, the definition of group assumes different dimensions depending on the type of group. One of the basic classificiations of groups has been that of the primary and secondary group. In American sociology, Charles H. Cooley is credited with coining the term "primary 143 group." For Cooley, the small informal face-to-face group was "primary" in the sense that the moral standards which constitute social reality are formed in such groups. A contemporary writer, Robert Mclver, defined the primary 1 40 Bently, op. cit., p. 211. 1 41 T, . , Ibid. 1 42 Truman, op. cit.. pp. 23-26, 143 Charles H. Cooley, Social Organization (New York: Scribner*s Sons, 1909); also see F. R. Clow, "Cooley*s Doctrine of Primary Groups," American Journal of Sociology. XXV, 3 (November, 1919), passim. 109 group as the face-to-face nucleus which exists in all social organizations for the individual— an individual's own unit cell within the greater organization— the most obvious 144 example being the family. For Edward A. Shills, the primary group is characterized by a high degree of soli darity, informality in its code of rules regulating the 1 45 numbers, and autonomy in the creation of these rules. . As a research tool, the primary group was "rediscovered" for the simple reason that other models of social behavior did not explain many observed phenomena. Research models centered on concepts of mass production, mass communication, and mass society on one hand; or on the 146 "independent" citizen, on the other hand, failed. Thus, recent studies of industrial organization, the United States Army in World War II, and the community employed, or found 1 47 to be important, the concept of the primary group. As 144 Robert Mclver, Society (New York: Farrar and Rinehart, Inc., 1937), p. 236. ^ ^Edward A. Shills, "The Study of the Primary Group," The Policy Sciences, ed. Daniel Lerner and Harold D. Lasswell (Stanford: Stanford University Press, 1951), p. 44. 146 For an extensive discussion of these models see: Elihu Katz and Paul F. Lazarsfeld, Personal Influence (Glencoe: The Free Press, 1955). 1 47 F. J. Roethlisberger and W. J. Dickson, Manage ment and the Worker (Cambridge: Harvard University Press, 1939); Samuel A. Stouffer, Edward A. Suchman, Leland DeVin- ney, Shirley A. Star, and Robert M. Williams, Jr., The American Soldier: Adjustment During Army Life, Vol. I, (Princeton: Princeton University Press,1949); and W. Lloyd Warner's Yankee City series (New Haven: Yale University Press, 1945-1959). 110 Truman pointed out, the social psychologist and the cultural anthropologist have done much to explain the manner and extent of group influence upon individual 148 behavior. But, while the role of the group becomes better understood, there are those who contend that its actual significance is being reduced by the atomization of society. The individual's family, religious connections, childhood friends are all reduced in significance due to social and residential mobility, and the industrialization 1 49 of society. As a result, according to Morris Janowitz, the secondary groups— formal large scale structures, commu nity organizations, and business bureaucracies— have acquired a significance today which they may have lacked before.^ All of this is most prevalent and significant in the urban areas, where the secondary group may well 1 51 replace the primary group. Whether one considers primary or secondary group, the relationship is still one of formal or informal achieved and recognized membership. 148 Truman, op. cit.. p. 17. 1 49 Lane, op. cit.« p. 110; Eric Fromm, Escape from Freedom (New York: Farrar and Rinehart, Inc., 1941); and Morris Janowitz, "Some Consequences of Social Mobility in the United States," Transactions of the Third World Cong ress of Sociology (London: International Sociological Association, 1956), pp. 191-201. ^ "^Janowitz, op. cit.. p. 193. 151 Louis Wirth, "Urbanism as a Way of Life," Reader in Urban Sociology, P. K. Hatt and A. J. Reiss, Jr., eds. (Glencoe: The F r e e Press, 1957), pp. 32-49. 111 Reference group theory Men frequently orient themselves to groups other than those they belong to— these nonmembership groups are commonly called reference groups. Further, as Tamotsu Shibutani pointed out, the concept of reference group includes groups which may be in fact imaginary— the vital dynamic is that real or imaginary, they are used as a frame of reference by the individual in the organization of his 1 52 perceptual field, and as a source of values and norms. The reference group cannot be equated with the primary- secondary group concept since the reference group may be imaginary; its controls on the individual stem from an entirely different relationship between the "group" and the individual. Since the individual may not be a member of the group, group members cannot interact with him but the individual acts as though they did. As S. M. Eisenstadt pointed out: . . . most of the choices of reference groups seem to be made in terms of status aspirations of the individual and his evaluation of the status-conferral possibilities of different groups . . .153 Vhat this comes to is that the individual assumes a 152 Tamotsu Shibutani, Society and Personality (Englewood Cliffs: Prentice Hall,’ Inc., 1961), p. 257; and see his article, "Reference Groups and Social Controls," op. cit.. p. 132. 1 53 S. M. Eisenstadt, "Reference Group Behavior and Social Integration: An Explorative Study," American Sociol ogical Review* XIX, 2 (April, 1954), p. 183. 112 psychological membership with the group to which he aspires. There is, of course, a reverse situation— the negative references group. Former group members often convert their earlier group membership into a negative referent during upward social mobility. Thus, the individual in the process of moving from one group to another may reject the group of origin's values and assume the values of the group of aspiration without being a member of either. While there is a limit to the number of primary and secondary groups to which one may belong, there is, in principle, an almost unlimited number of reference groups. Methodologically, in terms of analysis of behavior, the reference group offers major operational problems. While the researcher may ascertain an individual's primary and secondary group memberships through ascribed and achieved attributes, there are no such "objective" criteria for determining an individual's reference groups (whether they 1 54 be primary, secondary, real or imaginary). Patterns of group membership While the behavioral and attitudinal linkages 1 54 An excellent example of using primary and secon dary group affiliations to "understand" individual behavior is found in the IPP Index of Berelson's Voting, op. cit., pp. 125—126. For an excellent criticism of the IPP Index, see Peter H. Rossi's "Four Landmarks in Voting Research," American Voting Behavior, ed. Eugen Burdick and Arthur Brod- beck (Glencoe: The Free Press, 1959), pp. 5-54. 113 between group memberships and the individual offer many operational problems there has been a considerable amount of research done on the patterns of primary and secondary group membership and activity. These studies have been largely concerned with the impact of urbanization upon primary and secondary group memberships. Scott Greer and Ella Kube found that as urbanism increases, neighboring declines along with social participa- 1 55 tion, membership, and attendance in formal organizations. Greer and Kube also concluded that as urbanization increases kin and childhood friends become more significant in inter- 1 56 personal relations. In another study of a community’s friendship patterns, Leon Festinger, et al., found that differences in distance as little as twenty or thirty feet 1 57 played a major part in determining friendships. There are, traditionally, two major views of urban personal relations. The first emphasizes the impersonal 155 Scott Greer and Ella Kube, "Urbanism and Social Structure— A Los Angeles Study," Community Structure and Analysis. ed. Marvin B. Sussman (New York: Crowell, 1955), p. 109. Ibid.. pp. 109-110. This seems to conflict with the earlier (1954) findings of E. E. Lemasters (supra, P. 100). 1 57 Leon Festinger, Stanley Schachter, and Kurt Back, Social Pressures in Informal Groups (New York: Harper and Brothers, 1950), p"! 34. This study was of a homogeneous student population, the members of which had had no previous social contact with each other. 114 nature of the urban community and the decline of the kin ship group with the consequent increased importance of the secondary groups. The second theory accords informal friendship groups greater importance than the formal secondary groups. As several studies have demonstrated, the social and economic characteristics of the neighborhood 1 58 greatly influence these patterns. One survey of Detroit, Michigan, found that 63 per cent of the respondents were members of formal groups, but 1 59 that few belonged to more than one group. Only one-fifth of those surveyed were very active, held office, attended meetings, etc., within their groups. Morris Axelrod found that participation and membership rates varied directly 1 60 with education, family income, and ranking of occupation. Further, he found that status once achieved carried with it the obligation of participation in various community groups. In examining respondents' informal group relations, Axelrod found that kin were most significant followed by friends, 1 5 8 Wendell Bell and Marion D. Boat, "Urban Neigh borhoods and Informal Social Relations," American Journal of Sociology, LXII, 4 (January, 1957), pp. ^91-98. i 5 9 Morris Axelrod, "Urban Structure and Social Participation," American Sociological Review. XXI, 1 (February, 1956), pp. 14-15. Axelrod excluded church affiliation. Respondents were selected on an area proba bility sample, N =749. 160Ibid.. p. 15. 161 neighbors, and co-workers. Floyd Dotson found, in a study of fifty New Haven families (skilled and semi skilled), that the majority of working class people do not participate in formally organized volunteer organizations < j £ O and that kinship was the most significant relationship. Generally, those with higher socioeconomic status are more likely to be members of formal organizations and to actively participate in formal organizations and informal relationships while those with lower status are more likely to be limited to kin relationships. Social reality and norms Earl Latham began his work, in essence, with the statement that "the chief social values cherished by indiv- 1 /* O iduals in modern society are realized through groups." John Dewey rejected the concept of the abstract individual, asserting that the individual has meaning only in his 1 64 relations with others. Gestalt would go farther, 1 61 Axelrod found that those with the highest social status slightly reversed the order, having more frequent contacts with friends than kin, Ibid., p. 17. 162 Floyd Dotson, "Patterns of Voluntary Association Among Urban Working-Class Families," American Sociological Review. XVI, 5 (October, 1951), p. 693. 163 Earl Latham, "The Group Basis of Politics: Notes for a Theory," American Political Science Review. XLVI, 2 (June, 1952), pp. 376-97. 164 John Dewey, Human Nature and Conduct: An Intro duction to Social Psychology (New York: Henry Holt and Company, 1922), p. To! 116 arguing that the basic forms of knowing are collective. Dr. Karen Horney, rejecting the Freudian assumption of the singular person, asserts that what were at one time regarded as innate personality traits are, in fact, introduced traits 1 f\ s originating within the culture. As Robert Presthus put it: From bifth onward the individual is subjected to group norms that over a period of time mold his per sonality. Personality is defined as a consistent way of reacting or accommodating to interpersonal situa tions. The reduction of anxiety by compliance with the perceived wishes of authoritative persons . . . is a * critical mechanism in this process. Interpersonal psychiatry asserts that most behavior is the result of the individual’s search for relief from tension by con forming to authority. . . . The basis of tensions are (largely) socially d e r i v e d . 1 6 6 Truman put it more succinctly: . . . every individual, from infancy onward, tries to make himself an accepted participant in a group, or, more properly, a set of groups, that makes up his social environment. . . . Most individuals in any cultural set ting find it intolerably painful not to be accepted by the groups in which they move or in which they hope to move.167 Daniel Katz suggested three specific functions of groups: (1) identification through the incorporation of ^^See Karen Horney, The Neurotic Personality of Our Time (New York: Norton, 1937). 1 66 Robert Presthus, The Organizational Society (New York: Alfred A. Knopf, 1962), pp. 96-97. 1 67 Truman, op. cit.. pp. 18-19. For an example of conformity to group norms, see: Orvis Collins, Melville Dalton, and Donald Wray, "Restriction of Output and Social Cleavage in Industry," Applied Anthropology. V, 3 (Summer, 1946), passim. 117 group values into one's own value system, (2) reference in an ambiguous situation, and (3) identity as part of a 168 group. Katz and Lazarsfeld suggested a slightly dif ferent approach. They see the primary group as providing: (1) identity through conformity, (2) social reality in 1 69 giving meaning to situation, and (3) social norms. One may combine much of the meaning in both these definitions by saying that groups define social reality and establish social norms, and that acceptance of these realities and norms is impelled by the rewards of acceptance into the group. As Katz and Lazarsfeld pointed out: Interpersonal relationships seem to be "anchorage" points for individual opinions, attitudes, habits and values. That is, interacting individuals seem collec tively and continuously to generate and to maintain common ideas and behavior patterns which they are reluctant to surrender or to modify unilaterally.170 David Riesman's happy phrase, "other-directedness," summed up much of these relationships but unfortunately does not tell us "what other"? Herbert McCloskey and Harold Dahlgren assert: The belief that people who associate together come to think alike is now so thoroughly buttressed by 1 68 Daniel Katz, "The Functional Approach to the Study of Attitudes," Public Opinion Quarterly. XXIV, 2 (Summer, 1960), pp. 163-204. 1 69 Katz and Lazarsfeld, op. cit., pp. 50-59. 170Ibid., p. 44. 118 research and daily observation that it has become a commonplace.171 There is more than a nagging doubt that McClosky and Dahl- gren have confused cause with effect. There is some reason to suggest that people associate because they have values in common. Thus, Joseph A. Precker found that in college, students and faculty tended to choose peers and near 172 authority figures whose values were similar to their own. The relationship is probably two-way and depends, in addi tion, on the nature of the "reality" in question. Social reality differs from physical reality not only in terms of "hard" fact but in terms of testing the reality. Vhile the weight of a statue may be determined by a fairly objective "reality check," its value is determined by the more sub jective social reality. Since social opinion and attitudes often cannot be checked against physical reality, their value depends upon and is defined by one's reference groups. There is a considerable body of research data to substantiate the relationship between reality and group 173 consensus. The classic experiment was conducted by 171 Herbert McClosky and Harold D. Dahlgren, "Primary Group Influence on Party Loyalty," American Political Science Review, LIII, 3 (September, 1959), pi 757. 172 Joseph A. Precker, "Similarity of Valuings As a Factor in Selection of Peers and Near-Authority Figures," Journal of Abnormal and Social Psychology. XLII, 2 (April, 1952), passim. 173 See Kurt Lewin and Paul Grabbe, "Conduct, Know ledge, and Acceptance of New Values," The Journal of Social Issues, I, 3 (August, 1945), passim. 119 174 Muzafer Sherif using the autokinetic effect. Sherif found that a single individual placed in a new stimulus situation— unstructured and unstable— establishes a range 175 and norm. When the individual who has established a range and a norm is put into a group with other individuals who have also undergone the same stimulus experience and who have also established their own individual ranges and norms, the ranges and norms of each individual tend to 1 76 converge toward a group range and norm. The convergence of ranges and norms is not so close as when the individuals are first exposed to the autokinetic effect as a group. Further, when a member of the group, after the group has established a range and a norm, faces the effect alone he applies the group's range and norm. Sherifs experiments are only one of a number of similar examinations of social reality definitions by groups for individual members. Solomon Asch found that stated judgments of a group sometimes could alter an individual’s judgment as to the 174 The autokinetic effect consists of placing an individual in a dark room in which there is a single point of light. While the light is in fact stable it appears to the individual to move; for a full description of the auto- kinetic effect see: H. F. Adams, ’ ’ Autokinetic Sensations," Psychological Monograph No. 59 (1912), pp. 32-44. 175 After exposure to the "moving" point of light the individual establishes the range of its movement and a norm. 1^Muzafer Sherif, "Group Influences Upon the Forma tion of Norms and Attitudes," Readings in Social Psychology, ed. Theodore M. Newcomb and Eugene L. Hartly (New York: Henry Holt and Company, 1947), pp. 77-90. 120 177 length of a line. In this situation, reality is rela tively "hard" compared to the autokinetic effect. In fact, Asch found that the preponderance of individual judgments as to the length of a line were not altered in spite of 178 group pressure to modify those judgments. Asch's exper iment suggests that some of the "conformity" found in Sherif’s work was more of an acceptance of "information"— that individuals in the Sherif experiment were not too sure of their range and norms and were willing to accept the group range and norms as more accurate. As Sherif says: . . . other norms, very different from ours, are possi ble even in the most fundamental psychological cate gories. We have our existence in a cultural atmosphere as well as in a physiclal atmosphere. 179 Sherif cites several studies of a different culture in 1 80 which colors are classified differently. Sherif's general theme is that in the course of an individual’s life he internalizes his group's norms and that subsequently these norms determine his actions in concrete situations. 1 77 Solomon E. Asch, "Effects of Group Pressure Upon the Modification and Distortion of Judgments," Group Dynamics. ed. Dorwin Cartwright and Alvin Zender (Evanston: Row, Peterson and Company, 1953), pp. 151-62. 178 Ibid., p. 153. Asch conspired with the group to announce erroneous judgments— the individual being tested did not know of the conspiracy. 1 79 Sherif, The Psychology of Social Norms, op. cit., p. 9. 1^°Ibid., pp. 14-15. 121 Conformity and changes Truman pointed out that category groups are signif icant for behavior analysis- only as they exhibit some uniformity in behavior and that such uniformities depend not only on shared characteristics, but also on interaction 1 81 between members. M. Brewster Smith, et al.. wrote that: Demographic description of a person's income group or religious affiliation is not the same thing as a description of the role and status requirements that such membership entails.182 In order to conclude that observed uniformity in behavior among a number of individuals is the result of the operation of a group standard or social norm one must be able to show the existence of a psychological group. As Leon Festinger, et al., pointed out, a group's influence upon an individual depends upon the attraction of that group for the individual 1 83 and some form of communication (very broadly defined). "Social control" does not depend so much on deliberate influence or coercion but on the fact that each individual generally takes into account the group's expectations or the 1 84 expectations which he imputes to the group. 1 81 Truman, op. cit., pp. 23-24. 1 82 Smith, Bruner, and Vhite, op. cit., p. 14. 1 83 Festinger, Schachter, and Back, op. cit., p. 171. 1 84 Shibutani uses "impute" which seems a happy choice as it comprehends both reference group dynamics and the individual's perceptual defense mechanisms. Shibutani, "Reference Groups and Social Control," op. cit., p. 129. 122 However, the dynamics of social control differ, depending upon the nature of the reference group. If the reference group is also a membership group, the expressions of appropriate opinions become a required part of continued membership. Expressions of appropriate opinions are a part of the communication or interaction process of membership groups. If the reference group is not a membership group, it may be idealized, distant, or perhaps fictional. In the latter case, opinions need only be held, not expressed, to serve as a link. Holding an appropriate opinion symbolizes for the individual his solidarity or "affiliation" with a 185 nonmembership reference group. Opinions expressed may often contradict the observed group memberships of a given individual and suggests that the observed group membership does not always function as a reference. Analysis of poli tical behavior must take into consideration not only the gestalt or interpersonal opinion-attitude processes, but the reverse of these processes. From the individual’s point of view, the group exists to satisfy personal values or felt needs; projection may be as significant as identification. Nicholas T. Fouriezos, et al.. listed the various relationships which may exist between the individual and the group: 185 See: Smith, et al.. op. cit., p. 10. The holding of opinions appropriate to a reference group offers a sig nificant insight into the dynamics of subjective class affiliation and attitudes. 123 1) Dependency; a reliance on authority or a general need for support by conforming to general group attitudes, 2) Status; the achievement of title, or identifica tion with the group, 3) Dominance: the need to demonstrate intellectual superiority or dominance in social situa tions , 4) Aggression; against the authority of the leaders and, or, the group or a negative attitude toward group policy, and 5) Catharsis; participation in the group allows the person to t a l k . 1 8 6 The individual's relations with the group depend upon the degree to which one or more of these personal needs are ful filled by the group. The more significant the group in the individual's life, the more compelling will be the forces to conform to the appropriate norms or social reality as defined by the group and the relationship. Leon Festinger and Harold Kelley found that it is not information about the world that counts, but what other 1 87 people think about the world. Vhile a person can distort facts to conform to his attitudes (perceptual defense), it is difficult for him to distort his perception or others' attitudes without destroying his communications with them. Thus, the need to accurately perceive others' attitudes— 186 Nicholas Fouriezos, Max L. Hutt, and Harold Guetzkow, "Self—oriented Needs in Discussion Groups," Group Dynamics. op. cit.. pp. 355-57. 1 8 7 Leon Festinger and Harold H. Kelley, Changing Attitudes Through Social Contact (Ann Arbor: Research Center for Group Dynamics, 1951), pi 97 124 social reality— is rooted in the need to maintain communica tion. The need to accurately perceive "hard reality" often 1 88 enjoys no such sanction. Group pressures to conform to group norms have been found to modify attitudes when they change from privately held opinions to publicly expressed opinions. Raymond L. Gordon found that respondents' privately expressed opinions about a controversial subject (Russia) were modified towards the group's opinions when respondent publicly expressed his 1 89 opinions. Further, each respondent was able to estimate whether the group's opinion was more or less pro-Russian than his own. Communications within the group appear to harmonize with the group's values— the group exercises a form of defense— a selective process— which exposes individ uals within the group to opinions which reinforce the 1 90 group's norms. This "value homophily" (to use Merton's term) is hard to measure in terms of group interaction as cause or effect. As Precker found, people of like opinions and values seek each other out. In addition, the group once established would offer little attraction to those who did not share its values or opinions. Leon Festinger, "Informal Social Communication," op. cit.. pp. 199-200. 1 89 Raymond L. Gordon, "Interaction Between Attitudes and the Definition of the Situation in the Expression of Opinion," Group Dynamics, op. cit.. pp. 165-69. i go Katz and Lazarsfeld, op. cit.. p. 96. 125 An individual’s perceptual defense has been shown to protect him from perceived conflict with his own gfoup. Mary Monk and Theodore Newcomb found that a person's polit ical preference distorted his perception of how others would 1 91 vote. Working class respondents who would vote for Eisenhower reported that more than half of their fellow workers would also. Working class respondents would would vote for Stevenson reported that more than half of their 1 92 fellow workers would vote for Stevenson. Conversely, middle class respondents, whether they were Stevenson or Eisenhower voters, accurately perceived how the working 1 93 class would vote. But, when asked how sales workers would vote, the middle class voters tended to attribute 1 94 their own political preference to the group. Difficulty in perception may be partially due to ambiguity. The working class or sales worker's vote is a 195 fairly ambiguous phenomenon. In addition, it is an 191 Mary Monk and Theodore M. Newcomb, "Perceived Consensus Within and Among Occupational Classes," American Sociological Review. XXI, 1 (February, 1956), passim. 1 92 Ibid., p. 73. In fact, in the area surveyed, about one-third of the factory workers voted for Eisenhower. 1 93 This may not be an illustration of perceptual accuracy but of a belief common to the middle class. 1 94 y Ibid., p. 74. 1 95 Monk and Newcomb asked respondents how their friends would vote— since it is not known how their friends did, in fact, vote, there is no check on the accuracy of 126 unjustified reification. This is not to say, obviously, that opinions do not change when in conflict with group norms. In fact, even "hard fact" has been shown not to 196 survive group norms. How opinions are changed offers a substantial clue as to the dynamics of conformity. Phillip Moneypenny found that groups whose members were in conflicting positions— that is, composed of individuals who were members of groups 1 97 in conflict— tended to be unstable. Cross-pressures are thus exerted on groups just as on individuals and, just as individuals tend to be more unstable under cross-pressure, so do groups. But, as Shibutani pointed out, overlapping group affiliations seldom lead to difficulty since reference 1 98 groups, for most people, are mutually sustaining. Cross pressures, when they do operate, appear to affect a rela tively small number of people and for a short time only. respondents' perceptions. Another study of voters' percep tions of group voting patterns may be found in Angus Camp bell, Gerald Gurin, and Warren Miller, The Voter Decides (Evanston; Row, Peterson and Company, 1954), pp. 211-14. 1 9 6 Albert H. Hastorf and Hadley Cantril, "They Saw a Game," Journal of Abnormal and Social Psychology, LIX (1954), and reprinted in Outside Readings in Psychology, ed. Eugene L. Hartley and Ruth E. Hartley (New York; Thomas Y. Crowell Company, 1957), pp. 94-103. 1 97 Phillip Moneypenny, "Political Science and the Study of Groups: Note to Guide a Research Project," Western Political Quarterly, VII, 2 (June, 1954), p. 198. a 98 Shibutani, Society and Personality, op. cit.« p. 575. 127 Individuals do not change reference groups readily. Leonard I. Perlin has taken the position that exposure to different ideas and beliefs does not account solely for attitude change— that attitude change depends in large part on whether or not the "different attitude" is held by 1 99 an attractive group or individual. If, as has been previously posited, an individual's opinions and attitudes are created and maintained in association with others, then changes in those opinions and attitudes can not be purely an individual matter. The operational problem of ascertaining an indi vidual's reference groups for a particular opinion or attitude is a difficult one. It is, to some extent, an inherently circular relationship. The fundamental nature of opinion suggests, moreover, that cross-pressures are avoided— so that the individual through time tends to increase the value homophily between his reference groups.The problems inherent in this circular relation ship are pointed up in a recent study by Bernard C. Rosen 201 of parent-peer group cross-pressures. After examining 1 99 Leonard I. Perlin, "Shifting Group Attachments and Attitudes Toward Negroes," Social Forces, XXXIII, 1 (October, 1954), p. 50. ^^See Paul Lazarsfeld and R. K. Merton's "Friend ship and Social Process," in Morroe Berger, T. Abel, and C. H. Page (eds.); Freedom and Control in Modern Society (New York: D. Van Nostrand and Company, 1954), pp. 18-66. 201 Bernard C. Rosen, "Conflicting Group Membership: 128 the patterns of friendships between Jewish high school students, their observance of kosher food laws, and their parents’ observance of .kosher food laws, Rosen concluded that: In cases where the attitudes of the two membership groups are conflicting (parents and peers), the group with which the adolescents agree tends to be their reference group. . . .202 For Rosen, reference groups are identified by attitude agreement. With such a methodology, it is not surprising that the researcher finds individuals in agreement with their reference groups. Group Political Behavior John Dewey once pointed out that every governmental policy has its own public. More than forty years ago, Charles E. Merriam reminded readers that both Bodin (six teenth century) and Gierke (nineteenth century) had called attention to the factor of the social group in the political 203 process. And, while there have been many words of caution against imputing a one-to-one relationship between groups and "obvious" issue orientations, pragmatic politics A Study of Parent-Peer Group Cross—Pressures," American Sociological Review, XX, 2 (April, 1955), pp. 155—61. 202Ibid., p. 160. 203 Charles E. Merriam, "Progress Report of the Com mittee on Political Research: Recent Advances in Political Methods," American Political Science Review, VII, 2 (May, 1923), p. 279. 129 204 is often practiced as if there were. As Moses Rischin » has said, American electioneering has always given special attention to the ethnic and religious interests of the voters. History demonstrates that Americans bring to the polls their special backgrounds and pull down the levers congenial to their national origins and religious ties.205 Vhether or not some group, for example Catholics, votes Democratic because of common and unique religious values, or rather due to the dynamics of their early entry 206 into American politics is a debatable point. But, as 204 Phillip Moneypenny: "The imputable groups, based upon some theory of stable political identification such as income, neighborhood, religion, national origin, and occupa tion, are not likely to be the equivalents of the groups which are demonstrably active on a given issue to which these indices seem relevant." Op. cit.« p. 188. 205 Moses Rischin, Our Own Kind (Santa Barbara: Fund for the Republic, 1960), p. 3. 206 According to John Fenton, "... the Catholic Church's influence on the political attitudes of Catholics is pervasive and profound. Ranging from birth control where the Church's influence was direct and most apparent [to] . . . right—to—work and civil rights for Negroes where influence was subtle and derived from certain basic values of the Church, the Catholic Church in each case evidently had a greater impact than the Protestant churches on the thinking and thus the political behavior of its members." John H. Fenton, The Catholic Vote (New Orleans: The Hauser Press, 1960), p. '56. Angus Campbell and Homer Cooper, on the other hand, attribute Catholic support of the Democratic Party to political inheritance. Early Catholics settled in the big cities and were captured by the local Democratic organizations. These early Catholics passed on to their children their religious faith and their political affilia tion. See, Angus Campbell and Homer C. Cooper, Group Dif ferences in Attitudes (Ann Arbor: University of Michigan Press, 1956), p"I 36. Other recent studies of religion and political behavior have come to the conclusion that there is some 130 Rischin pointed out, Jews in New York voted for Robert Wagner (a Catholic Democrat) rather than for Jacob Javits (a Jewish Republican); while Negroes in Philadelphia voted for a white Democrat rather than a Negro Republican; and in Washington and Ohio voters have elected a Rosellini and a 207 Lausche to the Governor's mansions. The political party also functions as a group, as Rischin said: The "Jewish Vote" stuck to Wagner, as expected, though a number of Jewish Democrats and a handful of Liberals crossed party lines to vote for Javits. More must have seriously considered defecting but drew back in "ter minal terror" before pulling the Republican l e v e r . 208 It is clear that partisan identification, as such, has a tremendous impact upon political behavior— political party identification is probably the single most effective relationship between the two. See: Benton Johnson, "Ascetic Protestantism and Political Preference," The Public Opinion Quarterly. XXVI, 1 (Spring, 1962), pp. 35-46; and Wesley Allinsmith and Beverly Allinsmith, "Religious Affiliation and Politico-Economic Attitude: A Study of Eight Major U.S. Religious Groups," The Public Opinion Quarterly. XII, 3 (Fall, 1948), pp. 377-89. Both Johnson and the Allinsmiths found broad differences between various Protestant denomina tions in political and economic attitudes and behavior. 207 Rischin, op. cit.. p. 33. An interesting but in conclusive study of political motivations by Joan Moore sug gests that only the unusually deprived or greatly advantaged have strong interest group orientations. She found that Negroes but not whites were politically oriented by race. On this basis she reaches the dubious conclusion that whites (holding occupational types constant) were largely free of interest group motivations. Joan W. Moore, "Social Depri vation and Advantage as Sources of Political Values," West ern Political Quarterly. XXV, 2 (June, 1962), pp. 217-26. 208 Rischin, op. cit.. p. 29. 1 31 "reference group" in politics— and the following discussion O C lQ is presented with that as a basic, if unexplored, premise. Primary and secondary reference groups Sidney Verba has suggested that: Primary groups of all sorts mediate political relation ship at strategic points in the political process. They are the locus of most political decision-making, they are important transmission points in political communi cations, and they exercise a major influence on the political beliefs and attitudes of their m e m b e r s . 2 1 0 A major theme in the individual-reference group relationship is harmony (or "value homophily"), in Henry Riecken's words: . . . there is a general tendency for friendships to develop among people of similar values, tastes, and opinions and also a tendency for individuals who have formed a friendship to adjust differences in values . . . in the direction of greater agreement.211 209 Belknap and Campbell found that differences in attitudes on foreign policy issues between Democrats and Republicans persisted even after holding religion, urban-rural, income, and levels of information constant. This suggests that parties are, per se, reference groups for political attitudes and behavior. George Belknap and Angus Campbell, "Political Party Identification and Attitudes Toward Foreign Policy." The Public Opinion Quarterly. XXV, 4 (Winter, 1951), pp. 601-23. However, aside from political parties, "the most influential associations in the political sphere do not appear to be those which have for their specific objec tive the influencing of the electorate." Howard E. Freeman and Morris Showel, "Differential Political In fluence of Voluntary Associations," The Public Opinion Quarterly. XV, 4 (Winter, 1951), pp. 713-14. 210 Sidney Verba, Small Groups and Political Behavior (Princeton: Princeton University Press, 1961), p7 4. 211 Henry W. Riecken, "Primary Groups and Political Party Choice," American Voting Behavior, op. cit., p. 172. 132 Tested time and time again, it has been found that political discussion is made up more of agreement than dis agreement, that people generally vote with their friends and co-workers, and that this political homogeneity increases 212 with age. However, as Norman Kaplan pointed out, a valid examination of the impact of reference groups on voting behavior must take into account the concept of awareness. Is the individual aware of the actual behavior of the 21 3 particular reference group? Data presented in The Voter Decides indicates that different groups have different perceptions of how given groups will vote, and that these 214 perceptions are quite often inaccurate. Kaplan examined two primary groups (family and friends) and three secondary groups (co-workers, formal associations, and unions). His findings may be summarized as: (1) respondents were aware 21 2 William A. Glaser and Charles Kadushin, "Politi cal Behavior in Midterm Elections," Public Opinion and Con- f ressional Elections, ed. W. N. McPhee and W. A. Glaser New York: The Free Press of Glencoe, 1962), pp. 261-66. Glaser and McPhee list 208 propositions distilled from pre vious political behavior studies and in Chapter 11 evaluate them in terms of their own four state midterm Congressional election panel data. A similar, but earlier, listing of such propositions can be found in Bernard Berelson, et al., Voting (Chicago: University of Chicago Press, 1954). 21 3 Kaplan, op. cit.. pp. 23-25. Kaplan used 1948 Elmira Study data. 214 Campbell, Gurin, and Miller, The Voter Decides, op. cit.. pp. 70-73 for actual vote of different demographic groups and pp. 210-14 for respondents1 perceptions of these groups’ votes. 133 of the voting norm for each group, (2) the primary groups were more likely to be reference groups than were the secondary groups, (3) secondary group norms were ''mediated" through primary group norms, and (4) no one group was 21 5 crucial since there was little conflict between them. In measuring "awareness," Kaplan found that at least two—thirds of respondents were aware of how their family, friends, co-workers, fellow union members, and fellow 216 members of formal associations would vote. In examining the family and voting behavior, Kaplan discovered that the larger the family the greater the frequency of disagree- 217 ment. He also found that sons tended to vote like their fathers, and that when both father and mother agreed in vote, respondents would also vote the same way in about 21 8 90 per cent of the cases. Turning to friends, Kaplan found that respondents^ were more likely to report political agreement with friends 219 than family. Similarly, a study of Cornell students 21 5 Kaplan, op. cit.. pp. 1-2 (abstract). 216 Ibid.. p. 70. Like others, Kaplan found that politically interested respondents were more likely to know the politics of their friends, etc., than were the uninter ested . 217 Ibid.. p. 90. This would be in accord with probability theory. 21^Ibid., pp. 146-52. 219Ibid., pp. 91-96. 134 revealed that those whose friends preferred Stevenson also preferred Stevenson, while those whose friends preferred 220 Eisenhower also preferred Eisenhower. Alice Kitt and David Gleicher found that respondents who changed their vote intention did so most often toward the political inclination 221 of their friends. Thus, one might come to the conclusion that while the individual cannot "choose” his family or its values, he can choose his friends— friends whose values he is in accord with. Further, while one is aculturated by his family in an essentially unilateral process, one's selection of friends comprehends a multi-lateral value exchange. Given the tendency to avoid conflict it appears, again, that as one grows older he established a greater 222 value homophily. Finally, Kaplan found that the formal associations were perceived as politically mixed and were, therefore, not very significant as reference groups. Kaplan's data, here, agrees with the earlier findings of Freeman and Showel 220 Peter I. Rose, "Student Opinion on the 1956 Presidential Election," The Public Opinion Quarterly. XXI, 3 (Fall, 1957), p. 375. Survey based on a 4 per cent random sample of Cornell students. 221 Alice S. Kitt and David B. Gleicher, "Deter minants of Voting Behavior," The Public Opinion Quarterly, XIV, 3 (Fall, 1950), p. 400. 222 . Michigan SRC data supports this hypothesis in that Campbell, et al.. have found a steady increase in the proportion of strong party identifiers as age advances. See, The American Voter, op. cit., p. 494. 135 (supra, fn. 209). However, his other data is in conflict with the findings of Campbell et al. in The American Voter. Using 1952 SRC data, they found that the rates of agreement- between respondent's vote and that of his "primary" groups were, in order: spouse, 91 per cent; family, 87 per 223 cent; friends, 84 per cent; and co-workers, 77 per cent. From this it would appear that the value homophily of friends is not as great as that of family' and spouse. (The problem here may well be that of respondent's age distribu tions and their levels of political interest.) Cross pressures How do people in conflict choose among their several interests? Reference group theory does not answer this question— although it does indicate that such conflict exists for only a small proportion of the population and for them only a short time. Bently discusses this, calling it 224 the "conflict of crossed groups." This appears to be similar to the later thesis of cross-pressures which 225 Berelson, et al., postulated in their Elmira study. Results of the Kaplan, Berelson, and Lazarsfeld studies 223Ibid., p. 77. 224 Bently, op. cit.. p. 204. 225 Berelson, et al., op. cit., pp. 283-84. Valter Lippman also discusses this in his Public Opinion (New York: Harcourt, Brace, and Company, 1922), p. 183. Note that both Lippman and Bently (1908) preceded Berelson by a good number of years. 136 indicated that individuals under cross-pressure reduce the pressure in at least one of two significant ways. First, one may withdraw from the area of conflict— manifested by a declining interest in politics. Secondly, one may shift his attitude(s) toward conformity with one of the reference 226 groups and withdraw from the other group. However, there seem to be some situations in which other patterns of behavior occur. Withdrawal from group membership may not always be possible— e.g., for a Negro. Lane reports that if conflict occurs within the family, 227 political discussion— interest— increases. Here, as in the case of the Negro, it is a situation in which withdrawal from membership is not readily available and the conflict cannot be ignored— so it must be resolved. Conflict among 228 friends is not so structured. The classic case of cross-pressure is one in which the individual is objectively a member of one class but 226 Kaplan, op. cit., pp. 100-101; Berelson, op. cit.. p. 27; and Paul F. Lazarsfeld, Bernard Berelson, and Hazel Gaudet, The People's Choice (New York: Columbia Uni versity Press, 1955),pp. xvi and 62. Also, see Kitt and __ Gleicher, op. cit.. supra. Lazarsfeld, et al., found that there were no per sonality characteristics which distinguished party changer from party constants. However, the personality of the changer more often resembled the personality of the adher ents of the party changed to than from, p. 70. 227 Lane, op. cit.. pp. 191-92. 228 Eleanor E. Maccoby, Richard E. Mathews, and Anton S. Morton, "Youth and Political Change," The Public Opinion Quarterly. XVIII, 1 (Spring, 1954), passim. 137 229 subjectively a member of another. Such cross-pressures need not lead to a reduction in political activity but may actually increase it. For example, an unskilled worker who identifies with the middle class may well become more polit- 230 ically active in conformity with middle class norms. And, as was shown earlier, in the discussion of status crystallization, response to low status crystallization depends largely on the relative levels of ascribed and achieved positions (p. 145, this chapter). Attitude stability Broadly speaking, an individual1s attitudes are stable. According to Lazarsfeld, neither "stubbornness" nor "inertia" explains such stability. Instead, the preser vation of existing attitudes provides a source of great satisfaction and success to individuals in their group relations. By maintaining these attitudes they are able to 231 avoid or minimize conflicts within their social reality. There appear to be two major dimensions involved in the relationship between the group and individual political behavior. Lane suggested that they are: (1) valence, and 229 See Eulau, Class and Party in the Eisenhower Years, op. cit. . for an extensive discussion of this phenomenon. 230 See Arthur Kornhauser, et al., When Labor Votes (New York: University Books, 1956), pp. 110-14. 231 Lazarsfeld, et al., The People's Choice, op. cit., p. xx. 138 232 (2) group relevance. By these Lane meant that a group will influence an individual's political behavior if the group is important to the individual's needs (valence), or if the political issue is relevant to the raison d'etre of the group. Lane's discussion is concerned with formal groups but the concept of valence would certainly apply to informal primary and secondary groups too. The concept of valence, in fact, would seem to be in line with the high level of political agreement found among spouses, family, friends, and co-workers. Relevance would seem to support the theory that the large, formal, and heterogeneous groups are likely to have little political influence aside from a specific and relatively narrow range of issues. Lane suggested that politically influential groups appear to be those defined along the religious, occupational, and ethnic 233 lines. These particular categories are value—laden and most often the origin of basic primary and secondary group relationships. Political attitudes and actions are only a part of the total of individual attitudes and behavior. As such, they too fall within the same dynamics which define social reality. Generally, groups define the content of morality and duty, structure belief about social 232 Lane, op. cit.. pp. 191-92. 233 Ibid.. p. 192. Lane also suggested income and residential groupings, but the ambiguity of these two categories leaves them open to challenge in this respect. environment, influence self-images, affect life goals, and suggest means of attaining these goals. More specifically — politically— groups define what is public and what is private, provide new associations for political partisan ship, provide new insights into the individual's stake in l politics, and reveal the relevance of specific policy 234 matters to the individual. In all these patterns, there is a powerful force for equalibrium and political stability. Some limits The research discussed above has led to various efforts to perfect models of voting behavior based on reference group membership. Thus, as a result of data reported in The People's Choice, a crude Index of Political Predisposition (IPP) was set up, designed to indicate the basic inclination of each respondent. The factors which Lazarsfeld, et al., used were religion, SES level, and 235 residence (urban—rural). However, V. 0. Key and Frank Munger have argued that: The style set in the Erie County study of voting, The People's Choice, threatens to take the politics out of the study of electoral behavior. The theoreti cal heart of The People's Choice rests in the conten tion that "social characteristics determine political preference."236 234 * Ibid.. p. 195. 235 Lazarsfeld, et al.« The People's Choice, op. cit.. Appendix B and pp. 25-27. 236 V. 0. Key and Frank Munger, "Social Determinism and Electoral Decision: The Case of Indiana," American Voting Behavior, ed. Burdick and Brodbeck, op. cit.. p. 281. 140 In their discussion of the Erie County study, Morris t Janowitz and Varren E. Miller report that according to Lazarsfeld the IPP can only be considered an index to a portion of the population and that the IPP indicates only 237 how voting intentions will be modified. Another attempt to set up a model for predicting voting behavior by membership in secondary groups having political norms can be seen in Philip Converse's Ph.D. 238 dissertation, "Group Influence in Voting Behavior." Like Lazarsfeld, et al., he found a relationship between 239 membership in the groups examined and voting behavior. Converse looked at four groups: labor union members, Jews, Catholics, and Negroes and found that individuals with high group identification were more inclined to be Democratic 240 than those with low identification. H. C. Cooper, in his Ph.D. dissertation, also examined the interrelationship 237 Morris Janowitz and Warren E. Miller, "The Index of Political Predisposition in the 1948 Elections," The Journal of Politics. XIV, 4 (November, 1952), p. 713. Janowitz and Miller cite a personal communication from Lazarsfeld. 238 Philip E. Converse, "Group Influence in Voting Behavior," (unpublished Ph.D. dissertation, University of Michigan, Social Psychology, 1958), microfilm. Converse used 1948, 1952, and 1956 SRC data. 239 Ibid.. p. 47. 240 Ibid.. p. 1. One should note that membership in these groups is largely involuntary and, except for union membership, ascribed. 141 of SES and reference groups as they affect voting 241 behavior. Cooper found that reference group identifica tion is a more immediate determinant of voting behavior than is SES.242 Recently, Ithiel de Sola Pool and R. P. Abelson reported on a more extensive (though less sophisticated) attempt to simulate 1960, electoral behavior on the basis of selected demographic characteristic which, when 243 cross-classified, produced 480 "voter types." If voting behavior and attitudes are determined to some extent by reference groups, then examination of the non-voter and the "don't know" respondent ought to reveal a population with little or no social linkage— a group of social isolates. Indeed, this is exactly what V. 0. Key suggested in saying that "... persons less completely integrated into the life of the community are less disposed 244 to participate in its politics." 241 Homer C. Cooper, "Interaction of Socioeconomic Characteristics and Reference Groups with Respect to Voting Behavior," (unpublished Ph.D. dissertation, University of Michigan, Social Psychology, 1957), microfilm. Cooper used 1952 SRC data. 242T. . A - Ibid., p. 3. 243 Ithiel de Sola Pool and R. P. Abelson, "The Simulmatics Project," Public Opinion Quarterly, XXV, 2 (Summer, 1961), pp. 167-83. 244 V. 0. Key, Politics, Parties, and Pressure Groups (New York: Thomas Y. Crowell Company, 1960), 4th ed., p. 636. 142 Philip K. Hastings, in examining the relationship between voluntary group membership and voting found a direct relationship. Those who belonged to at least one voluntary group tended to be voters more often than did 245 those who did not belong to any group. However, Hastings’ handling of the data obscures the point at hand— * note that his table defined voting rate as the independent variable. A reorganization of the data, defining group membership as the independent variable (as V. 0. Key sug- y gested) presents a more clear-cut relationship. Examination of both the IPP of Lazarsfeld and the Hastings data suggests that other factors in addition to, or modifying, reference group dynamics might lead to a more complete grasp of the determinants of voting behavior and attitudes. Among these might be considered a particu lar type of reference group— class; or, on the other hand, the broad range of factors included in the psychology of political behavior. The latter is, however, beyond the scope of this study. Class Political Behavior In spite of the doubts and criticism of class as a meaningful concept, in spite of the different definitions 245 Philip K. Hastings, "The Non-Voter in 1952: A Study of Pittsfield, Massachusetts," The Journal of Psychology, XXXVIII (July, 1954), p. 303. 143 TABLE 16 GROUP MEMBERSHIP BY VOTING RATE* Extent of Group Membership t. t ■ Occasional Non—voter „ . Voter Voter At least one 48# 64 # 71# N: 111 115 524 *Differences statistically significant TABLE 17 at p. < .001. VOTING RATE BY GROUP MEMBERSHIP Voting Rate Member Non-member Non—voter Occasional voter Voter 1196 15 74 23# 16 61 N: 499 251 144 of class, and in spite of the ambiguity of many class symbols, all voter research indicates that there is a class relationship (by whatever definition) in political behavior. As Lipset commented, the emphasis on "classlessness" in American political ideology has led many foreign and American political observers to conclude that party divisions in America are less related to class than in other 247 western nations. Certainly, the issues within the political arena influence the extent of class—related political behavior. As Arthur Kornhauser pointed out in 1950, a broad reading of the available survey data indicates that class political differences are greatest in regard to issues that directly and differently affect people in their 246 Campbell, et al.. The Voter Decides, op. cit.: Campbell, et al., The American Voter, op. cit.; Centers, The Psychology of Social Classes, op. cit.; Alfred de Grazia, The Western Public (Stanford; Stanford University Press, 1954); Eulau, Class and Party in the Eisenhower Yearst op. cit.; S. M. Lipset. Political Man (New York: Doubleday and Company, 1960); McPhee and Glasser, Public Opinion and Con gressional Elections, op. cit.; and Mills, White Collar. op. cit. 247 Lipset, Political Man. op. cit.. p. 285. This question may be viewed as a function of the individual’s position in society. Ralf Dahrendorf suggests that those at the bottom of the social scale look at society in terms of class while those at the top look at it in terms of strata. This is a significant difference in point of view as "strata” implies continuity, while "class" implies antagonistic groups. See Ralf Dahrendorf, Class and Class Conflict in Industrial Society (Stanford: Stanford University Press, 1959), pp. 206 and 283; originally published as: Soziale Klassen und Klassenkonflikt in der industriellen Gesell- schaft in 1957 and translated by the author. 145 respective class levels.Issues such as economic reform, regulation of economic affairs, the distribution of income, and the role of government in such areas evoke quite differ ent responses depending on the class position of the respondent. Other issues, such as foreign policy, race relations, and state-church relations failed to produce 249 such clear-cut differences along class lines. Eulau suggested that class position influences the role of class in political behavior, with the working class voting on the basis of perceived class interests more often than the middle class and that interest—oriented voters vote Demo cratic more often than non—interest—oriented voters, regardless of class position. Variations in political behavior by class are modified by other factors in the United States— for example, region. Campbell, et al., reported considerable differences in party identification by class for the different 251 regions. The greatest differences in party 248 Kornhauser, "Public Opinion and Social Class," op. cit., p. 334. 249 Ibid. Kornhauser wrote this before civil rights assumed a dominant position in the area of domestic issues. 250 Heinz Eulau, "Perception of Class and Party in Voting Behavior: 1952," American Political Science Review, XLIX, 2 (June, 1955), pp. 372-73. 251 Campbell, et al. , The American Voter, op. cit.. p. 158. This supports the Dahrendorf thesis cited in fn. 247, but the Centers, Strunk, and Campbell data, cited infra, pp. 145—148 are in conflict. 146 identification by class are to be found in the west, while the least differences are found in the northeast and south, TABLE 18 PARTI IDENTIFICATION BY CLASS BY REGION 252 ■ b J b Region Party Identification SDa WD I WR SR N Northeast Working class 20fo 22 27 16 15 492 Middle class 10?S 14 30 21 25 317 Midwest Working class 22fo 25 27 15 11 649 Middle class 1 3 % 17 25 19 26 435 West . Working class 2 1 % 30 26 11 6 265 Middle class 16 io 14 25 18 27 145 South Working class 34 io 36 13 9 8 531 Middle class 29^ 34 20 9 8 311 SD = Strong Democrat; WD = Weak Democrat; I = Independent; WR = Weak Republican SR = Strong Republican 252 Calculation of correlations of class by party for the four regions produced low figures of .151 for the northeast and .048 for the south; a moderate correlation of .283 for the midwest; and a higher correlation of .315 for the west. Correlations were calculated by the formula: Y = ni fx'y' - ( ifx’) (Ify1) t/n(f(x.*) - f x')2) n(f(y'2) - (ify1)2) This is a modification of the standard linear formula and is required because the data examined is grouped; see Lillian Cohen, Statistical Methods for Social Scientists (New York: Prentice Hall, Inc., 1954), pp. 152-54. Actual 147 Other data indicate that class as a factor in political behavior is reduced in rural areas as compared to urban areas. Just as there is only one party in the south, the Democratic, so there tends to be only one party— Repub- 253 lican— in many rural areas (outside the south). One party environments appear to reduce party division by class. Philip H. Ennis' data tend to support the Epstein findings. Results from Washington and Colorado panel surveys show a relation between size of voter's community and difference in vote for congressmen by blue and white 254 collar workers. While there was a difference of 43 percentage points between the white collar and blue collar Democratic vote in Washington metropolitan areas, there was only a 16 percentage point spread in Washington rural areas. 255 A similar pattern was also found in Colorado. These class-political relationships are known to the public as well as to political scientists. Mildred calculation was done on the Honeywell 800, University of Southern California's Computer Sciences Laboratory, with a program written by the author. 253 See Leon D. Epstein's "Size of Place and the Division of the Two-Party Vote in Wisconsin," Western Political Quarterly. IX, 1 (March, 1956). 254 Philip H. Ennis, "The Contextual Dimension in Voting," Public Opinion and Congressional Elections, ed. McPhee and Glasser, op. cit., p. 187. 255 Ibid. The one party, classless politics of rural areas may not be so in fact. The correlations cited in fn. 252 supra suggest that there may be unexplained inter vening variables at work. 148 Strunk reported findings of the American Institute of Public Opinion that people in 1949 saw each of the two major polit ical parties as serving the interests of different groups 236 within the nation (see Table 19). Asking members of each group which party would best serve the interests of the group produced a similar 217 pattern (see Table 20). Not only did most people perceive a clear-cut group or class partisan interest for others but for themselves as well— with one major exception, the white collar class. Respondents as a whole saw the Democratic party as serving the interests of the white collar class slightly better than would the Republican party, while white collar respon dents saw the Republican party as serving their interests slightly better. Except for the white collar respondents, each group's members and respondents as a whole agreed on which party would best serve the interests of each group. Later data, reported in The Voter Decides, shows the same pattern. The respondents as a whole viewed most 218 groups as favoring one party or another (see Table 21). For a large majority (about 80 per cent) of the ^"^Mildred Strunk (ed.), "The Quarter's Polls," The Public Opinion Quarterly, XIII, 1 (Spring, 1949), p. 169. ^^Mildred Strunk (ed.), "The Quarter's Polls," The Public Opinion Quarterly. XIV, 1 (Spring, 1950), p. 183. -^CO Campbell, et al.. The Voter Decides, op. cit., p. 211. ------------------ 149 TABLE 19 PERCEPTION OF PARTY BEST SERVING GROUPS, PARTY BY GROUPS Group Demo Repub No Dif No crat lican ference Opinion Unskilled workers 6 9 % 8 10 13 Farmers 6 8 % 10 9 13 Skilled workers 6 2 % 10 12 16 White collar '39% 26 17 18 Business owners and professional people 2 7 % 46 11 16 TABLE 20 PERCEPTION OF PARTY BEST SERVING PARTY BY GROUPS OWN GROUP Group Demo crat Repub lican No Dif ference No Opinion Unskilled workers 6 4 % 10 10 16 Farmers 6 5 % 16 10 9 Skilled workers 5 9 % 12 11 18 White collar 3 3 % 37 19 11 Business owners and professional people 24 59 9 8 TABLE 21 PERCEPTION OF PARTY BY PREFERENCE OF VARIOUS GROUPS, PARTY BY GROUP Group Perception of Party Preference8, Demo crat Repub lican Split Don* t Know Not As certained N Farmers 35$ 14 27 24 _ 1,614 Labor union members 61$ 5 12 22 - 1,614 Vorking-class people 51$ 7 25 17 - 1,614 Middle-class people 19$ 24 31 26 - 1,614 Big-businessmen 12$ 55 9 24 - 1,614 Negroes 34$ 10 15 41 - 1,614 Protestants 4$ 11 40 44 1 1,614 Catholics 17$ 8 25 49 1 1,614 Jews 11$ 8 19 61 1 1,614 Partial table, ''total" omitted. V J l o / 151 respondents, religion had little or no political content. On the other hand, a clear majority of respondents saw a relationship between being a member of the working class, or being a businessman, and voting for one party or the other. Further, the perceived choice of these two groups was quite clear! But, once again, the middle class appears to have been a fairly ambiguous concept (31 per cent "split" and 26 per cent "don't know")— only 43 per cent of all respon« dents could attribute a political choice to the middle class and that choice was not as clear-cut as for business men or working class members. Turning to respondents' perception of their own group's voting behavior, we find that, as in the Quarter's polls, respondent's own class or economic group was a 259 meaningful political frame of reference. Interestingly, where "middle class" had been ambiguous for respondents as a whole (see Table 21), middle class members themselves saw the middle class as politically meaningful and perceived a clear-cut choice for their class between the two parties (see Table 22).260 259Ibid.. p. 214. 2^Vhile AIPO data (1948 and 1949) showed white collar members split in party preference, SRC data for 1952 showed a definite party choice. This may have been an Eisenhower phenomenon, see The Voter Decides, op. cit., p. 72 for the impact of Eisenhower on the white collar vote and Robert R. Alford's Party and Society (Chicago: Rand McNally and Company, 1963),pT 352 which indicates that some of the middle class returned to the Democratic Party in 1960. 152 TABLE 22 PERCEPTION OF PARTY PREFERENCE OF VARIOUS GROUPS, PARTY BY GROUP MEMBERS AND BY NON-MEMBERS Group Perception of Party Preference0, Demo crat Repub lican Split Don't Know N Farmers Members*3 26# 18 37 18 183 Others0 34?6 14 27 25 1 ,431 Labor Union Members 72# 3 14 11 440 Others 57# 5 11 27 1 ,174 Vorking class Members 54# 8 22 16 504 Others 48 8 26 18 1,110 Middle class Members 15# 47 29 9 200 Others 19# 21 32 28 1 ,414 Partial table; "not ascertained" and "total" have been omitted. ^Members of the group. cNon-members of the group. 153 Asking people which attitude or opinion they would ascribe to members of various classes produced a definite 1 variation by class. Both in—class and out—class respon dents ascribed definite differences to the four classes, as described in Tables 23 and 24. Trends in class-politics Whether or not the class basis of political behavior is increasing or decreasing, or changing its orientation (e.g., from domestic-economic to international) is an unresolved question. The authors of The American Voter came to the conclusion that class voting has decreased in the past few years— with the decreasing salience of domestic-economic issues, and the increase in "non-class" issues of nationalism and foreign policy. On the other hand, Robert Alford found that class voting has not decreased. The American Voter describes how, since 1948, the correlation between "status polarization" (used to index class) and vote has decreased— for both presidential and i i +• 263 congressional elections. 261 Richard Centers, "Social Class, Occupation, and Imputed Belief," The American Journal of Sociology. LVIII, 6 (May, 1953), p. 554. See Campbell, et al.. The American Voter* op. cit.. pp. 346-50; and Robert R. Alford, "The Role of Social Class in American Voting Behavior," Western Political Quarterly. XVI, 1 (March, 1963), passim. 263 Campbell, op. cit.. p. 347. TABLE 23 154 THE CLASS ASCRIBED TO PERSONS REPUTED TO BELIEVE THAT "STRIKES SHOULD BE COMPLETELY FORBIDDEN BY LAV" Respondent's Class Ascription®- Own Class Membership Upper Middle Vorking T Other and Lower Undecided N Upper 62%> 10 7 7 14 29 Middle 5 7 % 15 10 9 10 407 Vorking 4 6 % 14 13 13 14 700 Lower 3 3 % — 22 — 44 9 Other and undecided 3 3 % 4 8 8 44 24 All R's 5 0 % 14 12 11 13 1,169 Percentages recalculated from a partial table. TABLE 24 THE CLASS ASCRIBED TO PERSONS REPUTED TO BELIEVE THAT "GOVERNMENT SHOULD GUARANTEE EMPLOYMENT AND DECENT STANDARD OF LIVING TO ALL"264 Respondent's ________________ Class Ascriptiona Own Class Membership Upper Middle Vorking Lower Other and Undecided N Upper 1 296 12 21 41 1 5 34 Middle 2 11 31 48 8 41 5 Vorking 5 8 38 37 12 711 Lower 9 18 27 — 45 11 Other and undecided 4 8 16 28 44 25 All R's 4 9 34 40 12 1,196 Percentages recalculated from a partial table. 264 Centers, 'Social Class, Occupation, and Imputed Belief," op. cit., p. 554. Campbell, et al., looked at the presidential elec tions between 1944 and 1956 and found a decreasing impact of class starting with a high in 1948 and a low in 1956. Alford's later article examined class voting from 1936 through 1960 and suggested that the "low" class vote in 1952 and 1956 was unique for -the period 1936—1960. Alford also found that the 1960 presidential election "restored" class voting. This would indicate that the trend observed by Campbell, et al.. may have been an Eisenhower phenomenon. Alford found that the differences, or "gaps," in the support by each class for the two parties remained fairly constant, and that the changes in class support for each party's C candidates were usually in the same direction. Arthur N. Holcomb suggested that there can be no national majority for a party which scorns the support of 266 the middle class. However, the middle class is such an ambiguous and broad grouping that one suspects each major party in the United States appeals to its own "middle class." Thirty years ago Harold Lasswell predicted a world revolution in the rise of the middle class provided that it Alford, "The Role of Social Class," op. cit.. p. 184. The 1948 pattern appears to be an anomaly in that the "gap" widened due to divergent trends in class support for the two parties. 266 Arthur N. Holcomb, Our More Perfect Union (Cambridge: Harvard University Press, 1950) , p"! 419. 156 acquired a unifying symbol——he suggested that the symbol might well be its sacrifices in acquiring socially useful skills.267 Writing a little later than Lasswell, Arthur W. Kornhauser found that the middle class had a split person ality, tending to agree with the lower class on economic questions but with the upper class on maintenance of Q status. As will be shown in the following section— this bipolarization continues to exist. Objective class and political behavior A great deal of work has been done in examining the relationship between objective class and political behavior — most of which uses occupation as an index to class. Alfred de Grazia, in examining the voting behavior of western America, found that political identification 269 varied by both occupation and income. (See Table 25.) De Grazia suggested that class predicts political affiliation better than does income; calculation of the correlations 267 Harold Lasswell, "The Moral Vocation of the Middle-Income Skill Group," International Journal of Ethics. XLV (1935), pp. 127—37. See Grayson's definition of the middle class as the class of aspiration, Chapter II, p. 18. 268 Arthur W. Kornhauser, "Attitudes of Economic Groups," Public Opinion Quarterly. II, 2 (April, 1938), p. 262. 269 Partial table, de Grazia, op. cit.« p. 107, SRC data for 1952. i 15? TABLE 25a PARTY IDENTIFICATION BI OCCUPATIONAL STRATA BY INCOME Occupation Strong Dem. Weak Dem. Weak Rep. Strong Rep. N Self-employed business men and artisans, managers and officials $-2,999 — 2 8 % — 27 11 3,000-4,999 14 % 29 7 22 14 5,000+ 24 % 14 14 19 21 Clerical and sales $-2,999 2 0 % 20 — 20 5 3,000-4,999 21^6 26 13 5 39 5,000+ M % 17 24 18 21 Skilled and semi-skilled $-2,999 ^ 4 % — 29 — 7 3,000-4,999 3 3 % 29 17 — 24 5,000+ 15 % 28 15 6 33 Unskilled and farm labor $-2,999 3 2 % 21 5 21 19 3,000-4,999 6 7 % 16 — 17 6 5,000+ 3 3 % 33 — 34 3 £L Partial table. "Independent Democrat," "Indepen dent," and "Independent Republican" have been omitted. 158 270 for de Grazia's data supports his conclusion. Correla tion of class to party was .172 while correlation of income to party was .037. Holding income constant, a slightly higher correla tion of .211 was found for class and party in the middle income bracket ($3 ,000—$5,000). There were also low correlations, .165 and .202, between class and party for the high and low income groups (above $5,000 and below $3,000, respectively). Very low correlations— both plus and minus-— were found for various income levels with class held constant. This analysis of de Grazia's data suggests that only within one income segment of the population, the middle income group, is there much correlation between 271 class and party. But, more importantly, income was obviously a very poor prediction of political behavior within classes. A similar result was reported by Andrew R. Baggaley, 270 however, who calculated correlations (unlike de Grazia). Comparing the vote for Villiam Proxmire (Dem.) and Walter 270 Correlations calculated by formula given in fn. 252. 271 As early as 1940, Elmo Roper had pointed out that cash income was a very poor measure of status in correlation with political behavior. See Elmo Roper, "Classifying Respondents by Economic Status," Public'Opinion Quarterly. IV, 2 (Summer, 1940), passim. 272 Andrew R. Baggaley, "White Collar Employment and Republican Vote," Public Opinion Quarterly, XX, 2 (Summer, 1956), passim. 159 Kohler (Rep.) in the 1954 Wisconsin gubernatorial election by wards in Milwaukee, he found that the correlation between median income and vote for Kohler was +.35 (low, 273 but in the expected direction). However, he found a very high correlation of +.93 between white collar 274 occupation and Kohler vote. Other studies tell the same story. Angus Campbell, et al., reports that the higher the occupation status the 275 greater the Republican vote in 1948 and 1952. William Glasser and Charles Kadushin, in their study of off-year congressional elections, report that occupation, even with 0 7 SES controlled, correlated with party choice. Ennis' article, based on the same data, comes to the same con- 277 elusion. Other data, supplied by Dr. Glasser, which 273Ibid., p. 472. 274 Ibid. Fortunately, Milwaukee ward boundaries co incided with the 1950 census tract boundaries. Baggaley used a product moment correlation, defining white collar as: professional, managerial, sales, and clerical. 275 Angus Campbell, Gerald Gurin, and Warren E. Miller, "Political Issues and the Vote; November, 1952," American Political Science Review. XLVII, 2 (June, 1953), p. 379; and also see The Voter Decides, op. cit.. p. 72. 0 7 f \ William A. Glasser and Charles Kadushin, "Polit ical Behavior in Midterm Elections," Public Opinion and Congressional Elections, op. cit. . p. 259. Another study indicates that occupation with subjective class controlled still relates to political affiliation. Raymond J. Murphy and Richard T. Morris, "Occupational Situs, Subjective Class Identification, and Political Affiliation," American Sociological Review. XXVI, 3 (June, 1961), p. 390. 277 Ennis, op. cit.. p. 186. follows the design described in Appendix A of Public Opinion and Congressional Elections further substantiates 278 the significance of occupation in political behavior. Mills, in his study, White Collar, provides data for the 1936, 1940, and 1944 presidential elections which shows a 279 relationship between occupation and vote. Since Mills relies upon published Gallup data, he does not include the 1948 figures. However, a press release from the University of Michigan News Service of February 21, 1957, cites 1948 survey data which follows the patterns of earlier and sub- 280 sequent presidential elections. There is, then, data drawn from national, regional, and state samples, and for presidential, congressional, and gubernatorial elections, which consistently shows a relationship between occupation and voting behavior. Occupation also relates to a sense of political efficacy, according to Campbell, et al.. and to a sense of political relatedness (a combination of efficacy and 278 Personal communication from Dr. Glasser, May, 1963. Dr. Glasser very kindly supplied occupation by party for each of the four states, for which grateful acknow ledgment is made. 270 Mills, White Collar, op. cit.. p. 332. 280 University of Michigan News Service, Press Release, February 21, 1957 (mimeographed), p. 3. 16t 281 responsibility). Vayne Thompson and John Horton found a negative correlation between occupational rank and political alienation, which they define as a lack of per ceived power in community affairs and a distrust of those 282 in power positions. Finally, while most studies have been concerned with partisan political actions, a recent study of bond issue voting behavior indicates that there is a similar relationship between occupation and choice in 283 such non-partisan, non-personality, elections. Subjective class and political behavior Maclver, writing in Society in 1937, took the posi tion that the concept of class loses its sociological sig nificance if it is defined by only objective criteria— that class does not unite people (or separate them) unless they 284 feel such unity (or separation). Eulau has suggested 281 Campbell, et al.. The Voter Decides, op. cit.. p. 191; and Heinz Eulau and Peter Schneider, "Dimensions of Political Involvement," Public Opinion Quarterly, XX, 1 (Spring, 1956), pp. 131-32. 282 Vayne E. Thompson and John E. Horton, "Political Alienation as a Force in Political Action," Social Forces. XXXVIII, 3 (March, 1960), p. 195. 283 Alvin Boskoff and Harmon Zeigler, Voting Patterns in a Local Election (Philadelphia: J. P. Lippin— cott Company, 1964), pp. 105-106. 284 Maclver, Society, op. cit.. p. 167. See Chapter II, this dissertation, for a discussion of the content and dimensions of subjective class. 162 that the relationship between objective class and political behavior is often affected by the intervening variable of 9OC subjective class affiliation. He found about one-fourth of the respondents in both the 1952 and 1956 SRC surveys identified with one class while being objectively assigned to another (see Chapter II, p. 70); he also found differ ences in attitudes which related to class self- 286 identification. Eulau warned, however, that identifica tion is not the same thing as class consciousness. Mills, in his White Collar, defined class conscious ness as identification with a class, belief in the existence of class interest, and belief in the existence of other class interests in opposition to one's own class. Class-consciousness has always been understood as a political consciousness of one's own rational class interests and their opposition to the (or by the) interests of other classes. Economic potentiality becomes politically realized: a "class in itself" becomes a "class for itself.?287 Oscar Glantz recently examined the "consciousness" and politics of two traditionally opposed classes— laborers 288 and businessmen. He found that 40 per cent of those who Heinz Eulau, "Identification with Class and Political Role Behavior," Public Opinion Quarterly, XX, 3 (Fall, 1956), p. 516. 286Ibid., pp. 519-20. 287 Mills, op. cit., p. 325. This concept of class consciousness suggests that those who are upwardly mobile could not identify with their objective class origins. 288 Oscar Glantz, "Class Consciousness and Political Solidarity," American Sociological Review, XXIII. 4 (August. 1»5»). ---------------------------------------- 163 were oriented to big business and 28 per cent of those 289 oriented to labor were class conscious. Class conscious big businessmen were unanimous in their support of Dewey (1948) and Eisenhower (1952), while the class conscious laborers supported Truman (91 per cent) and Stevenson (1952) 86 per cent.29* " * Thirty years ago, Hadley Cantril stated that the overwhelming majority of the American people iden tify themselves with some category of the great middle class.291 There is, however, good reason to conclude that a majority of people are subjectively (as well as objectively) members 292 of the working class. This is not to fault Cantril; the polls conducted prior to 1943 offered respondents their choice of: upper, upper-middle, middle, lower-middle, and 293 lower. In 1946, when AIPO offered respondents their choice of: upper, middle, lower, and worker, 51 per cent 294 chose "worker." Much, then, depends on the categories 289Ibid.. p. 382. 290 Ibid. These are higher levels of support than Campbell, et al., report; see The Voter Decides, op. cit., pp. 72-73. 291 Hadley Cantril, "Identification with Social and Economic Class," The Journal of Abnormal and Social Psychol ogy. XXXVIII, 1 (January, 1943), p. 78. 292 See Chapter II, p. 293 Hadley Cantril, Public Opinion 1935-1946 (Prince ton: Princeton University Press, 1951), p. 116. ^9^Ibid. See also, Chapter II, p. 78, this disser tation . 164 used. For example, Dahrendorf cited the definitions of Popitz ("wirklich arbeiten und nich wirklich arbeiten"), Willener ("ceux qui travaillent et ceux qui ne travaillant pas”), and Centers ("working for a living and not working 295 for a living"). These definitions are essentially the dichotomous heads-or—hands criteria previously cited (Chapter II, p. 75). Centers examined the different patterns of response to two four-choice class scales with alternatives or "working" and "laboring": upper-middle-working—lowert or upper-middle-laboring-lower. He found a consistently greater self-identification with "working" than with 296 "laboring." Centers also found, in a study published earlier, that class identification has an effect upon political behavior. If people's class identifications are the same, their attitudes tend to be similar even though their objective occupational positions are d i f f e r e n t .297 Centers shortly thereafter followed up this article by a whole work devoted to subjective class behavior and 295 Dahrendorf, op. cit.. p. 286, fn. 5. 296 Richard Centers, "Nominal Variation and Class Identification: The Working and Laboring Classes," The Journal of Abnormal and Social Psychology« XLV, 2 (April, 1950), pp. 198-203. 297 Centers, "The American Class Structure: A Psychological Analysis," Readings in Social Psychology, ed. T. M. Newcomb and E. L. Hartley (New York: Henry Holt and Company, 1947), p. 490. 165 298 attitudes. In this he reported the relations between a Conservative-Radical (C-R) scale with both strata (occupa tional ranking) and subjective class; attitudes of different occupational strata; class identification of different occupational strate; 1944 presidential vote by class and strata; and several 6ther attitude differences between class (subjective) and strata (objective). For example, he found that there was a positive correlation between C-R 299 and strata (see Table 26). He found similar patterns with subjective class (see Table 27). A calculation of the correlations between C-R scales and the two variables produces a higher correlation between occupational strata and C-R (.453), than between 301 subjective class and C-R (.318). Other data which Centers provided substantiates these correlations (Centers 302 did not calculate the correlations). (See Table 28.) Centers thus held strata "constant” while examining the relationship between subjective class and his C—R scale. For the data, as given by Centers, the correlation between 298 Centers, The Psychology of Social Class, op. cit. 299Ibid.. p. 57. ~^°Ibid. , p . 1 20. ^^See fn. 24 for formula. Also, see Kahl' s statement, this dissertation, Chapter II, p. 79. 302 • Centers, The Psychology of Social Class, op. cit*» p. 126. 166 TABLE 26 CONSERVATISM-RADICALISM BI OCCUPATIONAL STRATA „ Conser- Inter- D ,. , Ultra- „ Conser- .- ,. , Radical n ,. , N .. vative mediate Radical vative Large business 56# 32 11 0 2 54 Professional 30# 40 19 4 7 73 Small business 46# 28 18 7 2 131 White collar 24# 31 29 11 5 172 Skilled manual 12# 26 34 17 10 163 Semi-skilled manual 5# 16 29 29 21 174 Unskilled manual 3# 21 39 21 17 77 TABLE 27 CONSERVATISM-RADICALISM BI CLASS Ultra Conser vative Conser vative Inter mediate Radical Ultra- Radical N Upper 42# 24 17 10 7 29 Middle 35# 33 21 7 4 467 Vo rking 1 2# 23 33 19 1 3 564 Lower — 23# 31 31 31 15 167 TABLE 28 CONSERVATISM-RADICALISM BI OCCUPATIONAL STRATA BY CLASS* Cons ervative Intermediate Radical N Business, profes sional, and white collar Middle class 74 $ 20 6 298 Working class 30 23 100 Manual workers Middle class 37 $ 30 33 83 Working class 25$ 34 41 318 *Partial table 168 class and C-R for the upper strata was .283 and for the lower strata +.308. By reorganizing the data it was possible to hold class constant and find the correlations between strata and C-R. With the middle class "constant," the strata-C-R correlation was .386, and for the working class "constant" the strata-C-R correlation was .299. On the whole, strata was a slightly better predictor of 303 C-R than was class. Without the use of correlations Centers came to the same conclusion. The conviction has gradually emerged during the course of this survey that class affiliations and politico-economic orientations are both to a very large extent direct consequences of the differing socio economic positions of people in our system of produc tion and exchange of goods and services.304 . . . occupational stratification is a distinctly more important index to class identification and conserva tism- radical ism than any of those that have been examined.305 In spite of the better predictive powers of strata than class, Centers offered firm evidence of the existance and impact of subjective class on political attitudes and behavior. He also offered further confirmation of the 303 There are several weaknesses in Centers’ methods, not the least of which is that his C-R Scale is multi dimensional. See, Herman M. Case, "G-uttman Scaling Applied to Centers’ Conservatism-Radicalism Battery," The American Journal of Sociology. LVIII, 6 (May, 1953), pp. 556-63. 304 Centers, The Psychology of Social Class, op. cit., p. 160. 3Q5Ibid.. p. 191. 169 earlier NORC data that people do define class in subjective as well as objective terras.Since Centers was more con cerned with the subjective nature of class, generally, than the correlates between subjective class and political behavior specifically, it is not surprising that his work is inconclusive in regard to the latter. A later work directly concerned with the relationship of subjective class and political behavior is that of Heinz Eulau, Class 307 and Party in the Eisenhower Years. Eulau's findings are significant in two dimensions. First, he found that about one quarter of the respondents were in conflicting categories, that is, they identified with one class while being objectively members of the other 308 (using a two-class system). And, secondly, he found that while "... class identification had some effect on party identification and other political attitudes, . . ."it was not very effective as a determinant of political 309 behavior. This appears to conflict with an earlier assessment. Class identification was found to differentiate significantly between those who identify themselves as Democrats or Republicans . . . In general, regard less of whether they are objectively working-class or 306 Ibid.. p. 91; and see this chapter, p. 72. 307 Eulau, op. cit. 308 Ibid.. Table 3, p. 62. 3Q9Ibid.. p. 139. 170 middle class, -those identifying with the working class tend to identify themselves as Democrats whereas those identifying with the middle- class tend to identify themselves as Republicans.310 In this earlier study Eulau also found a relationship 311 between class identification and political behavior. Vhy the apparent conflict? The key lies in the difference between looking at the discovered relationships as causal or as parts of a larger system. While, as Eulau stated, there are clear- cut relationships between class identification and party identification, this hardly established a causal relation ship. And, indeed, as he illustrated in his examination of the relationship eight years later, class identification was not very effective in predicting political behavior. But Eulau's data clearly indicated some relation ship between subjective class and political identification (see Table 29).312 Similar results were found when class configura tions were related to respondents' presidential vote (see Table 30).313 3^Heinz Eulau, "Identification with Class and Political Perspective," The Journal of Politics, XVIII, 2 (May, 1956), p. 251. 311 'ibid., p. 252. 31 2 Eulau, Class and Party, op. cit.. p. 62. The three tables are literal translations of Eulau's data. 31 3 ,JIbid., p. 66. 171 TABLE 29 PARTY IDENTIFICATION BY SUBJECTIVE-OBJECTIVE CLASS PATTERNS Class Configuration Democratic Party Identification Subj ective Obj ective 1952 1956 Working Working 59% 50% Working Middle 49 % 48% Middle Working 39% 41% Middle Middle . 32% 36% TABLE 30 PRESIDENTIAL VOTE BY SUBJECTIVE-OBJECTIVE CLASS PATTERNS Class Configuration Democratic Presidential Vote Subj ective Obj ective 1952 1956 Working Working 60% 41% Working Middle 43% 39% Middle Working 37% 31% Middle Middle 23% 31% 172 And, also, for attitudes between class and 31 4 governmental activities (see Table 31). TABLE 31 ISSUE ATTITUDE BY SUBJECTIVE-OBJECTIVE CLASS PATTERNS Class Configuration Subjective Objective "Government 1952 Should Do More" 1956 Vo rking Working 31 % 5 0% Working Middle 3 4 % 2 8 % Middle Working 23 % 3 1 % Middle Middle 2 4 % 2 2 % Looking at these three tables, it is clear that the polar values were established by class consistents— that is, where respondents' objective and subjective class support each other. The third table indicates something of a muddled relationship, with the polar values established by respondents in "mixed" categories for 1952. By first holding objective class constant and then subjective class constant for each of the tables, it was observed that in five out of six cases, subjective class modified respondents' positions to a greater extent than did objective class. The conclusion is, therefore, that for the Eulau political data, subjective class appeared to 314T. .. Ibid., p. 64. 173 31 5 play a larger role than does objective class. Conclusions , No firm conclusions may be drawn from the research to date as to whether subjective or objective class is the best predictor of politibal behavior. First, because the data have not been subjected to causal analysis— correla tions have been discovered but not causal relationships. Secondly, it may well be that for some aspects of political behavior— for example, the vote and party identification— that objective class plays a larger role than does subjec tive class; while for other aspects of political behavior— for example, issue stands or perceptions of government policy— that subjective class plays the larger role. Thirdly, it may well be that in some situations the independent variable is not subjective class but rather party. For example, the upwardly mobile may retain his party identification while changing his objective class. After becoming middle class by objective criteria he may find himself in conflicting political positions— conflicts which he resolves not by changing party identification but by changing class identification (a change, probably, that conforms with his parents' objective class). Vhile this 31 5 Another, earlier, article which lends weight to this thesis is one by Homer C. Cooper, "Social Class Identification and Political Party Affiliation," Psychologi cal Reports. V, 2 (June, 1959), pp. 337-340. 174 is the exact opposite of the prevailing popular theories about the relationship between social mobility and political behavior, there are some reasons for suggesting that it may, in fact, occur. In the first place, the patterns of occupational distribution in our expanding industrial society have been such that many people have literally found themselves pushed, willy-nilly, into the lower ranks of the middle class without either much effort, aspiration, or 316 motivation. Those in such positions may find that in objective terms of income and, perhaps, occupation they are middle class, but that their style of life and attitudes preclude easy access into the "real1 1 middle class. It might well be that such individuals find psychic satis faction in continuing their identification with the working class and manifesting it in their political behavior. Lenski's research lends some support to this hypothesis— wherein he found that those with status inconsistencies 317 were more often Democrats than Republicans. And, in the study of transplanted automobile workers by Bennett M. Berger, it was found that one may acquire the trappings 318 of suburbia and still vote Democratic. If party 316 See the comments in this chapter, p. 93. 317 Lenski, "Status Crystallization . . . ," op. cit. 31 g Bennett M. Berger, Working Class Suburb (Berke ley and Los Angeles: University of California Press, 1960). identification is the enduring independent variable which it appears to be, then subjective class may, in some cases, be the dependent variable. CHAPTER III j i SUBURBIA | j The patterns of suburban political behavior and the patterns of political behavior as related to variables commonly associated with suburbia have led, generally, to the conclusion that suburbs vote Republican. However, as was suggested in Chapter I, the ''Republican suburb" may be a phenomenon largely dependent upon a candidate (e.g., Eisenhower) or the particular definition of suburbia. Suburbia Defined Previous definitions Suburbs have been most often defined as the "ring" around the central city. Such a definition has been employed by the Bureau of the Census; the Republican and Democratic national committees; Thomas R. Dye; Lewis Mumford; Jean Gottman; R. E. Park, E. V. Burgess, and R. D. McKenzie; Amos Hawley and Basil Zimmer; Scott Greer; 1 and William Zeckendorf. Fred Greenstein and Raymond ^U.S. Bureau of the Census, U.S. Census of Popula tion: 1960. Number of Inhabitants, United States Summary. Final Report PC (1)-1A. U. S. Government Printing Office, Washington, D.C., 1961. Research Division, Republican 177 Wolfinger defined suburbia as a dormitory at the edge of 2 central city, as did Lawrence Fuchs. John Seeley, R. A. * Sim, and Elizabeth Loosley used a narrower definition, calling suburbia an elite dormitory, as did Alfred 3 de Grazia, and Leo Schnore and David Varley. However, both Bernard Lazerwitz and Bennett Berger found that 4 suburbia could be working class as well as elite. Schnore National Committee, "The 1960 Elections" (Washington, D.C.: Mimeograph, 1961); Research Division, Democratic National Committee, "The I960 Election Report" (Washington, D.C.: Mimeograph, 1961); Thomas R. Dye, "Certain Political Cor relates of the Areal Differentiation and Stratification of Sub-populations in a Metropolitan Area" (unpublished Ph.D. dissertation, Department of Political Science, Pennsylvania University, 1961); Lewis Mumford, The City in History (New York: Harcourt, Brace and World, Inc., 1961);Jean Gottman, , Megalopolis (New York: Twentieth Century Fund, 1961); R. E. : Park, E. W. Burgess, and R. D. McKenzie, The City (Chicago: University of Chicago Press, 1925); Amos H. Hawley and Basil G. Zimmer, "Resistance to Unification in a Metropolitan Area," Community Political Systems, ed. Morris Janowitz (Glencoe: Free Press, 1961), pp. 146-84; Scott Greer, "Catholic Voters and the Democratic Party," Public Opinion Quarterly. XXV (Winter, 1961), 4, pp. 611-25; William Zeckendorf, "Cities Versus Suburbs," The Atlantic Monthly. July 1952, pp. 24-28. 2 Fred I. Greenstein and Raymond E. Wolfinger, "The Suburbs and Shifting Party Loyalties," Public Opinion Quarterly. XXII, 4 (Winter, 1958), pp. 473-82; Lawrence H. | Fuchs, "American Jews and the Presidential Vote," American ! Political Science Review. IL, 2 (June, 1955), pp. 385-401. 3 John R. Seeley, R. Alexander Sim, and Elizabeth W. Loosley, Crestwood Heights (New York: Basic Books, Inc., 1956); Alfred de Grazia, The Western Public (Stanford: Stanford University Press, 1954); Leo F. Schnore and David W. Varley, "Some Concomitants of Metropolitan Size," Ameri can Sociological Review. XX, 4 (August, 1955), pp. 408-14. Bernard Lazerwitz, "Some Characteristics of Resi dential Belts in the Metropolitan Community" (unpublished 178 ; later found that suburbs could also have many degrees of i industrialization and were not just necessarily dormi- f 5 ' tories. Lazerwitz has suggested that suburbia is defined j gj by recent growth, as have Stuart Queen and David Carpenter. , The authors of The American Voter and T. J. Anderson have both suggested that suburbs may also be defined by their 7 high population mobility. And, finally, both Sylvia Fava and George Lundberg, et al., have suggested that suburbia Q is really a "way of life." For this work An earlier definition, used at the beginning of this definition to demonstrate the fallacy of equating Ph.D. dissertation, University of Michigan, 1958); Bennett M. Berger, ¥orking-Class Suburb (Berkeley and Los Angeles: University of California Press, 1960). ■*Leo F. Schnore, "The Functions of Metropolitan Suburbs," American Journal of Sociology. LXI, 5 (March, 1956), pp. 453-58. ^Bernard Lazerwitz, "Metropolitan Community Resi dential Belts," American Sociological Review. XXV, 2 (April, 1960), pp. 245-52; Stuart Queen and David B. Carpenter, The American City (New York: McGraw-Hill Book Co., Inc., 1955). j 7 1 Angus Campbell, Philip E. Converse, Varren E. ! Miller, and Donald E. Stokes, The American Voter (New York: j John Wiley and Sons, Inc., 1960); Totton J. Anderson, i "California: Enigma of National Politics," Western Politics. j ed. Frank H. Jonas (Salt Lake City: University of Utah i Press, 1961), pp. 69-112. j 8 Sylvia Fava, "Suburbanism as a Way of Life," Ameri-i can Sociological Review. XXXI, 1 (February, 1956), pp. 34-37. George A. Lundberg, Mirra Komarovsky, and Mary Alice Mclnerny, Leisure: A Suburban Study (New York: Columbia University tress, 1 9,34) . suburbia with the Republican voter, employed four factors: (1) high percentage of home ownership, (2) high percentage of dwelling units recently constructed, (3) high rate of j ! recent population increase, and (4) high percentage of popu- 9 lation in family units. The use of some such combination of factors appears to be fruitful in examining suburbia*s political behavior for two reasons: (1) areas ranking high on one variable may not be similar in any other respect, and (2) there appears to be little political content in at least three of the variables. On the other hand, home ownership may be a measure of higher socioeconomic status and thus have a political connotation which might automatically define suburbia as Republican. The problem is, of course, that the concept of suburbia loses its meaning if such fac- ! tors as home ownership are not included. As a result, four variables were initially selected by which suburbs in Los Angeles County were to be defined. These were: 1. Percentage of owner-occupied dwelling units; 2. Percentage of dwelling units recently builtj 3. Percentage of housing units recently moved into; and 4. Percentage of single unit structures. Data for each of these variables were available by census tract on IBM tape.^ Each of the 1,297 census [ j i _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ _ j ^Supra. p. 6. ^Dwaine Marvick, "Los Angeles Political Statistics,"! a tape recorded dictionary containing population and housing! data for 1960 by census tract to which has been added elec tion and registration data for 1958, 1960, and 1962. Grate ful acknowledgment is made to Dr. Marvick for use of the tape. tracts in Los Angeles County was examined for each of the four variables* It was hypothesized that those tracts which were in the upper decile for each of the four variables could be defined for the purposes of this work as suburban areas. Unfortunately, no census tract survived all four criteria. In fact, there appeared to be a negative relationship between one of the variables, housing units moved into, 1958-1960* and two of the other variables, dwelling units owner—occupied and percentage of single unit structures. On the other hand, there appeared to be positive relationships between the rest of the variables. A preliminary analysis of relationships between the four 11 definitive variables is presented in Table 32. A later analysis, employing Pearsonian product moment correlations, substantiated the preliminary analysis 1 2 (Table 33). As a result, the definition of suburbia employed in this dissertation was those census tracts in the upper decile for dwelling units owner-occupied, single unit structures * and dwelling units recently (1950-1960) built. Of the 1,297 census tracts in Los Angeles County, 50 were in 11 All census tracts in Los Angeles County were ranked by each of the four variables. The highest and lowest decile were then sorted out. 1 2 The Pearsonian was calculated by use of a program written by the author for use on an IBM 1620-11. TABLE 32 PRELIMINARY ANALYSIS OF FOUR .VARIABLES TO BE USED TO DEFINE THE SUBURBAN AREAS OF LOS ANGELES COUNTY Variables A B--1 C E--2 N High* Low High Low High Low High Low A Dwelling units owner occupied High Low - 58.2 1.6 2.1 41.1 60.5 5.2 0.0 51.6 5.6 44.8 39.6 5.9 144 136 B-1 Housing units recently built High Low 58.2 2.1 1.6 41.1 21.8 16.3 0.7 21.6 14.7 17.0 19.4 17.7 129 141 C Single unit structures High Low 60.5 0.0 5.2 51 .6 21.8 0.7 16.3 21 .6 11.7 26.6 27.4 2.6 153 139 C M I w Housing units recently moved High Low into 5.6 39.6 44.8 5.9 14.7 19.4 17.0 17.7 11.7 27.4 26.6 ’ 2.6 132 127 N: 144 136 129 141 153 139 132 127 *Cell figures are the percentage of tracts falling into the appropriate cross categories. Per cents do not add up to 100 since some tracts, while being high or low for one variable, were neither for the other. Note, for example, the high mortality rate for tracts in the cross-categories for B-1 and E-2. Only a third survived both criteria. ________ _______ __ 182 : the upper decile for all three definitive variables and forty-five were in the lowest decile. The fifty high tracts were defined as suburban, while the lowest forty-five were j i defined as "anti-suburban." TABLE 33 CORRELATIONS BETWEEN THE FOUR PRELIMINARY DEFINING VARIABLES* Variable B-1 C E-2 A Dwelling units owner-occupied .553t .846+ -.616+ B-1 Housing units built 1950- 1960 ,402t .054 C Single unit structures -.848f E-2 Housing units moved into, 1958-1960 i *Based on a random sample of Los Angeles County Census tractsj N = 320. * Statistically significant at p. < .05 by the critical value of t. See, William L, Hays, Statistics for Psychologists (New York: Holt, Rinehart, and Winston, 1963), p. 529. In addition to the four variables discussed above, eleven other variables were extracted from the tape on file I at the Western Data Processing Center, and, in addition, one variable was generated at a later date, making a total of I sixteen. The following is a complete listing of the variables analyzed:. 183 A Percentage of Dwelling Units Owner-Occupied; B-1 Percentage of Housing Units Built 1950-1960; B-2 Percentage of Housing Units Built 1940-1950; and generated from B-1 and B-2, B-3 Percentage of Dwelling Units Built before 1940; C Percentage of Single Unit Structures; D Median Value of Property; B-1 Percentage Living in Same House As in 1955; E-2 Percentage of Housing Units Moved into, 1958-1960; E-3 Percentage Living Elsewhere in Los Angeles County in 1955; E-4 Percentage Living Elsewhere in California in 1955; F-1 Percentage of Families Having an Income from $3,000 to $10,000 per Tear; F-2 Percentage of Families Having an Income of More than $10,000 per Tear; G-1 Percentage of Male Working Force in the Professional Category; G-2 Percentage of Male Working Force in the Managerial Category; H-1 Percentage of Population White; H-2 Percentage of Population Negro. Fourteen variables were ranked by each of the definitive variables. Median values for each dependent 184 variable were then located for each of three groups (sub- 1 3 urbia, anti-suburbia, and both combined). Thus, in Table-34, the definitive (independent) variable is j dwelling units owner occupied— all other variables are defined as dependent. The median value for the percentage of housing units built 1950-1960 in the high decile tracts was eighty-nine, the median value in the low decile tracts was eleven, while the median value for the high and low tracts combined was forty-two. Applying the chi square test produced a value of 219.9 for one degree of freedom, which means that the difference in distributions around the two medians— 89 per cent and 11 per cent— was statistically significant at the .001 level. The same process was followed for each of the fourteen variables in tables 34, 35, and 36. The Three Definitive Variables An extensive examination of the two groups of census tracts established by the three definitive variables indicates that there is a striking difference between suburbia and "anti-suburbia." Each of the three definitive variables relates to the other fourteen variables in such a j way as to produce, almost without exception, statistically - i significant differences. t 13 * The median rather than the mean was used since the values involved were based on different sized populations. 185 Dvelling units owner—occupied A ranking of the census tracts in Los Angeles County by dvelling units ovner-occupied produced an upper decile of 144 tracts ranging between 90 and 100 per cent 14 owner occupancy* The lowest decile of 136 tracts ranged from zero to 19 per cent owner occupancy. Examination of the two groups of tracts showed a statistically significant difference in the distributions around the medians for twelve of the variables.^ For those tracts in the upper decile of dwelling units owner-occupied, compared to the lowest decile, it appears that housing is newer, there are more single unit structures, there are no Negroes, the population is less mobile, there is a larger proportion of families in the median and high family income brackets, and the percentage of males occupied in managerial positions is higher. On the other hand, there was no statistically significant difference in the value of property or the percentage of males employed in the professional fields. Perhaps the single most striking result of this first analysis is that suburbia, as defined by percentage 14 Of the 1,297 tracts in Los Angeles County, 11 had no data for this variable. Vhile 10 per cent of 1,297 is 130, the distribution of values was such that a cut-off point at 144 was used for the upper decile and 136 for the lower decile. Similar situations developed for the distributions of the other variables. 1 5 The test for statistical significance was taken from Hays, op. cit.. pp. 620-23 and is, essentially, a chi square test. 186 TABLE 34 MEDIAN VALUES FOR FOURTEEN SELECTED VARIABLES BY THE HIGH AND LOV DECILE OF DWELLING UNITS OCCUPIED (A) triable Medians for Both Groups Medians for High Decile Medians for Low Decile Chi Square df Stat Sig N=280 N=144 N=1 36 B-1 ,42?S .8995 .1196 219.9 1 yes B-2 .08 .11 .06 10.08 2 yes C .95 1.00 .19 216.9 2 yes D $15838 $16235 $15536 1 .38 1 no E—1 .3495 .4795 .3195 50.62 2 yes E-2 .39 .30 .54 179.1 2 yes E-3 .41 .38 .42 9.47 2 yes E-4 .03 .03 .04 11 .77 2 yes F-1 .65 .69 .62 27.38 2 yes F-2 .18 .25 .10 77.19 2 yes G-1 .15 .17 .12 3.39 2 no G-2 .11 .13 .09 10.87 2 yes H-1 .99 .99 .95 96.53 2 yes H-2 .00 .00 .01 74.48 2 yes 187 i of dwelling units owner-occupied, is less mobile than anti suburbia. Secondly, there was no statistically significant difference in the median value of the housing units. Thirdly, owner-occupants appear to be more often of the j managerial class but there appeared to be no difference j between the two groups as to professional occupations. Dwelling units recently built Again the census tracts were ranked, this time by percentage of dwelling units built since 1950. One hundred twenty—nine tracts fell into the upper decile, with a percentage range from 100 to 90 per cent of new housing. One hundred forty—one tracts fell into the lowest decile, with a range of zero to 6 per cent of new housing. Comparing these two groups (see Table 35), we see that for the high decile group there is a statistically significant higher median value for percentage of dwelling units owner-occupied, percentage of single unit structures, property values, percentage of population white, residential mobility (residence elsewhere in Los Angeles and elsewhere , i in California in 1955), high family income, and percentage J of males in the professional and managerial occupations. Perhaps the most striking result of this analysis is that while there was a negative relationship between age j of housing and its value, there appeared to be no relation- ; ship with the percentage of families in the median income 188 TABLE 35 MEDIAN VALUES FOR FOURTEEN SELECTED VARIABLES BT THE LOV AND HIGH DECILE OF HOUSING UNITS RECENTLI BUILT (B-1) triable Medians’ for Both Groups Medians for High Decile Medians for Low Decile Chi Square df Stat Sig N=270 N=1 29 N=141 A .70% .90% .27% 204.1 1 yes B-2 .04 .03 .06 19.1 2 yes C .89 1.00 .60 128.4 1 yes D $14197 $16336 $12652 43.5 1 yes E-1 .38% .33% .41% 8.18 1 yes E-2 .37 .35 .39 5.72 2 no E-3 .41 .45 .39 17.57 2 yes E-4 .03 .04 .02 27.35 2 yes F-1 .67 168 .65 2.22 2 no F-2 .16 .27 .08 114.4 2 yes G-1 .12 .19 .07 112.5 1 yes G-2 .18 .13 .06 50.89 2 yes H-1 .98 .99 .82 189.4 2 yes H-2 .00 .00 .09 167.9 1 yes 189 ; category (but there was a positive relation to the percentage of families in the high income category). Thus, areas with new housing are areas with more expensive j i housing, there is a greater likelihood that the occupants ■ I will own the dwelling, there are fewer multiple dwelling units (these are built in the older areas), very few Negroes, and a higher population mobility. This last fact illustrates the difficulty in constructing the suburban typology. Owner-occupancy was negatively related to mobility, new housing was positively related to mobility. The question may well be raised: What kind of mobility? Mobility in and out, or only mobility into the new housing but not out of it? Earlier analysis (Table 33) of the \ Los Angeles County sample produced the significant negative correlation betweeq mobility and owner occupancy of -.616, while later we will see that there is no relationship 1 between the two variables within the suburban typology. Single unit structures Once more, the census tracts were ranked, this time j by the percentage of single unit structures. One hundred ! fifty-three tracts had 100 per cent single unit structures; ! i these were defined as upper decile. One hundred thirty-ninei tracts were in the zero to 33 per cent range; these were j i defined as the lowest decile. 16 See Table 38 infra for this correlation. 190 TABLE 36 MEDIAN VALUES FOR FOURTEEN SELECTED VARIABLES BT THE HIGH AND LOW DECILE OF SINGLE UNIT STRUCTURES (C) triable Medians for Both Groups Medians for High Decile Medians for Low Decile Chi Square df Stat Sig N=292 N=1 53 N=1 39 A .49# .9196 .11# 240.4 1 yes B-1 .48 .89 .15 145.5 1 yes B-2 .10 .11 .08 6.17 2 yes D $16344 $15750 $17188 3.29 1 no E-1 .33 .37 .31 14.31 2 yes E-2 .48 .33 .53 125.0 2 yes E-3 .41 .43 .41 18.07 2 yes E-4 .04 .04 .04 1 .75 2 no F-1 .63 .68 .60 21 .86 2 yes F-2 .20 .25 .13 46.1 2 yes G-1 .16 .17 .15 1 .89 2 no G-2 .12 .13 .11 6.26 2 yes H-1 .99 .99 .98 54.16 2 yes H-2 .00 .00 .00 43.5 2 yes 191 i i l Compared to the low decile single unit structure tracts, the high decile tracts were statistically significantly higher in owner-occupancy, new housing, i white population, median family income, high family ! income, and managerial occupations. They were statisti cally significantly lower in population mobility. There was no difference in the median value of the property or professional occupations. The suburban typology Sorting by all three definitive variables markedly reduced the number of tracts in the "suburbia" category. Fifty tracts survived the upper decile for all | three criteria; forty-five survived for the lower decile. j After the sorting routine, the same procedure was followed as in the three previous tables. The following table gives the median values for each of the twelve variables as they relate to the Suburban and Anti-suburban typologies. Suburbia, compared to anti-suburbia, has more new houses, and once people have moved into them they do not move out as frequently as do the residents of anti suburbia, the property is worth more, both the median ! and high family income groups are more in evidence, and ! there are more people whose work is in the professional or managerial fields. TABLE 37 MEDIAN VALUES FOR TWELVE SELECTED VARIABLES BY THE SUBURBAN AND ANTI-SUBURBAN TYPOLOGY Medians Variable for Both Groups Medians for Suburbs Medians for Anti- Suburbs Chi Square Stat. Sig. N=95 N=50 N=45 B-2 . 02 % .03 # CM O • 2.902 2 no D $14130 $15387 $11771 8.38 1 yes E-1 .31# .37# .30 # 1.599 1 no E-2 .41 .33 .55 43.76 1 yes E-3 .42 .43 .42 10.132 2 yes E-4 .04 .03 .04 8.87 2 yes F-1 .64 .71 .61 13.25 1 yes F-2 .13 .23 .05 54.70 1 yes G-1 .11 .17 .08 28.65 1 yes G-2 .09 .11 .06 24.39 1 yes H—1 .99 .99 .92 60.9 1 yes H-2 .00 .00 .05 35.4 2 yes 193 The Relationship Between Variables The previous examination of the variables, and of the suburban and anti-suburban typologies, while presenting a fairly good picture of what suburbia is and is not, did j i j not describe the interrelationship between variables, nor could it have revealed if those relationships are a function of the communities involved. One suspects that some of the relationships may be a product of the communities— the intrinsic differences between suburbia and the rest of the county. The county sample For the purposes of examining the relationship between variables for Los Angeles Coimty as a whole, a random sample of 320 tracts was taken from the total of 1,297 tracts. The following tables (tables 38 through 53), present the Fearsonian correlations between each of the variables for the three communities— the county, suburbia, and anti-suburbia. As before, the tables have been arranged so that the same type of phenomena are grouped j i together. Summary for the County I Owner-occupancv Owner-occupancy was positively related to the incidence of single unit structures and unrelated to the TABLE 38 194 CORRELATION BETVEEN OWNER OCCUPANCY (A) AND FIFTEEN SELECTED VARIABLES BY THREE COMMUNITY TYPOLOGIES VARIABLES . SAMPLE8 , r. SUBURBb r. ANTI SUBURB' r. B Age of housing Built, 1950-1960 .553* . 668* .398* Built, 1940-1950 .176* -.219 -.137 Built before 1940 (old) -.632* -.576* .106 C Single unit structures .846* t .477* D Value of property .072 .055 .078 E Mobility Moved into house, 1958-1960 -.616* -.049 -.390* Residence same as in 1955 .296* -.168 .354* Residence elsewhere in 1955 Los Angeles -.069 .290* -.090* California -.097 -.187 -.343* F Income Family income $3,000-$10,000 .031 .1 33 .213 Family income $10,000+ .371* -.075 .350* 0 Occupation Professional .112* -.134 .367* Managerial .182* -.049 .104 H Race White .207* -.258 -.090 Negro -.181* t .036 N: 320 50 36d *Statistically significant at p. < .05. Correlation must be at least .110 by t test to be statistically significant at p. < .05. j ^Correlation must be at least .281 by t test to be ! statistically significant at p. < .05. Correlation must be at least ,330 by t test to be statistically significant at p. < .05. i This is less than the original 45 due to incomplete! data fields. No correlation figure since there are no Negroes in suburbia; and suburbia is by definition 100 per cent single unit structures. TABLE 39 195 CORRELATION BETWEEN HOUSING BUILT 1950-1960 (B-1) AND FIFTEEN SELECTED VARIABLES BY THREE COMMUNITY TYPOLOGIES VARIABLES SAMPLE® r. ' SUBURBb r. ANTI- SUBURB1 r. A Owner occupancy .553* .668* .398* B Age of housing Built, 1940-1950 -.203* -.537* .095 Built before 1940 (old) -.848* -.683* -.169 C Single unit structures .402* t .234 D Value of property .162* -.373* .344* E Mobility Moved into house, 1958-1960 .054 -.044 -.045 Residence same as in 1955 -.332* -.269 .060 Residence elsewhere in 1955, Los Angeles .305* .253 -.239 California .252* .059 -.177 F Income Family income S3,000-S10,000 -.004 .557* .344* Family income $10,000+ .318* -.496* .490* G Occupation Professional .269* -.443* .453* Managerial .193* -.443* .391* H Race White .344* -.314* .157 Negro -.312* t -.139 N; 320 50 36d *Statistically significant at p. < .05. Correlation must be at least .110 by t test to be statistically significant at p. < .05. Correlation must be at least .281 by t test to be statistically significant at p. < .05. Correlation must be at least .330 by t test to be statistically significant at p. < .05. This is less than the original 45 due to incomplete data fields. No correlation figure since there are no Negroes in suburbia; and suburbia is by definition 100 per cent single unit structures. TABLE 40 196 CORRELATION BETWEEN HOUSING BUILT 1940-1950 (B-2) AND FIFTEEN SELECTED VARIABLES BY THREE COMMUNITY TYPOLOGIES VARIABLES SAMPLE8 , r. SUBURBb r. ANTI SUBURB0 r. A Owner occupancy .176* -.219 -.137 B Age of housing Built, 1950-1960 -.203* -.537* .095 Built before 1940 (old) -.340* -.250 -.997* C Single unit structure .203* t .234 D Value of property -.070 .034 -.173 E Mobility Moved into house, 1958-1960 -.21£* .085 .534* Residence same as in 1955 .387* .102 -.482* Residence elsewhere in 1955, Los Angeles -.337* -.052 -.568* California -.102 .090 .549* F Income Family income $3,000-$10,000 .062 -.094 -.015 Family income $10,000+ -.001 .151 -.281 G Occupation Professional -.129* -.112 -.117 Managerial -.049 -.045 -.229 H Race White .042 -.061 -.197 Negro -.037 t .277 N: 320 50 36d *Statistically significant at p. < .05. £ L Correlation must be at least .110 ty t test to be statistically significant at p. < .05. ^Correlation must be at least .281 by t test to be statistically significant at p. < ,05. Correlation must be at least .330 by t test to be statistically significant at p. < .05. This is less than the original 45 due to incomplete data fields. +No correlation figure since there are no Negroes in suburbia; and suburbia is by definition 100 per cent single unit structures. TABLE 41 197 CORRELATION BETWEEN HOUSING BUILT BEFORE 1940 (B-3) AND FIFTEEN SELECTED VARIABLES BY THREE COMMUNITY TYPOLOGIES VARIABLES SAMPLE* r. SUBURB2 r. ANTI SUBURB r. A Owner occupflLncy -.632* -.576* .106 B Age of housing Built, 1950-1960 -.848* -.683* -.169 Built, 1940-1950 -.340* -.250 -.997* C Single unit structure -.482* t -.248 D Value of property -.132* .398* .145 E Mobility Moved into house, 1958-1960 -.482* -.023 -.248 Residence same as in 1955 .118* .220 .473* Residence elsewhere in 1955, Los Angeles -.113* -.245 .580* California -.188* -.145 -.530* F Income Family income $3,00-$10,000 -.018 -.550* -.011 Family income $10,000+ -.310 .530* .241 G Occupation Professional -.202* • 605* .082 Managerial -.161* .547* .197 H Race White -.357* .413* .183 Negro .323* t -.264 N: 320 50 36d ♦Statistically significant at p. < .05. a Correlation must be at least .110 by t test to be statistically significant at p. < .05.. Correlation must be at least .281 by t test to be statistically significant at p. <r .05. Correlation must be at least .330 by t test to be statistically significant at p. < .05. cL This is less than the original 45 due to incomplete data fields. ^No correlation figure since there are no Negroes in suburbia; and suburbia is by definition 100 per cent single structures. TABLE 42 198 CORRELATION BETWEEN SINGLE UNIT STRUCTURES (C) AND FIFTEEN SELECTED VARIABLES BY THREE COMMUNITY TYPOLOGIES SAMPLE* SUBURBb r. r. r. A Owner occupancy .846* t .477* B Age of housing Built, 1950-1960 .402* t .234 Built, 1940-1950 .203* f .233 Built before 1940 (old) -.482* t -.248 D Value of property -.178* t -.081 E Mobility Moved into house, 1958-1960 -.484* + -.032 Residence same as in 1955 .243 t -.026 Residence elsewhere in 1955, Los Angeles -.025 t -.316 California -.034 + .094 F Income Family income $3,000-$10,000 .099 t .361* Family income $10,000+ .123* + -.011 G Occupation Professional -.106 t -.005 Managerial -.037 t -.356* H Race White .011 t .257 Negro .023 t .177 N: 320 50 36d -^Statistically significant at p. < .05. Correlation must be at least .110 by t test to be statistically significant at p. < .05. ^Correlation must be at least .281 by t test to be statistically significant at p. < .05. Correlation must be at least .330 by t test to be statistically significant at p. < .05. dThis is less than the original 45 due to incomplete data fields. A No correlation figures since there are no Negroes in suburbia; and suburbia is by definition 100 per cent single unit structures. TABLE 43 199 CORRELATION BETWEEN MEDIAN VALUE OF PROPERTY (D) AND FIFTEEN SELECTED VARIABLES BY THREE COMMUNITY TYPOLOGIES VARIABLES SAMPLE8 , r. SUBURB^ r. ANTI SUBURB r. A Owner occupancy .072 .055 .078 B Age of housing Built, 1950-1960 • 1 62* -.373* .344* Built, 1940-1950 -.070 .034 -.173 Built before 1940 (old) -.132* .398* .145 C Single unit structure -.178* + -.081 E Mobility Moved into house, 1958-1960 -.094 .047 -.151 Residence same as in 1955 -.032 .111 .034 Residence elsewhere in 1955 Los Angeles .070 .264 .175 California .044 -.294* -.241 F Income Family income $3,000-$10,000 -.804* -.913* .246 Family income $10,000+ .863* .931* .590* G Occupation Professional .772* .879* .432* Managerial .806* .856* .188 H Race White .288* .233 .238 Negro -.275* t -.105 N: 320 50 36d *Statistically significant at p. < .05. Correlation must Be at least .110 by t test to be statistically significant at p. < .05. Correlation must be at least .281 by t test to be statistically significant at p. < .05. Correlation must be at least .330 by t test to be statistically significant at p. < .05. dThis is less than the original 45 due to incomplete data fields. A No correlation figure since there are no Negroes in suburbia; and suburbia is by definition 100 per cent single structures. TABLE 44 200 CORRELATION BETWEEN RESIDENCY SAME AS IN 1955 (E-1) AND FIFTEEN SELECTED VARIABLES BY THREE COMMUNITY TYPOLOGIES VARIABLES SAMPLE0 , r. SUBUKBb r. ANTI SUBURB r. A Owner occupancy .296* -.168 .354* B Age of housing Built, 1950-1960 -.332* -.269 .060 Built, 1940-1950 .387* .102 -.482* Built before 1940 (old) .118* .220 .473* C Single unit structure .243* t -.026 D Value of property - .032 -.111 .034 E Mobility Moved into house, 1958-1960 -.774* -.669* -.818* ,Residence elsewhere in 1955 Los Angeles -.795* -.941* .034 California -.535* -.528* -.760* F Income Family income $3,000-$10,000 -.058 -.124 -.124 Family income $10,000+ -.091 .095 .127 G Occupation Professional -.096 .028 -.059 Managerial .079 .028 .337* H Race White -.086 .188 -.249 Negro .091 t .129 N: 320 50 45d ^Statistically significant at p. < .05. Correlation must be at least .110 by t test to be statistically significant at p. < .05. ^Correlation must be at least .281 by t test to be statistically significant at p. < .05. Correlation must be at least .330 by t test to be statistically significant at p. < ,05. dThis is less than the original 45 due to incomplete data fields. ^No correlation figure since there are no Negroes in suburbia; and suburbia is by definition 100 per cent single unit structures. TABLE 45 201 CORRELATION BETWEEN MOVED INTO HOUSE, 1958-1960 (E-2) AND FIFTEEN SELECTED VARIABLES BY THREE COMMUNITY TYPOLOGIES VARIABLES SAMPLE8 , r. SUBURBb r. ANTI SUBURB0 r. A Owner occupancy -.616* -.049 -.390* B Age of housing Built, 1950-1960 .054 -.044 -.045 Built, 1940-1950 -.212* .085 .534* Built before 1940 (old) .053 -.023 -.525* C Single unit structure -.484* t -.032 D Value of property -.094 .047 -.151 E Mobility Residence same as in 1955 -.774* —.669* -.818* Residence elsewhere in 1955 Los Angeles .472* .542* -.251 California .430* .358* . 661 * F Income Family income $3,000-$10,000 .154* .067 -.050 Family income $10,000+ -.282* -.066 -.309 G Occupation Professional -.039 -.008 -.110 Managerial -.220* -.030 -.333* H Race White .079 00 00 . 1 .163 Negro -.096 t -.068 N: 320 50 36d ♦Statistically significant at p. < .05. Correlation must be at least .110 by t test to be statistically significant at p. < .05. ^Correlation must be at least .281 by t test to be statistically significant at p. < .05. Correlation must be at least .330 by t test to be statistically significant at p. < .05. dThis is less than the original 45 due to incomplete data fields. ^No correlation figures since there are no Negroes in suburbia; and suburbia is by definition 100 per cent single structures. TABLE 46 202 CORRELATION BETWEEN RESIDENCE ELSEWHERE IN 1955, LOS ANGELES (E-3) AND FIFTEEN SELECTED VARIABLES BY THREE COMMUNITY TYPOLOGIES VARIABLES SAMPLE8, r. SUBURBb r. ANTI SUBURB' r. A Owner occupancy -.069 .290* -.095 B Age of housing Built, 1950-1960 .305* .253 -.239 Built, 1940-1950 -.337* -.052 -.568* Built before 1940 (old) -.113* -.245 .580* C Single unit structure -.025 t -.316 D Value of property .070 .264 .175 E Mobility Moved into house 1958-1960 • S ' .472* .542* -.251 Residence same as in 1955 -.795* .941* .034 Residence elsewhere in 1955 California .170* .357* -.394* F Income Family income $3,000-$10,000 -.019 -.016 -.057 Family income $10,000+ .008 .054 -.012 G Occupation Professional .012 .077 .115 Managerial .024 .116 .017 H Race White .001 -.173 .031 Negro .012 t .050 N: 320 50 36d ^Statistically significant at p. < .05. O Correlation must be at least .110 by t test to be statistically significant at p. < .05. ^Correlation must be at least .281 by t test to be statistically significant at p. < .05. c Correlation must be at least .330 by t test to be statistically significant at p. < .05. dThis is less than the original 45 due to incomplete data fields. ^No correlation figure since there are no Negroes in suburbia: and suburbia is by definition 100 per cent single unit structures. TABLE 47 203 CORRELATION BETWEEN RESIDENCE ELSEWHERE IN 1955, CALIFORNIA (E-4) AND FIFTEEN SELECTED VARIABLES BT THREE COMMUNITY TYPOLOGIES VARIABLES SAMPLE0 , r. SUBURBb r. ANTI SUBURB0 r. A Owner occupancy -.097 -.187 -.343* B Age of housing Built, 1950-1960 Built, 1940-1950 Built before 1940 (old) .252* -.102 -.188* .059 .090 -.145 -.177 .549* -.530* C Single unit structure -.034 + .094 D Value of property .040 -.294* -.241 E Mobility Moved into house, 1958-1960 Residence same as in 1955 Los Angeles .430* -.535* .170* .358* -.528* .357* .661* -.760* -.394* F Income Family income $3,000-$10,000 Family income $10,000+ -.049 .040 .330* -.349* .061 -.300 G Occupation Professional Managerial .194* .094 -.295* -.247 -.119 -.229 H Race White Negro .237* -.239* -.029 t .053 -.034 N: 320 50 36d *Statistically significant at p. < .05. £ L Correlation must be at least .110 by t test to be statistically significant at p. < .05. b ^ Correlation must be at least .281 by t test to be statistically significant at p. < .05. Correlation must be at least .330 by t test to be statistically significant at p. < .05. dThis is less than the original 45 due to incomplete data fields. ^No correlation figure since there are no Negroes in suburbia; and suburbia is by definition 100 per cent single unit structures. TABLE 48 204 CORRELATION BETWEEN FAMILY INCOME, $3,000-$10,000 (F-1) AND FIFTEEN SELECTED VARIABLES BY THREE COMMUNITY TYPOLOGIES VARIABLES SAMPLE8 , r. SUBURBb r. ANTI SUBURB r. A Owner occupancy .031 .133 .213 B Age of housing Built, 1950-1960 -.004 .551* .344* Built, 1940-1950 .062 -.094 -.015 Built before 1940 (old) -.018 -.550* -.011 C Single unit structure .099 t . 361# D Value of property -.804* -.913* .246 E Mobility Moved into house 1958-1960 • 1 54* .067 -.050 Residence same as in 1955 -.058 -.124 -.124 Residence elsewhere in 1955 Los Angeles -.019 -.016 -.057 California -.049 .330* .061 F Income Family income $10,000+ -.837* -.992* .344* G Occupation Professional -.675* -.898* .324 Managerial -.787* -.893* -.002 H Race White -.101 -.375* .276 Negro .079 T -.395* N: 320 50 36d *Statistically significant at p. < .05. o Correlation must be at statistically significant at p. least .110 by t test < .05. to be ^Correlation must be at statistically significant at p. least .281 < .05. by t test to be Correlation must be at least .330 by t test to be statistically significant at p. < .05. dThis is less than the original 45 due to incomplete data fields. +No correlation figure since there are no Negroes in suburbia; and suburbia is by definition 100 per cent single unit structures. TABLE 49 205 CORRELATION BETWEEN FAMILY INCOME, $10,000 PLUS (F-2) AND FIFTEEN SELECTED VARIABLES BY THREE COMMUNITY TYPOLOGIES VARIABLES SAMPLE8 , r. SUBURBb r. ANTI SUBURB r. A Owner occupancy .371* -.075 .350* B Age of housing Built, 1950-1960 .318* -.496* .490* Built, 1940-1950 -.001 .151 -.281 Built before 1940 (old) -.310* .530* .241 C Single unit structure .123* t -.011 D Value of property .863* .931* .590* E Mobility Moved into house 1958—1960 -.282* — .066 -.309 Residence same as in 1955 .091 .095 .127 Residence elsewhere in 1955 Los Angeles .008 .054 -.012 California .040 -.349* -.300 F Income Family income $3,000-$10,000 -.837* -.992* .344* G Occupation Professional .805* .908* .524* Managerial .847* .900* .262 H Race White .369* .367* .338* Negro -.342* t -.324 Ns 320 50 36d ♦Statistically significant at p. < .05. Correlation must be at least .110 by t test to be statistically significant at p. < .05. Correlation must be at least .281 by t test to be statistically significant at p. < .05. Correlation must be at least .330 by t test to be statistically significant at p. < .05. dThis is less than the original 45 due to incomplete data fields. No correlation figure since there are no Negroes in suburbia; and suburbia is by definition 100 per cent single unit structures. TABLE 50 206 CORRELATION BETWEEN OCCUPATION, PROFESSIONAL (G-1) AND FIFTEEN SELECTED VARIABLES BY THREE COMMUNITY TYPOLOGIES VARIABLES SAMPLE8 , r. SUBURBb r. ANTI SUBURB r. A Owner occupancy .112* -.134 .367* B Age of housing Built, 1950-1960 .269* -.443* .453* Built, 1940-1950 -.129* -.112 -.117 Built before 1940 (old) -. 202* .605* .082 C Single unit structures -.106 + -.005 D Value of property . .772* .879* .432* E Mobility Moved into house 1958-1960 -.039 -.008 -.110 Residence same as in 1955 -.096 -.028 -.059 Residence elsewhere in 1955 Los Angeles .01 2 .077 .115 California .194* -.295* -.119 F Income Family income $3,000—$10,000 -.675* -.898* .324 Family income $10,000+ .805* .908* .524* G Occupation Managerial .661* .827* .348* H Race White .354* .363* .452* Negro -.338* t -.418* N: 320 50 36d *Statistically significant at p. < .050. Correlation must be at least .110 by t test to be statistically significant at p. < .05. ^Correlation must be at least .281 by t test to be statistically significant at p. < .05. Correlation must be at least .330 by t test to be statistically significant at p. < .05. dThis is less than the original 45 due to incomplete data fields. A No correlation figure since there are no Negroes in suburbia; and suburbia is by definition 100 per cent single unit structures. TABLE 51 207 CORRELATION BETWEEN OCCUPATION, MANAGERIAL (G-2) AND FIFTEEN SELECTED VARIABLES BY THREE COMMUNITY TYPOLOGIES VARIABLES SAMPLE0 , r. SUBURBb r. ANTI SUBURB r. A Owner occupancy .182* -.049 .104 B Age of housing Built, 1950-1960 .193* -.443* .391* Built, 1940-1950 -.049 -.045 -.229 Built before 1940 (old) -.161* .547* .197 C Single unit structure .037 t -.356* D Value of property .806* .856* .188 E Mobility Moved into house 1958-1960 -.220* -.030 -.333* Residence same as in 1955 .079 .028 .337* Residence elsewhere in 1955 Los Angeles .024 .116 .017 California .094 -.247 -.229 F Income Family income $3,000-$10,000 -.787* -.893* -.002 Family income $10,000+ .847* .900* .262 G Occupation Professional .661* .827* .348* H Race White .342* .277 .287 Negro -.329* t .316 N: 320 50 36d *Statistically significant at p. < ,05. a Correlation must be at least .110 by t test to be statistically significant at p. < .05. ^Correlation must be at least .281 by t test to be statistically significant at p. < .05. Correlation must be at least ,330 by t test to be statistically significant at p. < .05. dThis is less than the original 45 due to incomplete data fields. ^No correlation figure since there are no Negroes in suburbia; and suburbia is by definition 100 per cent single unit structures. TABLE 52 208 CORRELATION BETWEEN RACE, WHITE (H-1) AND FIFTEEN SELECTED VARIABLES BY THREE COMMUNITY TYPOLOGIES VARIABLES SAMPLE* r. SUBURBb r. ANTI SUBURB r. A Owner occupancy .207* -.2513 .090 B Age of housing Built, 1950-1960 .344* -.314* .157 Built, 1940-1950 .042 -.061 -.197 Built before 1940 (old) .357* .413* .183 C Single unit structure .011 t -.257 D Value of property .288* .233 .238 E Mobility Moved into house, 1958-1960 .079 -.188 .163 Residence same as in 1955 -.086 .188 -.249 Residence elsewhere in 1955 Los Angeles .001 -.173 .031 California .235* -.029 .053 F Income Family income $3,000-$10,000 -.101 -.375* .276 Family income $10,000+ .369* .367* .338* G Occupation Professional .354* .363* .452* Managerial .342* .277 .287 H Race Negro -.987* t -.914* N: 320 50 36d *Statistically significant at p. < .05. G b Correlation must be at least .110 by t test to be statistically significant at p. < .05. ^Correlation must be at least .281 by t test to be statistically significant at p. < .05. Correlation must be at least .330 by t test to be statistically significant at p. < .05. dThis is less than the original 45 due to incomplete data fields. ^No correlation figure since there are no Negroes in suburbia; and suburbia is by definition 100 per cent single unit structures. TABLE 53 209 CORRELATION BETWEEN RACE, NEGRO (H-2) and FIFTEEN SELECTED VARIABLES BT THREE COMMUNITY TYPOLOGIES VARIABLES SAMPLE8 , r. SUBURB^ r. ANTI SUBURB r. A Owner occupancy .181* t .036 B Age of housing Built, 1950-1960 -.312* t -.139 Built, 1940-1950 -.037 t .277 Built before 1940 (old) .323* t —r264 C Single unit structure .023 t .177 D Value of property -.275* t -.105 E Mobility Moved into house, 1958-1960 -.096 t -.068 Residence same as in 1955 .091 t .129 Residence elsewhere in 1955 Los Angeles .012 + .050 California -.239* t -.034 F Income Family income $3,000-$10,000 .079 + -.395* Family income $10,000+ -.342* t -.324 G Occupation Professional -.338* + -.418* Managerial -.329* t -.316 H Race White -.987* + -.914* N: 320 50 36d *Statistically significant at p. < .05* a Correlation must be at least .110 by t test to be statistically significant at p. < .05. Correlation must be at least .281 by t test to be statistically significant at p. < .05. Correlation must be at least .330 by t test to be statistically significant at p. < .05. dThis is less than the original 45 due to incomplete data fields. +No correlation figure since there are no Negroes in suburbia; and suburbia is by definition 100 per cent single unit structures. 210 17 value of the property. It was negatively related to mobility, and positively related to high income, but was * unrelated to median income. Vhile owner-occupancy was positively related to both professional and managerial occupations, the relationship was so low as to be of little predictive value. Finally, owner-occupancy did relate to race, negatively to Negro and positively to white. Age of housing The younger the housing, the more likely it was to be single unit structures and worth more. The single unit residents were less mobile and had higher incomes, being more likely to be of the professional or managerial class and most likely to be white rather than Negro. Single unit structures The incidence of single unit structures related very positively with owner-occupancy and very negatively with the age of housing. There was also a low negative relationship with value of property and a strong negative relationship with mobility. There was no significant relationship to professional or managerial occupations, race, or high income, but there was a low positive- relationship to median income. 1 7 Where the statement is made that there is no relationship between two variables it means that while there was a value for the Fearsonian, it was not statisti cally significant; see footnotes at bottom of Table 38. 211 Summary for Suburbia Owner-occupancy The ^relationship of owner-occupancy to the other variables showed marked change in suburbia from that of the county as a whole. Unlike the rest of the county, owner- occupancy did not relate to mobility except for a low positive relationship to those who moved from some other place in Los Angeles. Owner-occupancy did not relate to income or occupation. There were no Negroes in suburbia. Like the county, there was a negative relationship to age of housing and no relationship to the value of the property. Age of housing The relationship between age of housing and the other variables also markedly changed in suburbia from what it was in the county as a whole. There was a positive relationship between value of property and age of housing in suburbia as well as with the incidence of high income— whereas both relationships were negative in the county. While there was no relationship between the incidence of median income and the age of housing for the county, there was a negative one in suburbia. Apparently middle income suburbanites buy new homes, high income suburbanites buy old homes. Both the professional and managerial class buy the older homes in suburbia— but there was no relationship for the county as a whole. On the other hand, there was 212 no relationship between mobility and age of housing in suburbia but there was a negative one in the county. Single unit structures * In suburbia all housing units are single unit struc tures; there are, therefore, no correlation figures for this category (just as there are none for the relationship for Negro population since there are no Negroes in suburbia). Summary for Anti-Suburbia Owner-occupancy The relationship between‘owner-occupancy and age of housing was negative, as it was for both the county and suburbia. Also, like the county and suburbia, there was no relationship between owner-occupancy and the value of the property, or the incidence of median income. Like the county, anti-suburbia showed a positive relationship between owner-occupancy and single unit structures, while the relationship to mobility was negative. Like the county, and unlike suburbia, there was a positive relationship between owner-occupancy and high income and also to profes sional occupations. Like suburbia and unlike the county, there was no relationship to the managerial occupations. Age of housing Unlike the county as a whole, there was no relation ship between age of housing and type of housing (single 213 unit structures). Like suburbia, there was no relationship between mobility and age of housing, although it was nega tive for the county. And, like•suburbia, the relationship between age of housing and the incidence of median income was negative while there was no relationship for the county. Like the county, and unlike suburbia, value of property, the incidence of high income, professional occupations and managerial occupations were all negatively related to the age of housing (in each case the relationship was positive in suburbia). Single unit structures The incidence of single unit structures related positively with owner-occupancy, as in the county. On the other hand, it related positively to high income and negatively to managerial occupations, while for the county there was no relationship to either. For the value of property and mobility there was no relationship, although there were negative ones for the county. For median incomes there was also no relationship but there was a positive one for the county. And, for professional occupa tions and single unit structures there was no relationship either for the anti-suburban area or the county as a whole. Table 54, which follows, sums up the patterns of relationships between the three definitive variables and all other variables for the three communities. TABLE 54 214 THE RELATIONSHIP BETWEEN VARIABLES BY THREE COMMUNITY TYPOLOGIES Variable Sample Suburb Anti- Suburb Por Owner Occupancy (A) Age of housing -* Single unit structures +* #t + . Value of property 0 - Mobility of residents - 0 - High income + + Median income Professional + Managerial + For Age of Housing {B—1, -2, and -3) Single unit structures - # Value of property - + - Mobility of residents - $ High income - + - Median income - - Professional - + - Managerial - + - For Single Unit Value of property Structures (C) # Mobility of residents - # High income ** # + Median income + # Professional # 0 Managerial $ # - *The signs: , , , +,t, and "0" indicate negative positive, and no relationships respectively. tlndicates no Pearsonian for the co-variables. 215 There were, obviously, differences not only in the degree of the relationships hut in the sign of these rela tionships for each of the three communities: the county, suburbia, and anti-suburbia. These differences involved, moreover, other variables in addition to the three definitive variables. (See Table 55.) The Effects of Community Examination of the median values illustrated that suburbia, compared to anti-suburbia, was different in that more people lived in new housing, they had higher incomes, their property was worth more, they were more likely to be either professionals or managers, they were less likely to move, and there were no Negroes. If these differences were due only to the inter relationship between the variables— e.g., the coincidence of high income and owner-occupancy— there would be nothing unique about suburbia except a concentration of certain types of people and property. However, it appeared that the relationship between variables was different depending on whether one looked at the county as a whole or only suburbia. For example, while the incidence of high income and owner-occupancy was "normal" for the county as a whole, that is positive; for suburbia there was no such relationship. As was seen in the listing of correlations between variables by community, these relationships changed not TABLE 55 CHANGES IN SIGN FOR STATISTICALLY SIGNIFICANT CORRELATIONS* BY THREE COMMUNITY TYPOLOGIES Variable Sample Suburb Anti Subur' Housing built 1950-1960 x Value of property .162 -.373 .344 Housing built 1950-1960 x Income $10,000+ .318 -.496 .490 Housing built 1950-1960 x Professional .269 -.443 .453 Housing built 1950-1960 x Managerial .193 -.443 .391 Housing built 1940-1950 x Moved 1958-1960 -.212 .534 Housing built before 1940 x Residence elsewhere in Los Angeles in 1955 -.113 .580 Housing built before 1940 x Value of property -.132 .398 Housing built before 1940 x Income $10,000+ -.310 .530 Housing built before 1940 x Professional -.202 .605 Housing built before 1940 x Managerial -.161 .547 Residence elsewhere in Residence elsewhere in Los Angeles in 1955 California in 1955 .170 .357 -.394 Residence elsewhere in _ „ , n i * • inrr x Professional California m 1955 .194 -.295 Income $3,000--$10,000 x Income $10,000+ -.837 -.992 .344 *In each instance the correlations are statistically significant. 216 217 * only in degree but in sign, suggesting (1) either a funda mental change in the relationships between variables within the different communities or (2) that the communities them selves were intervening variables. Early in this chapter the point was made that owner-occupancy and high socio economic status were probably positively related and that this in turn related positively to voting Republican. But what will happen if owner-occupancy and socioeconomic status (a relative value) do not always co-relate? For example, as we have seen, income was not the reliable guide it may have appeared to be to owner-occupancy. Or, perhaps, there was some co-relation between residential mobility and voting behavior. But, as we saw, the relationship between mobility and the incidence of owner-occupancies changed markedly from the county as a whole, where it was negative, to suburbia, where there was no relationship. If, then, community alters relationships, we might expect political behavior within suburbia to exhibit relationships with variables which would be different from the relationship within the county as a whole. CHAPTER IV SUBURBAN POLITICAL BEHAVIOR Introduction Examination of suburban political behavior has been both extensive and, in some instances, intensive. The general impression which one gains from these studies is that the place theory has laid claim to more adherents,^ There is no doubt, of course, that the suburbs vote differ- 2 ently than do the central cities. The problem is that time and region alter the pattern. According to Edward Janosik, many suburbs prior to World War II were Democratic, while Bernard Lazerwitz has suggested that Southern suburbs 3 may be more Democratic than their central cities. 1 See, for example: Leon Epstein, "Size of Place and the Division of the Two-Party Vote in Wisconsin," Western Political Quarterly. IX, 1 (March, 1956), pp. 138-150. 2 For a sweeping survey, comparing the Democratic vote in central cities and suburbs for thirteen metropolitan areas, see: Appendix A, Table 118, this dissertation. 3 G. Edward Janosik, "The New Suburbia," Current History. XXXI, 180 (August, 1956), p. 92; and Bernard Lazerwitz, "Some Characteristics of Residential Belts in the Metropolitan Community" (unpublished Ph.D. dissertation, University of Michigan, 1958), see Appendix A, Table 122, this dissertation. 218 219 In addition to the time and regional factors, differences between suburban party preference and suburban vote must be considered. Scott Greer's data on St. Louis illustrated the point well. While suburban Catholics did not differ in their party preference from central city 4 Catholics, their voting behavior did differ markedly. This chapter examines suburban census tract registration and election data in two ways. First, the differences in Democratic Party registration, votes cast for various partisan and non-partisan candidates, and levels in voter turn-out are compared between the suburban and "anti-suburban” typologies. Secondly, the relationships, expressed as Pearsonian correlations, between the demo graphic and political data are presented and discussed. Finally, an analysis is made holding three of the key intervening variables constant. The Political Variables Registration Democratic registration percentages for each tract for September, 1958; September and December, 1960; and September and December, 1962» were examined. 4 Scott Greer, "Catholic Voters and the Democratic Party," Public Opinion Quarterly, XXV, 4 (Winter, 1962), pp. 611-25; see Appendix A, Tables 123 and 124, this dissertation. 220 Partisan voting data The Democratic candidate*s percentage of votes received for the two major party candidates, including Edmund G. Brown (Gubernatorial), 1958 and 1962; John F. Kennedy (Presidential), 1960; Richard Richards (United States Senate), 1962; were examined as indications of the performance of leading Democratic political figures. In addition, the votes received by Bert A. Betts for State Treasurer, 1958 and 1962; and Don Rose, Secretary of State, 1962, were examined since both men were unknown outside of Democratic party circles and were running against Republican incumbents; hence their vote was probably due in large measure to the party label on the November ballot. The vote cast for Thomas M. Rees, candidate for State Senate from Los Angeles County was included since (1) there was no incumbent, and (2) it was a local, but county-wide, race. Non-partisan elections Two elections were chosen of the type which are supposed to have special impact on suburban voters. The 1962 election for State Superintendent of Public Instruction featured a contest between Ralph Richardson, a "liberal" member of the Los Angeles School Board, and Max Rafferty, an avowed "three-R" educational conservative. The other non-partisan race was for Los Angeles County Tax Assessor. 221 While the Assessor has nothing to do with the tax rate, his office quite often becomes the target of irate property tax payers. The election in 1962 was between Phillip Watson, who appealed to the property taxpayer and Phil Gibson, who appeared largely to offer a continuation of past policies. Mr. Watson and Dr. Rafferty won their elections. Ballot propositions Devoid of both partisan label and personality, ballot propositions offer perhaps a more narrowly defined range of alternatives to the voter. Two such propositions were employed. The Subversive Controls Constitutional Amendment, No. 24, in 1962, better known as the Francis Amendment, would have set up loyalty review boards across the state empowered to investigate rather loosely-defined concepts of "loyalty" and to dismiss government employees for such "cause." The campaign raised again the whole issue of Communism, internal subversion, and loyalty. The Proposition was defeated. The other ballot measure, No. 1, in 1960 was the billion dollar—plus state-wide Water Bond Issue designed to bring water from over-supplied northern California to parched southern California. The vote for and against, however, was not only one of North vs. South, but of whether or not the State should assume an all-time record debt. The measure passed. 222 Participation Democratic and Republican turn-out in the elections of 1960 and 1962 are used to measure participation. The figure is derived by dividing the September Registration (those eligible to vote) by the December Registration (those who did vote). The Party Vote Use of election data to measure party strength presents several rather formidable problems. One way to avoid these problems is to add up the votes for all of the party*s candidates and divide by the number of candidates. Such a method produces an average figure but muddles rather 5 than clarifies the problems involved. Many of these problems are rooted in the fundamentally different meaning of the word "party" to different individuals involved. Edmund Burke*s definition - of a party being a body of men united in principle obviously does not apply. While the more functional approach of considering all those who participate in the party*s primary as members of the party seems to fall short at the other end. Heinz Eulau has offered a role-definition of party which possibly comes closest to the truth but offers little help. In his words, 5 See: Edward F, Cox, "The Measurement of Party Strength," Western Political Quarterly. XIII, 4 (December, 1960), p. 1028. 223 party is . .a system of mutually interdependent roles which range from that of the professional functionary to one of an occasional supporter." As Eulau pointed out, such a system is not only differentiated but does not even demand much cooperation or solidarity, particularly at the edges. While party leaders might not consider voting for anyone but a party member, it is quite obvious that a considerable porportion of the general electorate can and does cross party lines with ease. As Neil A. McDonald observed, "... for most people party identification is not the same thing as current intention to vote for one 7 party or another." On the other hand, both parties do have rather large pools of voters who can be counted upon to vote for their party regardless of campaign rhetoric or circumstances. In addition, all of this takes place within differ ent contexts: national, state, and local. Duncan MacRae was of the opinion in 1955 that party division from one Congressional election to the next is more stable than that g for Presidential elections. Another, later, study found ^Heinz Eulau, Class and Party in the Eisenhower Years (New York: Free Press of Glencoe, 1962), p. 29^ 7 Neil A. McDonald, The Study of Political Parties (Garden City, New York: Doubleday and Company, Inc•, 1955), p. 56. g Duncan MacRae, Jr., "Occupation and the Congres sional Vote, 1940-1950," American Sociological Review, XX. 3 (June, 1955), p. 149. 224 that 84 per cent of all votes cast in the 1958 Congressional elections were cast by party identifiers who supported their own party. Further, this vote was devoid of any issue con tent; only 7 per cent of the voters gave any issue reason for their vote while half did not even know which party had had a majority in the previous Congress (85th, 1957-1959).^ John W. Meyer suggested that minor candidates (for 11 state-wide office) get their vote on the basis of party. A state can be competitive in Presidential vote while having a one-party record in voting for state offices. Donald Stokes suggested that party loyalties are the all-important 1 2 factor at the state level. David Truman summed it up when he said: . . . the relationships that produce the vote for a President in a state or locality may be quite different from those that elect Senators and Congressmen, to say n o t h i n g o f G o v e r n o r s . 1 3 Q Donald E. Stokes and farren E. Miller, "Party Government and the Saliency of Congress," Public Opinion Quarterly, XXVI, 4 (Vinter, 1962), p. 534. ^ ^Ibid., p. 536. ^John V. Meyer, "A Reformulation of the 'Coattails' Problem," Public Opinion and Congressional Elections," ed. by Villiam N. McPhee and William A. Glasser (New York; The Free Press of Glencoe, 1962), p. 64. ^Donald E. Stokes, "1960 and the Problem of Deviating Elections," a paper delivered at the 1961 annual meeting of the American Political Science Association, St. Louis, Missouri, September 6-9, 1961, p. 12. 1 3 David S. Truman, The Governmental Process (New York: Alfred A. Knopf, 1951 ), p. 275 225 Each of the political variables, then, measures in some degree different factors and dimensions. Registration can be considered the more long range affiliation, while the vote for President Kennedy represents a somewhat less permanent commitment. The Brown vote in 1958 is pro.bably a conglomeration of his own incumbency as Attorney General for eight years, a bitter primary fight in the Republican Party, and the party label. Bert Betts* candidacy for Treasurer can, on the other hand, be properly considered as a measure of the effect of party label. None of the elections is a perfect measure of party strength, but each offers some insight into the dimensions. The following table, Table 56) presents different median values for political variables by suburb, anti suburb, and for both combined. Comparison of the distribu tion of values above and below the median for each variable by a modified chi square was used to test the level of 14 statistical significance. In only two of the five cases for registration data in suburbia and "anti-suburbia" were the differences statistically significant. In every instance, however, the median Democratic registration was lower in suburbia than in anti-suburbia. 1 4 William L. Hays, Statistics for Psychologists (New York: Holt, Rinehart, and Winston, 1963), pp. 620-23. 226 TABLE 56 MEDIAN VALUES FOB SELECTED POLITICAL VARIABLES Variables Medians Chi df Stat. Both Suburbs Anti- Suburbs Square Sig. Democratic regis tration, Sept. 1958 66.0?6 64.0fo 70.0?6 1 .99 1 no Democratic regis tration, Sept. 1960 67.0 65.0 72.0 2.02 1 no Democratic regis tration, Dec. 1960 67.0 65.0 71 .0 2.17 1 no Democratic regis tration, Sept. 1962 67.0 62.5 72.0 7.91 1 < .01 Democratic regis tration, Dec. 1962 66.0 61 .5 71.0 3.92 1 < .05 Brown, 1958 67.0 66.0 68.0 0.22 1 no Betts, 1958 62.0 60.0 62.0 0.89 1 no Kennedy, 1960 58.0 54.5 62.0 8.30 1 < .01 Brown, 1962 61 .0 54.0 64.0 10.86 1 <.001 Richards, 1962 55.0 52.0 59.0 1 5.04 <.001 Rees, 1962 61.0 56.0 65.0 7.91 1 < .01 Rose, 1962 53.0 48.0 55.0 8.10 1 < .01 Betts, 1962 66.0 62.0 69.0 3.69 1 no Richardson, 1962 52.0 47.0 59.0 39.37 1 <•001 Gibson, 1962 44.0 39.0 54.0 47.85 1 <•001 Yes, No. 24, 1962 52.0 55.0 51.0 8.07 2 < .02 Yes, No. 1, 1960 59.0 58.0 60.0 6.18 2 < .05 Democratic Turn out, 1960 83.0 88.0 76.0 33.05 1 <.001 Republican Turn out, 1960 85.0 90.0 80.0 33.74 1 <.001 Democratic Turn out, 1962 72.0 78.0 66.0 37.73 1 <.001 Republican Turn out, 1962 N: 76.0 73 81 .0 40 70.0 33 27.88 1 <.001 227 Much more clear-cut vas the election data. With three exceptions out of eight, the suburbs were more Republican (at p. < .05): for President (Kennedy, 1960), Governor (Brown, 1962), U. S. Senator (Richards, 1962), State Senator (Rees, 1962, Los Angeles County), and State Secretary of State (Rose, 1962). Two elections which produced no statistically significant differences were the Gubernatorial 1958 (Brown) and State Treasurer 1958 (Betts)o This lack of any difference in the 1958 election may be due at least in part to the bitter Republican 1 5 primary. The third election in which there was no difference was for State Treasurer 1962 (Betts). It appears that while suburbia and anti—suburbia are somewhat alike in party registration, they differ in party vote. For the non-partisan elections, State Superintendent of Public Instruction and County Tax Assessor, there were markedly different figures for suburbia and anti—suburbia. The voting behavior of suburbia differs, then, with party removed. This is further substantiated by the differences in vote for the Subversive Controls Amendment and Water Bonds propositions. Finally, there was a consistently higher level of voter turn-out in suburbia than in anti—suburbia. As later ^ ^Total Republican Primary Vote was larger than the total Democratic Primary Vote: 2,061,657 to 2,039,391. 228 analysis will indicate, this appears to be a product of the socioeconomic differences but it does favor Republican candidates. Population, Housing, and Politics Tables 57 through 67 definte several of the popula tion and housing variables examined in Chapter III as inde pendent political variables, each of which is correlated to fifteen selected dependent variables: registration (three measures); partisan elections (five measures); non-partisan (one measure); ballot propositions (two measures); and par ticipation (four measures). The correlations have been calculated for the sample (Los Angeles County), the suburban tracts, and the anti—suburban tracts. Examination of the eleven independent variables reveals substantial variations in the predictive powers of 1 6 each. High family income ($10,000+) and professional occu pation were the most consistent predictions, each producing thirty—six statistically significant correlations out of a possible forty-five. Further, they were consistent, each with eleven statistically significant correlations for the sample, twelve for suburbia, and thirteen for anti-suburbia. On the other hand, mobility as measured by residency else where in Los Angeles, 1955+ produced only two statistically significant correlations and must, on this basis, be con sidered as a very poor prediction of political behavior. ^Pearsonian correlation calculated on an IBM 1620 computer with a program written by the author. 229 TABLE 57 CORRELATION BETWEEN OWNER-OCCUPANCY (A) AND FIFTEEN SELECTED POLITICAL VARIABLES BY THREE COMMUNITY TYPOLOGIES Political Variables Sample8, Suburb^ Anti- Suburb r. r. r. I Registration Democratic registration, September 1958 Democratic registration, September 1960 Democratic registration, September 1962 -.062 -.062 -.087 .206 .203 .209 .003 .064 .062 II Partisan election Brown, Gubernatorial, 1958 Betts, Treasurer, 1958 Kennedy, Presidential, 1960 Brown, Gubernatorial, 1962 Rose, Sectly of State, 1962 -.065 -.070 -.148* -.179* -.152* .170 .175 .226 .240 .257 .062 .015 .026 .031 .031 III Non-partisan election Gibson, County Assessor, 1962 IV Ballot propositions "Yes" Prop. No. 24, 1962 (subversive controls) "Yes" Prop. No. 1, 1960 (water bonds) V Participation Democratic turn-out, Republican turn-out, Democratic turn-out, Republican turn-Qut, 1960 1960 1962 1962 N: 340* .284 -.370* 220* -.023 -.179 003 .263 -.336 072 016 518* 476* -.056 -.043 .286 .118 . 538* .401* .509* .537* 274 40 33 .119 (by t test) to be .313 (by t test) to be .343 (by t test) to be Correlation must be at least statistically significant at p. < .05. 1^ Correlation must be at least statistically significant at p. < .05. Q Correlation must be at least statistically significant at p. < .05. ^Statistically significant. 230 TABLE 58 CORRELATION BETWEEN HOUSING BUILT 1950-1960 (B-1) AND FIFTEEN SELECTED POLITICAL VARIABLES BY THREE COMMUNITY TYPOLOGIES Political Variables Sample8, Suburb^ Anti- Suburb r. r. r. I Registration Democratic registration, September 1958 Democratic registration, September 1960 Democratic registration, September 1962 -.080 -.093 -.143* .489* .477* .478* -.311 -.336 -.318 II Partisan election Brown, Gubernatorial, 1958 Betts, Treasurer, 1958 Kennedy, Presidential, 1960 Brown, Gubernatorial, 1962 Rose, Sect*y of State, 1962 -.051 -.034 -.163* -.240* -.198* .477* .475* .474* .476* .516* -.277 -.252 -.349* -.280 -.212 III Non-partisan election Gibson, County Assessor, 1962 -.448* -.143 -.267 IV Ballot propositions "Yes" proposition No. 24, 1962 (subversive controls) "Yes" proposition No. 1, 1960 (water bonds) .281* .221* .449* .441* -.359* -.581* V Participation Democratic turn-out, 1960 Republican turn-out, 1960 Democratic turn-out, 1962 Republican turn-out, 1962 -.071 -.063 .072 .103 -.185 .126 -.169 -.214 .456* .449* .621* .520* N: 274 40 33 Correlation must be at least .119 (by t test) to be statistically significant at p. < .05. ^Correlation must be at least .313 (by t test) to be statistically significant at p. < .05. Correlation must be at least .343 (by t test) to be statistically significant at p. < .05. *Statistically significant. 231 TABLE 59 CORRELATION BETWEEN SINGLE UNIT STRUCTURES (C) AND FIFTEEN SELECTED POLITICAL VARIABLES BY THREE COMMUNITY TYPOLOGIES Political Variables Sample0, Suburb** r. r. Anti- Suburb* r. I Registration Democratic registration, September 1958 Democratic registration, September 1960 Democratic registration, September 1962 II Partisan election Brown, Gubernatorial, 1958 Betts, Treasurer, 1958 Kennedy, Presidential, 1960 Brown, Gubernatorial, 1962 Rose, Sect*y of State, 1962 III Non-partisan election Gibson, County Assessor, 1962 IV Ballot propositions "Yes” Prop. No. 24, 1962 (subversive controls) "Yes" Prop. No. 1, 1960 (water bonds) V Participation Democratic Republican Democratic Republican turn-out, turn-out, turn-out, turn-out, 1960 1960 1962 1 962 N: .061 .056 .029 .049 .049 r»060 -.083 -.040 -.414* .373* .094 .047 .017 .331* .306* 274 + + + t t + t + t .228 .222 .254 .244 .231 .232 .211 .245 -.217 -.065 -.189 .369* .026 .542* .311 33 a Correlation must be at least .119 (by < .05. .343 statistically significant at p. Correlation must be at least .3T3 statistically significant at p. < .05. Correlation must be at least statistically significant at p. < .05, ♦Statistically significant. tTracts in this category are 100 per cent single unit structures; therefore, no meaningful r. may be calculated. (by (by test) to be test) to be test) to be 232 TABLE 60 CORRELATION BETWEEN MEDIAN VALUE OF PROPERTY (D) AND FIFTEEN SELECTED POLITICAL VARIABLES BX THREE COMMUNITY TYPOLOGIES Political Variables Sample8, Suburb** Anti- Suburb r. r. r . I Registration Democratic registration, September 1958 Democratic registration, September 1960 Democratic registration, September 1962 -.654* -.658* -.636* -.865* -.847* -.829* -o 227 -.172 -.105 II Partisan election Brown, Gubernatorial, 1958 Betts, Treasurer, 1958 Kennedy, Presidential, 1960 Brown, Gubernatorial, 1962 Rose, Sect*y of State, 1962 -.631* -.623* -.515* -.518* -.546* -.879* -.880* -.763* -.766* -.767* -.189 -.173 -. 1 26 -.051 .025 III Non-partisan election Gibson, County Assessor, 1962 .260* .481* -.069 IV Ballot propositions "Yes" Prop. No. 24, 1962 (subversive controls) "Yes" Prop. No. 1, 1960 (water bonds) -.580* -.128* -.790* -.236 -.437* -.429* V Participation Democratic turn-out, 1960 Republican turn-out, 1960 Democratic turn-out, 1962 Republican turn-out, 1962 .030 .086 .435* .444* .321* -.292 .620* .608* .442* .418* .289 .292 N: 274 40 33 f i t Correlation must be at least .119 statistically significant at p. < .05. (by t test) to be ^Correlation must be at least .313 statistically significant at p. < .05. (by t test) to be c Correlation must be at least .343 statistically significant at p. < .05. (by t test) to be *Statistically significant • 233 TABLE 61 CORRELATION BETWEEN RESIDENCY SAME IN 1955 (E—1) AND FIFTEEN SELECTED POLITICAL VARIABLES BY THREE COMMUNITY TYPOLOGIES Political Variables Sample0, Suburb*5 Anti- Suburb r. r. r. I Registration Democratic registration, September 1958 Democratic registration, September 1960 Democratic registration, September 1962 II Partisan election Brown, Gubernatorial, 1958 Betts, Treasurer, 1958 Kennedy, Presidential, 1960 Brown, Gubernatorial, 1962 Rose, Sect,y of State, 1962 III Non-partisan election Gibson, County Assessor, 1962 IV Ballot propositions "Yes" prop. No. 24, 1962 (subversive controls) "Yes" prop. No. 1, 1960 (water bonds) V Participation Democratic turn-out, 1960 Republican turn-out, 1960 Democratic turn-out, 1962 Republican turn-out, 1962 N: Correlation must be at least .119 (by t test) to be statistically significant at p. < .05. Correlation must be at least .313 (by t test) to be statistically significant at p. < .05. Correlation must be at least .343 (by t test) to be statistically significant at p. < .05. *Statistically significant. 01 8 .098 .1 58 017 .120 .138 006 .145 .164 050 063 040 007 020 .089 .096 .156 .1 50 .117 .181 .146 .157 .188 .166 108 -.021 -.278 069 -.225 -.280 304* -.340* -.158 162* 105 423* 340* .255 .194 .205 -.065 .260 .101 .243 .092 274 40 33 234 TABLE 62 CORRELATION BETWEEN MOVED INTO HOUSE, 1958-1960 (E-2) AND FIFTEEN SELECTED POLITICAL VARIABLES BY THREE COMMUNITY TYPOLOGIES Political Variables Sample0, Suburb** Anti-^ Suburb r. r. r. I Registration Democratic registration, September 1958 Democratic registration, September 1960 Democratic registration, September 1962 .140* .127* .116 -.034 -.026 -.031 -.172 -.138 -.178 II Partisan election Brown, Gubernatorial, 1958 Betts, Treasurer, 1958 Kennedy, Presidential, 1960 Brown, Gubernatorial, 1962 Rose, Sect*y of State, 1962 .153* .174* .1 33* .113 .121* -.053 -.051 -.045 -.051 -.034 -.162 -.157 -.221 -.197 -.21 3 III Non-partisan election Gibson, County Assessor, 1962 -.025 .141 .070 IV Ballot propositions "Yes" Prop. No. 24, 1962 (subversive controls) "Yes" Prop. No. 1, 1960 (water bonds) .071 .116 .029 .176 .239 .104 V Participation Democratic turn-out, 1960 Republican turn-out, 1960 Democratic turn-out, 1962 Republican turn-out, 1962 -.189* -.112 -.637* -.560* -.206 -.303 -.248 .024 -.512* -.105 -.460* -.135 N: 274 40 33 Correlation must be at least .119 (by t test) to be statistically significant at p. < .05. Correlation must be at least .313 (by t test) to be statistically significant at p. < ,05. Correlation must be at least .343 (by t test) to be statistically significant at p. < ,05. *Statistically significant. 235 TABLE 63 CORRELATION BETWEEN RESIDENCE ELSEWHERE IN 1955, LOS ANGELES (E-3) AND FIFTEEN SELECTED POLITICAL VARIABLES BY THREE COMMUNITY TYPOLOGIES Political Variables Sample8, Suburb*1 Anti- Suburb0 r. r. r. I Registration Democratic registration, September 1958 Democratic registration, September 1960 Democratic registration, September 1962 .030 .019 .011 -.215 -.204 -.219 -.006 .077 .168 II Partisan election Brown, Gubernatorial, 1958 Betts, Treasurer, 1958 Kennedy, Presidential, 1960 Brown, Gubernatorial, 1962 Rose, Sect*y of State, 1962 .047 .060 .054 .01 5 .033 -.221 -.229 -.226 -.214 -.182 -.003 -.007 .153 .184 .205 III Non-partisan election Gibson, County Assessor, 1962 -.018 .128 .459* IV Ballot propositions "Yes" Prop. No. 24, 1962 (subversive controls) "Yes" Prop. No. 1, 1960 (water bonds) .012 .199* .068 .242 .282 .342 V Participation Democratic turn-out, 1960 Republican turn-out, 1960 Democratic turn-out, 1962 Republican turn-out, 1962 -.109 -.062 -.097 -.069 -.228 -.177 .017 .194 -.165 -.277 -.291 -.206 N: 274 40 33 Correlation must be at least .119 statistically significant at p. < ,05. (by t test) to be Correlation must be at least .313 statistically significant at p. < .05. (by t test) to be Correlation must be at least .343 statistically significant at p. < .05. (by t test) to be ^Statistically significant. 0 236 TABLE 64 CORRELATION BETWEEN FAMILY INCOME, $3,0QQ-$10,000 (F-1) AND FIFTEEN SELECTED POLITICAL VARIABLES BY THREE COMMUNITY TYPOLOGIES Political Variables Sample8, Suburbb Anti- Suburbc r. r. r. I Registration Democratic registration, September 1958 Democratic registration, September 1960 Democratic registration, September 1962 .716* .714* .683* .892* .863* .845* -.104 -.045 -.027 II Partisan election Brown, Gubernatorial, 1958 Betts, Treasurer, 1958 Kennedy, Presidential, 1960 Brown, Gubernatorial, 1962 Rose, Sect*y of State, 1962 .704* .698* .576* .557* .593* .906* .895* .781* .787* .796* -.060 .060 .082 -.121 -.076 III Non-partisan election Gibson, County Assessor, 1 962 -.223* -.391* -.116 IV Ballot propositions "Yes" Prop. No. 24, 1962 (subversive controls) "Yes" Prop. No. 1, 1960 (water bonds) .576* .049 .819* .307 -.034 -.261 V Participation Democratic turn-out, 1960 Republican turn-out, 1960 Democratic turn-out, 1962 Republican turn-out, 1962 -.030 -.062 -.518* -.526* -.274 .231 -.612* -.565* .357* .164 .061 .011 N: 274 40 33 Correlation must be at least .119 (by t test) to be statistically significant at p. < .05. Correlation must be at least ,313 (by t test) to be statistically significant at p, < .05. Correlation must be at least .343 (by t test) to be statistically significant at p. < .05. *Statistically significant. 237 TABLE 65 CORRELATION BETWEEN FAMILY INCOME, $10,000 PLUS (F-2) AND FIFTEEN SELECTED POLITICAL VARIABLES BY THREE COMMUNITY TYPOLOGIES Political Variables Sample8, Suburb** Anti- Suburb0 r. r. r. I Registration Democratic registration, September 1958 Democratic registration, September 1960 Democratic registration, September 1962 -.719* -.718* -.708* -.889* -.865* -.848* -.564* -.539* -.494* II Partisan election Brown, Gubernatorial, 1958 Betts, Treasurer, 1958 Kennedy, Presidential, 1960 Brown, Gubernatorial, 1962 Rose, Sect*y of State, 1962 -.708* -.699* -.628* -.637* -.657* -.904* -.898* -.780* -.791* -.796* -.537* -.505* -.£02* -.450* -.355* III Non-partisan election Gibson, County Assessor, 1962 .099 .392* -.261 IV Ballot propositions "Yes" Prop. No. 24, 1962 (subversive controls) "Yes" Prop. No. 1, 1960 (water bonds) -.479* -.087 -.801* -.308 -.207 —.643* V Participation Democratic turn-out, 1960 Republican turn-out, 1960 Democratic turn-out, 1962 Republican turn-out, 1962 .063 .080 .633* .625* .265 -.227 .614* .570* .524* .580* .420* .522* N: 274 40 33 Correlation must be at least .119 (by t test) to be statistically significant at p. < .05. ^Correlation must be at least .313 (by t test) to be statistically significant at p. < .05. Correlation must be at least .343 (by t test) to be statistically significant at p. < .05. *Statistically significant. 238 TABLE 66 CORRELATION BETWEEN OCCUPATION, PROFESSIONAL (G-1) AND FIFTEEN SELECTED POLITICAL VARIABLES BT THREE COMMUNITY TYPOLOGIES Political Variables Sample0 , Suburb^ Anti- Suburb r. r. r. I Registration Democratic registration, September 1958 Democratic registration, September 1960 Democratic registration, September 1962 -.720* -.709* -.689* -.912* -.896* -.854* -.509* -.507* -.507* II Partisan election Brown, Gubernatorial, 1958 Betts, Treasurer, 1958 Kennedy, Presidential, 1960 Brown, Gubernatorial, 1962 Rose, Sect‘y of State, 1962 -.698* -.684* -.589* -.591* -.611* -.918* -.911* -.804* -.788* -.791* -.474* -.488* -.453* -.496* -.434* III Non-partisan election Gibson, County Assessor, 1962 .117 .235 .026 IV Ballot propositions "Yes" Prop. No. 24, 1962 (subversive controls) "Yes" Prop. No. 1, 1960 (water bonds) -.505* -.033 -.743* -.319* -.247 -.497* V Participation Democratic turn-out, 1960 Republican turn-out, 1960 Democratic turn-out, 1962 Republican turn-out, 1962 .012 .070 .414* .376* .248 -.223 • 605* .570* .479* .523* .416* .638* N: 274 40 33 Correlation must be at least .119 (by t test) to be statistically significant at p. < .05. Correlation'must be at least .313 (by t test) to be statistically significant at p. < .05. Correlation must be at least .343 (by t test) to be statistically significant at p. < .05. *Statistically significant. 239 TABLE 67 CORRELATION BETWEEN OCCUPATION, MANAGERIAL (G-2) AND FIFTEEN SELECTED POLITICAL VARIABLES BY THREE COMMUNITY TYPOLOGIES Political Variables Sample8, Suburb*5 Anti- Suburb r o r. r. I Registration Democratic registration, September 1958 Democratic registration, September 1960 Democratic registration, September 1962 -.678* -.690* -.684* -.913* -.895* -.871* -.261 -.292 -.274 II Partisan election Brown, Gubernatorial, 1958 Betts, Treasurer, 1958 Kennedy, Presidential, 1960 Brown, Gubernatorial, 1962 Rose, Sectty of State, 1962 —.669* -.659* -.598* -.605* -.620* -.903* -.911* -.829* -.820* -.831* -.227 -.231 -.210 -.171 -.153 III Non-partisan election Gibson, County Assessor, 1962 .118 • .31 3* .083 IV Ballot propositions "Yes" Prop. No. 24, 1962 (subversive controls) "Yes" Prop. No. 1, 1960 (water bonds) -.465* -.028 -.778* -.165 -.462* -.250 V Participation Democratic turn-out, 1960 Republican turn-out, 1960 Democratic turn-out, 1962 Republican turn-out, 1962 .008 .042 .538* .583* .226 -.177 .710* .698* .244 .183 .223 .084 N: 274 40 33 Correlation must be at least .119 (by t test) to be statistically significant at p. < .05. Correlation must be at least .3P3 (by t test) to be statistically significant at p. < .05. Correlation must be at least ,343 (by t test) to be statistically significant at p. < ,05. *Statistically significant. 240 High family income Examination of Table 65, family income $10,000 plus, reveals not only the consistent predictive ability of the variable but, more significantly, the relationship is the same for each of the three community typologies„ Whether one compares suburbia to its antithesis or to the county as a whole, the strong negative correlation between high family income and Democratic registration or Democratic voting remains constant. Professional occupation Examination of Table 66, occupation, professional, reveals another consistent predictor of political behavior. Like high family income, professional occupation correlates negatively to Democratic registration and Democratic voting. This relationship holds for the suburbs, the anti-suburbs, and the county. In looking at each of the previous tables it can be seen that (as in the relationship between population and housing variables in Chapter III) these correlations are different within the county, suburbia, or anti-suburbia for the various political—population and political-housing variables. However, most of the correlations differ in degree rather than sign. And, in many instances, the differences in degree mean for one typology a lack of statistical 241 significance. Thus, for example, in Table 57, owner- occupancy has a low negative statistically significant cor relation with the Kennedy Presidential vote but, while the correlation is positive in suburbia, it is not statistically significant. Thus, one must say that there is no correla tion in suburbia and a very low negative one for the County. Only one of the eleven population and housing variables exhibited a striking 'change in correlation between political variables by community. In Table 58, housing units built 1950-1960* it was found that for six political variables the relationships differed not only in degree 1 7 but in sign, thus: TABLE 68 CHANGES IN THE CORRELATION BETWEEN HOUSING BUILT 1950-1960 AND CERTAIN POLITICAL VARIABLES BY THREE COMMUNITY TYPOLOGIES Variable Sample Suburb Democratic registration, September 1962 -.143* .478* Kennedy, Presidential, 1960 -.163* .474* -.349 Brown, Gubernatorial, 1962 -.240* .476* Rose, Secretary of State, 1962 -.198* . 516* ’ ’Yes,’ 1 No. 24. .281* .449* -.359* "Yes," No. 1. .221* .441* -.581* *Statistically significant. 1 7 Table 58, p. 230, Chapter IV, this dissertation. 242 The relationship between housing age and political behavior is different within suburbia from within the county. As the percentage of new housing increases in * census tracts within the county generally, the percentage of Democratic registration and vote decreases. However, in suburbia, as the percentage' of new housing increases the percentage of Democratic registration and vote also increases. All other correlations were in agreement as to sign or were not statistically significant. It appeared, then, that except for the age of housing factor, the relationships between variables is similar for the county, suburbia, and anti-suburbia. Politics, Housing, and Population Reversing the process, tables 69 through 83 define fifteen political variables as independent and present their relationships (measured by a Pearsonian correlation) with other political, housing, and population variables. The same pattern emerges with the major changes in sign being related to the age of housing. The Political Impact of Three Selected Variables As was noted earlier in this chapter, two variables, high family income and professional occupation, were the most consistent predictors of political behavior for the 243 TABLE 69 CORRELATION BETWEEN DEMOCRATIC REGISTRATION, SEPTEMBER 1958, AND SELECTED POLITICAL, HOUSING, AND POPULATION VARIABLES BY THREE COMMUNITY TYPOLOGIES Variables Sample8 , Suburb^ Anti- Suburb * r. r. r. Political variables Brown, Gubernatorial, 1958 Engle, Senatorial, 1958 Richards, State Senate, 1958 Betts, State Treasurer, 1958 .981* .970* .979* .985* .987* .973* .979* .987* .976* • 966* .957* .986* Housing variables Dwelling units owner-occupied Dwelling units built, 1950-1960 Single unit structures Value of property -.062 -.080 .061 -.654* .206 .489* t -.865* .003 -.311 .228 -.227 Population variables Housing units moved into, 1958-1960 Residence same as in 1955 Residence elsewhere, Los Angeles, 1955 .140* -.018 .030 -.034 .098 -.215 -.172 .158 -.006 Family income, $3,000-$10,000 Family income, $10,000+ Occupation, professional Occupation, managerial .71 6* -.719* -.720* -.678* .892* -.889* -.912* -.913* -.104 -.564* -.509* -.261 N: 274 40 33 Correlation must be at least .119 (by t test) to be statistically significant at p. < .05. Correlation must be at least .313 (by t test) to be statistically significant at p. < .05. Correlation must be at least .343 (by t test) to be statistically significant at p. < .05. *Statistically significant. tTracts in this category are 100 per cent single unit structures; therefore, no meaningful r. may be calculated. 244 TABLE 70 CORRELATION BETWEEN DEMOCRATIC REGISTRATION, SEPTEMBER 1960, AND SELECTED POLITICAL, HOUSING, AND POPULATION VARIABLES BY THREE COMMUNITY TYPOLOGIES Variables Samplea Suburb*5 Anti- Suburb0 r. r. r. Political variables Kennedy, Presidential, 1960 State water bonds, yes, 1960 Democratic turn-out, 1960 Republican turn-out, 1960 .948* .034 .027 -.017 .967* .204 -.206 .226 .959* .481* -.083 .362* Housing variables Dwelling units owner-occupied Dwelling units built, 1950-1960 Single unit structures Value of property -.062 -.093 -.056 -.658* .203 .477* t -.849* .064 -.336 .222 -.172 Population variables Housing units moved into, 1958-1960 Residence same as in 1955 Residence elswhere in Los Angeles, 1955 Family income, $3,000-$10,000 Family income, $10,000+ Occupation, Professional Occupation, managerial .127* -.017 .019 .714* -.718* -.709* -. 6-90* -.026 .119 -.204 .863* -.865* -.896* -.895* -.138 .138 .077 -.045 -.539* -.507* -.293 N: 274 40 33 Correlation must be at least .119 (by t test) to be statistically significant at p. < .05. ^Correlation must be at least .313 (by t test) to be statistically significant at p. < .05. Correlation must be at least .343 (by t test) to be statistically significant at p. < .05. *Statistically significant. ^Tracts in this category are 100 per cent single unit structures; therefore, no meaningful r. may be calculated. 245 TABLE 71 CORRELATION BETWEEN DEMOCRATIC REGISTRATION, SEPTEMBER 1962, AND SELECTED POLITICAL, HOUSING, AND POPULATION VARIABLES BY THREE COMMUNITY TYPOLOGIES Variables Sample8 , Suburb*5 Anti- Suburb0 r. r. r. Political variables Brovn, Gubernatorial, 1962 Richards, Senatorial, 1962 Rees, State Senate, 1962 Rose, Secretary of State, 1962 Betts, State Treasurer, 1962 Richardson, Supt. Public Instruct 1 962 .957* .978* • 965* .968* .985* * .738* .976* .981* .963* .977* .989* .518* . 966* .969* .957* .952* .972* .739* Gibson, County Assessor, 1962 Subversive controls, yes, 1962 Democratic turn-out, 1962 Republican turn-out, 1962 .132* .226* -.354* -.539* -.169 .618* -.535* -.634* .171 -.1 55 -.171 -.507* Housing variables Dwelling units owner-occupied Dwelling units built, 1950-1960 Single unit structures Value of property -.087 -.143* .029 -.636* .209 .478* t -.829* .062 -.318 .254 -.105 Population variables Housing units moved into, 1958-1960 Residence same as in 1955 Residence elsewhere in Los Angeles, 1955 Family income, $3,000-$10,000 Family income, $10,000+ Occupation, professional Occupation, managerial .116 -.006 .011 .683* -.708* -.687* -.684* -.031 .145 -.219 .845* -.848* -.854* -.871* -.178 -.164 .168 -.027 -.494* -.507* -.274 N: 274 40 33 Correlation must be at least .119 (^.313; c.343) by t test to be statistically significant at p. < .05. ♦Statistically significant. tTracts in this category are 100 per cent single unit structures; therefore, no meaningful r. may be calculated. 246 TABLE 72 CORRELATION BETWEEN BROWN VOTE, 1958, AND SELECTED POLITICAL, HOUSING, AND POPULATION VARIABLES BY THREE COMMUNITY TYPOLOGIES Variables Sample8 , Suburb** Anti- Suburb0 r. r. r. . Political variables Democratic registration, September 1958 Engle, Senatorial, 1958 Richards, State Senate, 1958 Betts, State Treasurer, 1958 Democratic turn-out, 1960 Republican turn-out, 1960 .981* .977* .977* .990* -.374* -.515* .987* .983* .984* .992* -.634* -.695* .976* .948* .974* .983* -.117 -.520* Housing variables Dwelling units owner-occupied Dwelling units built, 1950-1960 Single unit structures Value of property -.065 -.051 .049 -.631* .170 .477* t -.879* .062 -.277 .244 -.189 Population variables Housing units moved into, 1958-1960 Residence same as in 1955 Residence elsewhere in Los Angeles, 1955 Family income, $3,000-$10,000 Family income, $10,000+ Occupation, professional Occupation, managerial .1 53* -.050 .047 .704* -.708* -.698* -.669* -.053 .089 -.221 .906* -.904* -.918* -.903* -.162 .181 -.003 -.061 -.537* -.474* -.227 N: 274 40 33 Correlation must be at least .119 (by t test) to be statistically significant at p. < .05. Correlation must be at least .313 (by t test) to be statistically significant at p. < .05. Correlation must be at least .343 (by t test) to be statistically significant at p. < .05. ♦Statistically significant. ^Tracts in this category are 100 per cent single unit structures; therefore, no meaningful r. may be calculated. 247 TABLE 73 CORRELATION BETWEEN BETTS VOTE, 1958, AND SELECTED POLITICAL, HOUSING, AND POPULATION VARIABLES BY THREE COMMUNITY TYPOLOGIES Variables Samplea Suburb*1 Anti- Suburb r. r. r. Political variables Democratic registration, September 1958 • 985* .987* .986* Kennedy, 1960 .937* .929* .921* Subversive controls, yes, 1962 .260* .685* -.262 State water bonds, yes, I960 .056 .235* .457* Housing variables Dwelling units owner-occupied -.070 .175 .01 5 Dwelling units built, 1950-1960 -.034 .475* -.252 Single unit structures .049 t .231 Value of property -.623* -.880* -.173 Population variables Housing units moved into, 1958-1960 .174* -.051 . -.157 Residence same as in 1955 -.063 .095 .146 Residence elsewhere in Los Angeles, 1955 .060 -.229 -.007 Family income, $3,000-010,000 .698* .895* -.060 Family income, $10,000+ -.699* -.898* -.505* Occupation, professional —.684* -.911* -.488* Occupation, managerial -.659* -.911* -.231 N: 274 40 33 Correlation must be at least .119 statistically significant at p. < .05. (by t test) to be T . Correlation must be at least .313 (by t test) to be statistically significant at p. < .05. Correlation must be at least .343 (by t test) to be statistically significant at p. < .05. *Statistically significant, ^Tracts in this category are 100 per cent single unit structures; therefore, no meaningful r. may be calculated. 248 TABLE 74 CORRELATION BETWEEN KENNEDY VOTE, 1960, AND SELECTED POLITICAL, HOUSING, AND POPULATION VARIABLES BY THREE COMMUNITY TYPOLOGIES Variables Sample8 , Suburb^ Anti- Suburb0 r. r. r. Political variables Democratic registration, September 1960 State water bonds, yes, 1960 Democratic turn-out, 1960 Republican turn-out, 1960 .948* .092 .058 .032 .967* .184 -.120 .246 .959* .499* -.021 -.370* Housing variables Dwelling units owner-occupied Dwelling units built, 1950-1960 Single unit structures Value of property -.148* -.163* -.060 -.515* .226 .474* t -.763* .026 -.349* .232 -.126 Population variables Housing units moved into, 1958-1960 Residence same as in 1955 Residence elsewhere in Los Angeles, 1955 Family income, $3,000-$10,000 Family income, $10,000+ Occupation, professional Occupation, managerial .133* -.040 .054 .577* -.628* -.589* -.599* -.045 .156 -.226 .781* -.780* -.804* -.828* -.227 .157 .1 53 -.082 -.502* -.453* -.210 N: 274 40 33 C L Correlation must be at least .119 (by t test) to be statistically significant at p. < .05. ^Correlation must be at least .313 (by t test) to be statistically significant at p. < .05. Correlation must be at least .343 (by t test) to be statistically significant at p. < .05. ^Statistically significant. ^Tracts in this category are 100 per cent single unit structures; therefore, no meaningful r. may be calculated. 249 TABLE 75 CORRELATION BETWEEN BROWN VOTE, 1962, AND SELECTED POLITICAL, HOUSING, AND POPULATION VARIABLES BI THREE COMMUNITY TYPOLOGIES Variables Sample8 , r. Suburb*3 r. Anti- Suburb r. Political variables Democratic registration, September 1962 Rose, Secretary of State, 1962 - Democratic turn-out, 1962 Republican turn-out, 1962 .957* .844* -.133* -.061 .976* .631* .258 .171 . 966* .761* .070 .044 Housing variables Dwelling units owner-occupied Dwelling units built, 1950-1960 Single unit structures Value of property -.180* -.240* -.083 -.518* .240 .476* t -.766* .031 -.280 .211 -.051 Population variables Housing units moved into, 1958-1960 Residence same as in 1955 Residence elsewhere in Los Angeles, 1955 Family income, $3,000—$10,0Q0 Family income, $10,000+ Occupation, professional Occupation, managerial .113 -.007 .01 5 .557* -.637* -.591* -.605* .051 .150 -.214 .787* -.791* -.788* -.820* -.197 .188 .184 -.121 -.450* -.496* -.171 N: 274 40 33 Correlation must be at least .119 (by t test) to be statistically significant at p. < .05. ^Correlation must be at least .313 (by t test) to be statistically significant at p. < .05. Correlation must be at least .343 (by t test) to be statistically significant at p. < .05. *Statistically significant. tTracts in this category are 100 per cent single unit structures; therefore, no meaningful r. may be calculated. 250 TABLE 76 CORRELATION BETWEEN ROSE VOTE, 1962, AND SELECTED POLITICAL, HOUSING, AND POPULATION VARIABLES BY THREE COMMUNITY TYPOLOGIES Variables Sample8 , Suburb^5 Anti- Suburb r. r. r. Political variables Democratic registration, September 1962 Brown, Gubernatorial, 1962 Gibson, County Assessor, 1962 Subversive controls, yes, 1962 Democratic turn-out, 1962 Republican turn-out, 1962 .968* .844* .219* .080 -.323* -.560* .977* .631* -.112 .551* -.478* -.628* .952* .761* .253 -.286 -.052 —.463* Housing variables Dwelling units owner-occupied Dwelling units built, 1950-1960 Single unit structures Value of property -.152* -.198* -.040 -.546* .257 .516* + -.767* .031 -.212 .245 .025 Population variables Housing units moved into, 1958-1960 Residence same as in 1955 Residence elsewhere in Los Angeles, 1955 Family income, $3,000-$10,000 Family income, $10,000+ Occupation, professional Occupation, managerial N: .121* -.034 -.21 3 .020 .117 .166 .033 -.182 .205 .593* .657* .611* .620* .796* -.796* -.791* -.831* -.076 -.355* -.435* -.1 53 274 40 33 .119 (by t test) to be .313 • (by t test) to be Correlation must be at least statistically significant at p. < .05 Correlation must be at least statistically significant at p. < .05 Correlation must be at least .343 (by t test) to be statistically significant at p. < *05. *Statistically significant. +Tracts in this category are 100 per cent single unit structures; therefore, no meaningful r. may be calculated. 251 TABLE 77 CORRELATION BETWEEN GIBSON VOTE, 1962, AND SELECTED POLITICAL, HOUSING, AND POPULATION VARIABLES BT THREE COMMUNITY TYPOLOGIES Variables Sample8 , Suburb^ Anti- Suburb r. r. r. Political variables Democratic registration, September 1962 Rose, Secretary of State, 1962 Subversive controls, yes, 1962 State water bonds, yes, I960 Democratic turn-out, 1962 Republican turn-out, 1962 .132* .219* -.494* -.094 .138* .002 -.169 -.112 -.460* .1 59 .252 .165 .171 .253 .041 .500* -.226 -.252 Housing variables Dwelling units owner-occupied Dwelling units built, 1950—1960 Single unit structures Value of property -.340* -.448* -.414* .260* .284 -.143 t .481* -.370* -.267 -.217 -.070 Population variables Housing units moved into, 1958-1960 Residence same as in 1955 Residence elsewhere in Los Angeles, 1955 Family income, $3,000~$10,000 Family income, $10,000+ Occupation, professional Occupation, managerial -.025 .108 -.01 8 -.223* .099 .117 .118 .142 -.021 .1 26 -.391* .392* .235 .31 3* .070 -.278 .460* -.116 -.261 .026 .083 N: 274 40 33 fl# / \ Correlation must be at least .119 (by t test) to be statistically significant at p. < .05. Correlation must be at least .313 (by t test) to be statistically significant at p. < .05. Correlation must be at least .343 (by t test) to be statistically significant at p. < .05. *Statistically significant. tTracts in this category are 100 per cent single unit structures; therefore, no meaningful r. may be calculated. 252 TABLE 78 CORRELATION BETWEEN "TES" VOTE ON SUBVERSIVES CONTROL (PROPOSITION NO. 24) AND SELECTED POLITICAL, HOUSING, AND POPULATION VARIABLES BY THREE COMMUNITY TYPOLOGIES Variables Sample8 , Suburbb Anti-C Suburb r. r. r. Political variables Democratic registration, September 1960 Betts, State Treasurer, 1958 Rose, Secretary of State, 1962 State water bonds, yes, I960 Democratic turn-out, 1962 Republican turn-out, 1962 .226* .260* .080 .060 -.425* -.230* .618* .685* .551* .204 -.790* -.725* -.155 -.262 -.286 .341 -.441* -.019 Housing variables Dwelling units owner-occupied Dwelling units built, 1950-1960 Single unit structures Value of property .220* .281* .373* -.580* 1 -.023 .449* t -.790* -.179 -.359* -.065 -.437* Population variables Housing units moved into, 1958-1960 Residence same as in 1955 Residence elsewhere in Los Angeles, 1955 Family income, $3,000-$10,000 Family income, $10,000+ Occupation, professional Occupation, managerial .071 -.070 .012 .576* -.480* -.505* -.465* .029 -.225 .068 .819* -.801* -.743* -.779* .239 -.280 .282 -.034 -.207 -.247 -.462* Ns 274 40 33 Correlation must be at least .119 (by t test) to be statistically significant at p. < .05. i_ "" Correlation must be at least .313 (by t test) to be statistically significant at p. < .05. Correlation must be at least .343 (by t test) to be statistically significant at p. < .05. *Statistically significant. tTracts in this category are 100 per cent single unit structures; therefore, no meaningful r. may be calculated. 253 TABLE 79 CORRELATION BETWEEN "YES" VOTE ON STATE WATER BONDS, (PROPOSITION NO. 1) AND SELECTED POLITICAL* HOUSING, AND POPULATION VARIABLES BY THREE COMMUNITY TYPOLOGIES Variables Sample0 , Suburb^ Anti-C Suburb r. r. r. Political variables Democratic registration, September 1960 Kennedy, Presidential, 1960 Democratic turn-out, I960 Republican turn-out, 1960 .034 .092 -.173* -.149* .204 .184 -.072 -.220 .481* .499* -.545* -.621* Housing variables Dwelling units owner-occupied Dwelling units built, 1950-1960 Single unit structures Value of property .063 .221* .094 -.128* .263 .441* t -.236 -.336 -.581* -.189 -.429* Population variables Housing units moved into, 1958-1960 Residence same as in 1955 Residence elsewhere in Los Angeles, 1955 Family income, $3,000-$10,000 Family income, $10,000+ Occupation, professional Occupation, managerial .116 -.304* .199* .049 -.087 -.033 -.028 -.176 -.340* .242 .307 -.308 -.319* -.165 .104 -.158 .342 -.261 -.643* -.497* -.250 N: 274 40 33 Correlation must be at least .119 (by t test) to be statistically significant at p. < .05. Correlation must be at least .313 (by t test) to be statistically significant at p. < .05. Correlation must be at least .343 (by t test) to be statistically significant at p. < .05. *Statistically significant. tTracts in this category are 100 per cent single unit structures; therefore, no meaningful r. may be calculated. 254 TABLE 80 CORRELATION BETWEEN DEMOCRATIC TURN-OUT, 1960, AND SELECTED POLITICAL, HOUSING, AND POPULATION VARIABLES BY THREE COMMUNITY TYPOLOGIES Variables Sample8 , r. Suburb*3 r. Anti- Suburb r. Political variables Democratic registration, September 1960 Kennedy, Presidential, 1960 Yes, State water bonds, 1960 Republican turn-out, 1960 Democratic turn-out, 1962 Republican turn-out, 1962 .027 .058 -.173* .727* .078 .016 -.206 -.120 -.072 .179 .338* .304 -.083 -.021 -.545* .784* .609* .376* Housing variables Dwelling units owner-occupied Dwelling units built, 1950-1960 Single unit structures Value of property * .072 -.071 .047 .030 -.056 -.185 t .321 .£38* .456* .369* .442* Population variables Housing units moved into, 1958-1960 Residence same as in 1955 Residence elsewhere in Los Angeles, 1955 Family income, $3,000-110,000 Family income, $10,000+ Occupation, professional Occupation, managerial -.189* .162* -.109 -.030 .063 .012 .008 -.206 .255 -.228 -.274 .265 .248 .226 -.512* .261 -.165 .357* .524* .479* .244 N: 274 40 33 a, Correlation must be at least .119 (by (by (by statistically significant at p. ^Correlation must be at statistically significant at p. < .05. least .313 < .05. test) to be test) to be test) to be Correlation must be at least .343 statistically significant at p. < .05. ^Statistically significant. tTracts in this category are 100 per cent single unit structures; therefore, no meaningful r. may be calculated. 255 TABLE 81 CORRELATION BETWEEN REPUBLICAN TURN-OUT, 1960, AND SELECTED POLITICAL, HOUSING, AND POPULATION VARIABLES BY THREE COMMUNITY TYPOLOGIES Variables Sample0 , Suburb*1 Anti- Suburbc r. r. r. Political variables Democratic registration, September 1960 Kennedy, Presidential, 1960 Yes, State water bonds, 1960 Democratic turn-out, 1960 Democratic turn-out, 1962 Republican turn-out, 1962 -.017 .032 -.149* .727* .470* .018 .226 .246 -.220 .179 .064 -.073 -.362* -.370* -.621* .784* .468* .467* Housing variables Dwelling units owner-occupied Dwelling units built, 1950-1960 Single unit structures , Value of property .016' -.063 -.017 .086 -,043 .126 t -.292 .401* .449* .026 .418* Population variables Housing units moved into, 1958-1960 Residence same as in 1955 Residence elsewhere in Los Angeles, 1955 Family income, $3,000-$10,000 Family income, $10,000+ Occupation, professional Occupation, managerial -.112' .105 -.062 -.062 .080 .070 .042 -.303 .194 -.177 .231 -.227 -.223 -.177 -.105 .101 -.277 .164 .580* .523* .183 N: 274 40 33 Correlation must be at least .119 (by t test) to be statistically significant at p. < .05. ^Correlation must be at least .313 (by t test) to be statistically significant at p. < .05. Correlation must be at least .343 (by t test) to be statistically significant at p. < .05. *Statistically significant. ^Tracts in this category are 100 per cent single unit structures; therefore, no meaningful r. may be calculated. 256 TABLE 82 CORRELATION BETWEEN DEMOCRATIC TURN-OUT, 1962, AND SELECTED POLITICAL, HOUSING, AND POPULATION VARIABLES BX THREE COMMUNITY TYPOLOGIES Variables Sample8 , Suburb** Anti- Suburb r. r. r. Political variables Democratic registration, September, 1962 Rose, Secretary of State, 1962 Gibson, County Assessor, 1962 Yes, Subversives Control, 1962 Democratic turn-out, 1960 Republican turn-out, 1960 Republican turn-out, 1962 -.354* -.323* .1 38* -.425* .078 .470* .814* -.535* -.478* .252 -.790* .338* .064 .803* -.171 -.052 -.226 -.441* .609* .468* .692* Housing.variables Dwelling units owner-occupied Dwelling units built, 1950-1960 Single unit structures Value of property .518* .072 .331* .435* .286 -.169 t .620* .509* .621* .542* .289 Population variables Housing units moved into, 1958-1960 Residence same as in 1955 Residence elsehwere in Los Angeles, 1955 Family income, $3,000-$10, Family income, $10,0004- Occupation, professional Occupation, managerial 000 -.637* .423* -.097 -.518* .633* .414* .538* -.248 .205 .017 -.612* .614* .605* .710* -.460* .243 -.291 .061 .420* , 41 6* .223 N: 274 40 33 Correlation must be at least .119 (by t test) to be statistically significant at p. < .05. ^Correlation must be at least .313 (by t test) to be statistically significant at p. < .05. Correlation must be at least .343 (by t test) to be statistically significant at p. < .05. ♦Statistically significant. tTracts in this category are 100 per cent single unit structures; therefore, no meaningful r. may be calculated. 257 TABLE 83 r CORRELATION BETWEEN REPUBLICAN TURN-OUT, 1962, AND SELECTED POLITICAL, HOUSING, AND POPULATION VARIABLES BY THREE COMMUNITY TYPOLOGIES Variables Sample8 , r. Suburb*3 r. Anti- Suburb r. Political variables Democratic registration, September 1962 Rose, Secretary of State, 1962 Gibson, County Assessor, 1962 Yes, Subversives Contol, 1962 Democratic turn-out, 1960 Republican turn-out, 1960 Democratic turn-out, 1962 -.539* -.560* .002 -.230 .016 .018 .814* -.634* -.628* .165 -.725* .304 -.073 .803* -.507* -.463* -.252 -.019 .376 .467* .692* Housing variables Dwelling units owner-occupied Dwelling units built, 1950-1960 Single unit structures Value of property .476* .103 .306* .444* .118 -.217 + .608* .537* .520* .311 .292 Population variables Housing units moved into, 1958-1960 Residence same as in 1955 Residence elsewhere in Los Angeles, 1955 Family income, $3,000-$10,000 Family income, $10,000+ Occupation, professional Occupation, managerial -.560* .340* .069 -.526* .625* .376* .583* .024 -.065 .194 -.565* .570* .570* .698* -.135 .092 -.206 .011 .522* .638* .084 N: 274 40 33 Correlation must be at least .119 (by t test) to be statistically significant at p. < .05. Correlation must be at least .313 (by t test) to be statistically significant at p. < .05. Correlation must be at least .343 (by t test) to be statistically significant at p. < .05. ■^Statistically significant. tTracts in this category are 100 per cent single unit structures; therefore, no meaningful r. may be calculated. 258 three groups of census tracts examined (suburbia, anti suburbia, and the county sample). Both variables correlated positively, consistently, and significantly with Republican political behavior. A third variable, median value of prop erty, correlated positively and significantly with Republican behavior for the county sample and for the suburban tracts. In order to more adequately assess the political impact of these three variables, each was held constant for a comparison of the political behavior of the two suburban typologies. In each of the following analyses the county sample was ranked by census tract for the variable being examined and the ranges for lowest third (N=91), middle third (N=92), and highest third (N=91) were established. Median value of property The range of the three thirds, as established by the sample, was: low third, under $13,638; middle third, $13,639-$17,444; and high third, $17,445 plus. For the high suburban and anti-suburban tracts the distribution was: TABLE 84 DISTRIBUTION OF VALUE OF PROPERTY FOR THE SUBURBAN AND ANTI-SUBURBAN TRACTS Property Value Suburb Anti-Suburb Under $13,638 11 14 $13,639-$17,444 17 13 $17,445 plus 12 6 N: 40 33 259 Table 85, Suburban Political Behavior with Median Property Value Constant, suggests that the value of property within suburbia is an important factor in suburban political behavior, but that the suburban atmosphere itself may also be significant. Those suburban tracts in the low third were very close (76 per cent compared to 77 per cent) to the anti-suburban tracts in Democratic registration, September 1960. In two instances, the suburban tracts had a higher percentage of Democratic vote than did the anti-suburban tracts (Brown and Betts, 1958). This pattern changed slightly for the tracts in the middle range. Here the suburban tracts were slightly higher in Democratic party registration and were, like the tracts in the lower third, more Democratic in their votes for Brown and Betts in 1958. For the upper third, the suburban tracts were con sistently more Republican than the anti-suburban tracts. Thus, the relationship between suburban residence and political behavior appears to be to some extent a product of the value of property. However, with the value of property held constant there was still an observable dif ference in the behavior of the two communities. High family income The percentage of high family income (families having an income of more than $10,000 per year) for the 260 TABLE 85 A COMPARISON OF SUBURBAN AND ANTI-SUBURBAN POLITICAL BEHAVIOR WITH VALUE OF PROPERTY HELD CONSTANT* Range of Property Value One Low i-third Mid One-third High One-third Suburbf md. Democratic registration, Sept. 1960 Anti-suburb, md. Democratic registration, Sept. 1960 76.0 % 77.0 67.0^ 66.0 56.0fo 71 .0 Suburb, md. Brown, 1958 Anti-suburb, md. Brown, 1958 77.0 74.5 68.0 63.5 54.5 65.5 Suburb, md. Betts, 1958 Anti-suburb, md. Betts, 1958 72.0 69.0 62.0 60.0 48.5 62.0 Suburb, md. Kennedy, 1960 Anti-suburb, md. Kennedy, 1960 63.0 71 .5 56.0 59.5 48.0 62.0 Suburb, md. Brown, 1962 Anti-suburb, md. Brown, 1962 64.0 72.5 54.0 62.0 47.0 62.5 Suburb, md. Rose, 1962 Anti-suburb, md. Rose, 1962 58.0 63.5 50.0 53.0 41 .0 54.5 Suburb, md. Gibson Anti-suburb, md. Gibson 40.0 55.0 39.0 52.0 40.0 57.0 Suburb, md. Prop. No. 24 Anti-suburb, md. Prop. No. 24 59.0 51 .5 54.0 51 .0 52.5 51 .5 Suburb, md. Democratic turn-out, 1960 Anti-suburb, md. Democratic turn-out, 1960 86.0 74.0 89.0 76.5 91 .0 77.0 Suburb, md. Republican turn-out, 1960 Anti-suburb, md. Republican turn-out, 1960 89.0 76.0 90.0 81 .0 92.0 80.0 *The number of tracts was too small to support a chi square test of statistical significance between medians, as v*s done in the earlier tables of this chapter. 261 county sample when ranked by thirds fell into the following distribution: TABLE 86 DISTRIBUTION OF HIGH FAMILY INCOME FOR THE SUBURBAN AND ANTI-SUBURBAN TRACTS Percentage Suburb Anti-Suburb Under 16 per cent 10 33 17 per cent to 28 per cent 16 — 29 per cent or more 14 — N: 40 33 Since no anti-suburban tract fell into either the middle or high range, only a comparison of the low range could be mad&. Table 88, A Comparison of Suburban and Anti-Suburban Political Behavior with High Income Held Constant, suggests that the suburban atmosphere is not so important as is high income in producing the suburban Republican vote. Indeed, with high income held constant, suburban tracts were more Democratic in registration than anti-suburban tracts and had a higher vote for Democratic candidates than did the anti-suburban tracts. The fact that no anti-suburban tracts fell in the middle and higher ranges suggests that at least part of the Republican political behavior of suburbia is a function of the income differences rather than any so-called "suburban effect." 262 Professional occupation The impact of a third variable, the percentage of those employed in a professional occupation, was also evaluated by holding its effects constant. The range, in thirds, as established by-the county sample was: TABLE 87 DISTRIBUTION OP PROFESSIONAL OCCUPATION FOR THE SUBURBAN AND ANTI-SUBURBAN TRACTS Percentage Suburb Anti^Suburb Under 10 per cent 11 26 11 per cent to 18 per cent 12 7 19 per cent or more 17 — N: 40 33 Table 89, A Comparison of Suburban and Anti-Suburban Political Behavior with Professional Occupation Held Constant, further supports the suggestion of the two previous tables that much of suburban Republican political behavior is in fact a product of the suburban screening effect. Unfortunately, no anti-suburban tracts fell in the upper third of professional occupationsj the comparison between suburban and anti-suburban tracts is limited therefore, to the middle and lower ranges. Again, in some instances, suburban tracts were higher than anti-suburban tracts in their support for the Democratic Party. With 263 TABLE 88 A COMPARISON OF SUBURBAN AND ANTI-SUBURBAN POLITICAL BEHAVIOR WITH HIGH INCOME HELD CONSTANT* Range of High Income Low One-third Suburb, md. Democratic registration, Sept. 1960 Anti-suburb, md. Democratic registration, Sept. 1960 76.0% 72.0 Suburb, md. Brown, 1958 Anti-suburb, md. Brown, 1958 77.5 * 68.0 Suburb, md. Betts, 1958 Anti-suburb, md. Betts, 1958 72.5 62.0 Suburb, md. Kennedy, 1960 Anti-suburb, md. Kennedy, 1960 64. 5 62.0 Suburb, md. Brown, 1962 Anti-suburb, md. Brown, 1962 66.5 64.0 Suburb, md. Rose, 1962 Anti-suburb, md. Rose, 1962 61 . 5 55.0 Suburb, md. Gibson Anti-suburb, md. Gibson 41 .0 54.0 Suburb, md. Prop. No. 24 Anti-suburb, md. Prop. No. 24 59.0 51 .0 Suburb, md. Democratic turn-out, 1960 Anti-suburb, md. Democratic turn-out, 1960 87.0 76.0 Suburb, md. Republican turn-out, 1960 Anti-suburb, md. Republican turn-out, 1960 89.5 80.0 *The number of "tracts was too small to support a chi square test of statistical significance between medians, as was done in the earlier tables of this chapter. 264 TABLE 89 A COMPARISON OF SUBURBAN AND ANTI-SUBURBAN POLITICAL BEHAVIOR VITH PROFESSIONAL OCCUPATION HELD CONSTANT* Range of Professional Occupation Mid One~third Low One—third Suburb, md. Democratic registration, September 1960 Anti-suburb, md. Democratic registration, September 1960 64.0 i° 63.0 76.6 % 73.5 Suburb, md. Brown, 1958 Anti-suburb, md. Brown, 1958 67.5 59.0 78.0 69.5 Suburb, md. Betts, 1958 Anti-suburb, md. Betts, 1958 61 . 5 54.0 72.0 65.5 Suburb, md. Kennedy, 1960 Anti-suburb, md. Kennedy, 1960 54.0 55.0 64.0 66.0 Suburb, md. Brown, 1962 Anti-suburb, md. Brown, 1962 54.0 58.0 66.0 70.0 Suburb, md. Rose, 1962 Anti—suburb, md. Rose, 1962 49. 5 51 .0 61 .0 59.5 Suburb, md. Gibson Anti-suburb, md. Gibson 39.0 55.0 42.0 53.4 Suburb, md. Prop. No. 24 Anti-suburb, md. Prop. No. 24 56. 5 51 .0 59.0 51 .5 Suburb, md. Democratic turn-out, 1960 Anti-suburb, md. Democratic turn-out, 1960 88.0 77.0 86.0 75.0 Suburb, md. Republican turn-out, 1960 Anti-suburb, md. Republican turn-out, 1960 90.0 82.0 89.0 78.5 *The number of tracts was too small to support a chi square test of statistical significance between medians, as was done in the earlier tables of this chapter. 265 professional occupation held constant, suburban tracts were more Democratic in their registration than were anti suburban tracts. They were also more Democratic in their vote in 1958. CHAPTER V CONCLUSIONS AND SUMMARY Introduction Review of literature concerned with suburban polit ical behavior appears, on balance, to favor the hypothesis that (1) suburbia is Republican, and (2) that this is due, at least in part, to center-city Democrats switching to the Republican Party after their move to suburbia,, Typical of the first point of view are Edward C. Banfield, G. Edward Janosik, and Louis Harris,^ Angus Campbell and Homer C. Cooper have told us that it is only in suburbia that strong Republicans outnumber strong 2 Democrats. But, even here, in so-called Republican sub urbia, the total of Democratic identifiers is greater than that of Republican identifiers. (See Table 90.) ^Edward C. Banfield, "The Politics of Metropolitan Area Organization," Midwest Journal of Politics. I, 1 (May, 1957), passim; G. Edward Janosik, "The New Suburbia," Current History. XXXI, 180 (August, 1956), passim; and Louis Harris, Is There a Republican M a j o r i t y ? (New York; Harper and Brothers"! 1954), pp. 133-34. 2 Angus Campbell and Homer C. Cooper, Group Differ ences in Attitudes and Votes (Ann Arbor; University of Michigan Press, 1956), p. 46. 266 267 TABLE 90 SUBURBAN PARTY IDENTIFIERS Party Affiliation Per Cent of Vote Strong Democrats 15 Weak Democrats 32 Independent Democrats 9 Total 56 Strong Republicans 20 Weak Republicans 9 Independent Republicans 9 Total 38 Fred Greenstein and Raymond Wolfinger, however, found a greater proportion of suburbanites voting Republican 3 than voting Democratic. On the other hand, Bennett M, Berger found workers continue to vote Democratic, even after they have moved to the suburbs. Charles Liebman found 3 Fred I. Greenstein and Raymond E. Wolfinger, "The Suburbs and Shifting Party Loyalties," Public Opinion Quar terly, XXII, 4 (Vinter, 1958-59), p. 477. Both Thomas Dye and Bernard Lazerwitz found that the act of registration was not a reliable guide to political behavior of the suburban ites and used identification instead. See Bernard Lazerwitz, "Some Characteristics of Residential Belts in the Metropoli tan Community," (unpublished Ph.D. dissertation, University of Michigan, 1958); and Thomas R„ Dye, "Certain Political Correlates of the Areal Differentiation and Stratification of Sub-populations in a Metropolitan Area" (unpublished Ph.D. dissertation, Department of Political Science, Pennsylvania University, 1961). 4 Bennett M. Berger, Working Class Suburb (Berkeley and Los Angeles; University of California Press, 1960). 268 5 several suburbs which vote Democratic. F. M. Wirt found one—third of the suburbs he examined were Democratic on the Ranney-Kendal typology.^1 And, earlier in this dissertation, examples were given of suburbs in Los Angeles County which 7 have a majority of voters registered as Democrats. Is Suburbia Republican? * Three criteria for measuring the polities of sub urbia have been used: (1) party registration (see Chapter I, this dissertation), (2) party identification (Greenstein and Wolfinger and Campbell and Cooper), and (3) voting (Berger). Two of these, registration and voting have been used in this dissertation to determine the political complexion of Los Angeles County suburbs. Use of registration as a measure of suburban polit ical behavior did not prove to be very successful. While the suburbs were less Democratic than their antithesis, the anti-suburbs, the differences were statistically significant in only two cases out of five (September and December, 1962). While it is interesting that the differences in all cases ^Charles S. Liebman, "Functional Differentiation and Political Characteristics of Suburbs," American Journal of Sociology. LXVI, 5 (March, 1961), pp. 485-90. ^Frederick M. Wirt, "Suburban Patterns in American Politics," a paper delivered at the 1960 American Political Science Association Convention (mimeographed). 7 See Chapter I, p. 6, footnote 17 of this disserta tion. 269 between suburb and anti-suburb were in the expected direc tion, with suburbs being less Democratic than the anti suburbs, it is even more interesting and important to note that the suburbs were more Democratic than the County* The median Democratic registration for the fifty suburban tracts was 62*5 per cent in September 1962 and 61.5 per cent in December. This compares to a County-wide Democratic regis tration of 59.2 per cent and 57*7 per cent, respectively. For all five cases, 1958 through 1962, the fifty suburban tracts were more Democratic in registration than was the County as a whole. TABLE 91 DEMOCRATIC REGISTRATION FOR LOS ANGELES COUNTY AND ITS SUBURBAN TRACTS* Los Angeles Countyt Suburban Tracts (Median Percentage) September 1958 59.3 64.0 September 1960 59.9 65.0 December 1960 59.2 65.0 September 1962 59.2 62.5 December 1962 57.9 61.5 *A11 figures based on two-party totals. tSource: September 1958 and September 1960 figures are based on data found in Frank M. Jordan, State of Cali fornia. Statement of Vote for the General Elections of 1958 and 1960 (Sacramento, California: State Printing Office, 1958 and 1960); December 1960 figures are based on a per sonal communication from Jim Allison, Deputy Registrar of Voters, Los Angeles County; September and December 1962 are from Frank M. Jordan, Report of Registration for those dates (Sacramento, California: State Printing Office, 1962 and 1963). 270 TABLE 92 DEMOCRATIC VOTE IN LOS ANGELES COUNTY AND ITS SUBURBAN TRACTS Los Angeles County* Suburban Tracts (Median Percentage) Brown, 1958 58.1 66.0 Betts, 1958 52.7 60.0 Kennedy, 1960 50.4 54.5 + Brown, 1962 52.4 54.0 Richards, 1962 47.2 52.0 Rees, 1962 54.6 56.0 Rose, 1962 44.9 48.0 Betts, 1962 56.4 62.0 *County-wide data. tWhile we found the suburbs giving Kennedy 54.5 per cent of the vote, the Democratic National Committee found that Kennedy got only 41.0 per cent of the suburban vote in Los Angeles County. See: Research Division, Democratic National Committee, "The 1960 Election Report" (Washington, D.C.: mimeographed, 1962), Table 8. Use of partisan election data produced somewhat more clearcut differences. Suburban voting behavior in 1958, as measured by the Brown and Betts vote was not statistically significantly different from the anti-suburbs. However, in 1960, and in 1962 for four out of five cases, the figures were statistically significantly different. Again, as for registration, the suburbs were less Democratic than the anti-suburbs (see Table 56) and again, as for registration, the suburbs were more Democratic in vote than the County. The suburban Democratic vote at its minimum was 1.4 percentage points higher than the County Democratic vote 271 (Rees, 1962) and at its maximum was 7.9 percentage points higher than the County Democratic vote (Brown, 1958). In only one out of eight partisan elections did the suburban Democratic vote fall below 50 per cent. For the period studied, 1958-1962, suburbs in Los Angeles County registered more Democratic and voted more Democratic than the County as a whole, but less Democratic than the anti-suburbs. This conflicts with the findings of Banfield, Janosik, and Harris but is in conformity with Campbell and Cooper, who found that a majority of suburban ites had Democratic identifications (of varying intensities). Roots of Suburban Political Behavior Vhile suburban Los Angeles is not Republican, it is also less Democratic than its antithesis— anti-suburbia. Those who have taken the position that suburbia is Repub lican usually offer one of three reasons. First, there is the chameleon effect— Democrats who move to the suburbs become Republicans like their neighbors. Banfield, Green- stein and Wolfinger, and Lou Harris represent this school of g thought. However, this concept of mass political conver sion finds little support in the vast literature of social- psychology. Looking back to the discussion of the individual in g Banfield, op. cit.; Greenstein and Wolfinger, op. cit.: and Harris, op. cit. 272 group life (Chapter II), it is clear that the most potent political primary group is the family. This most pervasive of political influences has been exercised long before the move to suburbia. In conflict with this political primary group, according to the chameleon theory, are the political secondary groups: fellow workers, friends, the community at large. Research in this area has still many questions to answer, but in summary of the findings to date it can be said that there is little impirical evidence to support the contention that secondary group pressures modify basic political norms to any appreciable extent for a very large segment of the population. In fact, several studies suggest that urbanization may even weaken secondary 9 influence and strengthen the influence of family. Secondly, there are those who subscribe to the segregation or screening thesis— that Republicans move to the Republican suburbs (and, accordingly, that Democrats move to Democratic suburbs). In this school one may find Alfred de Grazia, Scott Greer, V. 0. Key, Bennett Berger, 9 Scott Greer and Ella Kube, "Urbanism and Social Structure: a Los Angeles Study," Community Structure and Analysis, Marvin B. Sussman, ed. (New York:Crowel1,1955), pp. 93-112; Morris Axelrod, "Urban Structure and Social Participation," American Sociological Review, XXI, 1 (February, 1956), pp. 13-18; and Floyd Dotson, "Patterns of Voluntary Association Among Urban Working-Class Families," American Sociological Review. XVI, 5 (October, 1951), pp. 687-93. 273 1 f ) and Lou Harris (after some second thoughts). Immediately following World War II, those moving to suburbia may have been mostly Republicans. But, soon there after, Democrats— thanks to low down payments and long-term mortgages— followed. While Republicans "moved" to the Angelus Mesa area (for example) with homes valued in excess of $15,000, Democrats "moved" to Lawndale, where houses were in the $6,000 to $7,000 range.^ Thus, as V. 0. Key and Berger suggest, there appear to be two suburbias— one Repub lican and the other Democratic. The segregation thesis appears to have more support than the chameleon thesis. David Wallace found that for Westport, Connecticut, (a New York suburb) earlier immigrants were Republicans 10 Alfred de Grazia, The Western Public (Stanford: Stanford University Press, 1954); Scott Greer, "Catholic Voters and the Democratic Party," Public Opinion Quarterly. XXV, 4 (Winter, 1961), pp. 611-25; V. 0. Key, Politics. Parties, and Pressure Groups (New York: CrowelTJ 1960), pp. 275—78; and Bennett Berger, op. cit. In 1954 Harris subscribed to the chameleon effect, but later, writing in Life, he took the position that Democrats were moving to the suburbs and remaining Demo crats. Louis Harris, Life, May 11, 1962, p. 89. ^For Angelus Mesa, see Census Tract 364-A, where 87 per cent of the homes occupied in 1950 had been built after 1940 (essentially, after 1945, of course); for Lawndale, see Census Tract 354-A, where 73 per cent of the homes were built in the same period. U.S. Bureau of the Census, U.S. Census of Population: 1950. Vol. III. Census Tract Statistics. Chapter 28. U.S. Government Printing Office, Washington, D.C., 1952, p. 171 for Tract 364-A and p. 170 for Tract 354—A. 274 1 2 while the more recent were Democrats. This suggests a two-step process which relates to the age of the area*s housing may also take place. Thirdly, there is the possibility that suburbia, aside from either or both the chameleon and segregation effects, exerts an influence singularly its own— that 1 3 suburbia is a way of life. Criteria Before assessing the various hypotheses for suburban political behavior as they apply in Los Angeles County, examination needs to be made of the political bias intro duced by the criteria used for selection of the suburbs. The criteria used appear to have some partisan political load in that they correlate positively with other variables which, in turn, correlate positively with voting and registration. While the criteria variable of single unit structures exhibited no significant correlation to political behavior (see Table 59), dwelling units owner—occupied (Table 57) and particularly age of housing (Table 58) did correlate to political behavior. Thus, the attempt to 1 2 David Wallace, MThe Sociology of Stability and Change in One Suburbts Voting" (unpublished Ph.D. disserta tion. Department of Political Science, Columbia University, 1962). 1 3 Sylvia Fava, "Suburbanism as a Way of Life." American Sociological Review, XXI, 1 (February, 1956), passim. 275 define suburbia by variables having no political connotation was not successful. As a result, due to the Republican bias of the criteria one might expect Los Angeles suburbs to be Republican, or at least less Democratic than the County, However, what were low negative correlations within the- County as a whole become moderately positive correla tions for the suburbs— noticeably so for recently built housing. The positive correlations between Democratic political behavior and recently built housing agrees with the Vallace findings. For both the County as a whole and the suburban tracts, the two criteria variables— owner— occupancy and recently built housing— have partisan political loads, though of differing magnitude and signs. The difference in signs for the correlations suggests that the relationship between variables changes depending upon the community examined. The Two Suburbs The observed change in relationship between variables depending upon the community examined (changes range from statistically significant positive figures to statistically significant negative figures) suggests that the observed changes are not just a product of sample size. For example, age of housing correlates positively with Democratic voting for the County but negatively for the suburbs (see Table 58). Thus, the older the housing in the TABLE 93 276 CORRELATION BETWEEN SUBURBAN CRITERIA AND SELECTED POLITICAL VARIABLES IN LOS ANGELES COUNTY Political Variable Owner ■ Occupancy8, Recently BuiJ.t Housing*5 Single Unit Structure0 Registration Democratic, Sept. Democratic, Sept. Democratic, Sept. 1958 1960 1962 -.062 -.062 -.087 -.080 -.093 -.143* .061 .056 .029 Partisan elections Brown, 1958 Betts, 1958 Kennedy, 1960 Brown, 1962 Rose, 1962 -.065 -.070 -.148* -.179* -.152* -.051 -.034 -.163* -.240* -.198* .049 .049 -.060 -.083 -.040 Note: The key to bottom of Table 94. Table 93 will be found at the TABLE 94 CORRELATION BETWEEN SUBURBAN CRITERIA AND SELECTED POLITICAL VARIABLES IN LOS ANGELES SUBURBS Political Variable Owner Occupancy8 Recently Built Housing*5 Single Unit Structure8 Registration Democratic, Sept. Democratic, Sept. Democratic, Sept. 1958 1960 1962 .206 .203 .209 .489* .477* .478* + t + Partisan elections Brown, 1958 Betts, 1958 Kennedy, 1960 Brown, 1962 Rose, 1962 .170 .175 .226 .240 .257 .477* .475* .474* .474* .516* + + + + €t See Table 57 for complete data. See -Table 58 for complete data. CSee Table 59 for complete data. •^Statistically significant at p. < .05. +A11 tracts in suburbia were 100 per cent single unit structures. suburb the more Republican is the suburb. This hypothesis receives considerable support from some of the other correlation figures. Old housing (built before 1940) correlates negatively with high income (family income >10.000+) for the County, but within the suburbs the correlation is positive (-.310 and .530, respectively). Similar patterns can be seen in other correlations. TABLE 95 CORRELATION BETWEEN AGE OF HOUSING AND PROPERTY VALUE BY COUNTY AND SUBURBS Variable County Suburb Housing built 1950-1960 Correlated to: Value of property .162* -.373* Income, $10,000+ .318* -.496* Housing built before 1940 Correlated to: Value of property -.132* .398* Income, $10,000+ -.310* .530* *Statistically significant at p. < .05. For complete data see Table 39 and Table 41. One of two conclusions may be drawn from these facts: (1) that as houses in the suburbs age they become more valuable, or (2) that there are two kinds of suburbs involved. The family income data suggests that the second case is the most likely. 278 Examination of the relationship between family income and age of housing reveals the following patterns: TABLE 96 CORRELATION BETWEEN FAMILY INCOME AND AGE OF HOUSING BY COUNTY AND SUBURBS Variable County Suburb Family income $3,000-$10,000 by age of housing: Built 1950-1960 Built 1940-1950 Built before 1940 (old) -.004 .062 -.018 .551* -.094 -.550* Family income $10,000+ by age of housing: Built 1950-1960 Built 1940-1950 Built before 1940 (old) .31 8* -.001 -.310* -.496* .1 51 .530* *Statistically significant at p. < .05. For complete data see Table 48 and Table 49. In suburbia, for middle-range family income there is a strong positive correlation (.551) with new housing and a strong negative correlation (-.550) with old housing. The pattern is reversed for high family income, there being a strong negative correlation (-.496) to new housing and a strong positive correlation (.530) to old housing. In light of the very high negative correlations between the two income categories (-.837 and -.992) in both the County and suburb, it appears that the suburbs divide into two commu nities by income. 279 Not surprisingly, those families with high incomes have houses which are worth more than do those families with middle-range income. The correlations are approximately the same for both County and suburb. TABLE 97 CORRELATION BETWEEN VALUE OF PROPERTY AND FAMILY INCOME BY COUNTY AND SUBURBS Variable County Suburb Value of property by: Family income $3,000—$10,000 Family income $10,000+ -.804* .863* -.913* .931* *Statistically significant complete data see Table 43. at p. < .05. For Thus, family income, value of property, and age of housing relate in consistent patterns which support the hypothesis that there are two suburbs— one with high income families living in older more valuable property, the other with middle income families living in newer less valuable housing. This concept of two suburbs— differing from each other in their class configurations— agrees with the earlier 1 4 findings of many political scientists and sociologists. 14 Otis Duncan, "Research on Metropolitan Popula tions: Evaluation of Data," Journal of the American Statis tical Association. LI, 276 (December, 1956) ; Dye, op. cit."; Amos A. Hawley, Human Ecology: A Theory of Community Structure (New York: Roland Press, 1950); Key, op. cit.; 280 And, as will be demonstrated later, the political behavior of the suburbs fits into the two—suburbs hypothesis. Class in Suburbia One of the basic assumptions concerning suburbs is that they are peopled by homogeneous conglomerations of upwardly social mobile middle class homeowners. That this is not so appears to be the case. Using property value, family income, and profes sional occupation as indices of class, it is clear that Los Angeles suburbs are not identical clusters of middle class homogeneity. The table on the following page (a summary of data contained in tables 84 through 88) recapitulates the distribution of three variables for the suburban tracts by low, middle, and high thirds (the ranges were based on County data). Of forty suburban tracts, ten were in the high third for all three criteria; six were in the low third for all three criteria. Leslie Kish, "Differentiation in Metropolitan Areas," Ameri can Sociological Review, XIX, 4 (August, 1954), passim; Liebman, op. cit.; Samuel Lubell, "Class, Party, Divide Suburban Voters," Los Angeles Times. October 24, 1962; Leo F. Schnore, "The Functions of Metropolitan Suburbs." The American Journal of Sociology. LXI, 5 (March, 1956), pp. 453-58; Eshref Shevky and Wendell Bell, Social Area Analysis (Stanford: Stanford University Press, 1955); Oliver P. Williams, Thomas R. Dye, Harold Herman, and Charles S. Liebman, "Urban Differentiation and Political Integration," a paper delivered at the 1962 American Political Science Association Convention (mimeographed); and Wirt, op. cit. 281 TABLE 98 DISTRIBUTION OP THREE CLASS VARIABLES WITHIN SUBURBAN CENSUS TRACTS Class Variables Range Property Value High Family Income Professional Occupation High third 30.0% 35.0 $ 42. 5 % Middle third 42.5 40.0 o o o Low third 27.5 25.0 27.5 N: 40 40 40 Of the ten consistently high tracts, six were located in the San Fernando Valley—Burbank area, two in Palos Verdes, one in Monterey Park, and one in La Mirada. Of the six consistently low tracts, three were in Norwalk and three in the East San Gabriel Valley. Thus, the two - 15 suburbs are separated geographically as well as by class. TABLE 99 PARTISAN DIFFERENCES BETWEEN SUBURBS AND ANTI-SUBURBS Political Variable Suburb Anti-Suburb Difference Democratic registration September 1960 65.0 72.0 -7.0 Partisan elections Brown, 1958 66.0 68.0 -2.0 Betts, 1958 60.0 62.0 -2.0 Kennedy, 1960 54.5 62.0 -7.5* Brown, 1962 54.0 64.0 -10.0* Rose, 1962 48.0 55.0 -7.0* *Statistically significant at p. < .05. 1 5 See Appendix B for map of these suburban tracts. 282 As was shown earlier in this chapter, the suburbs are more Democratic than is the County. But, compared to anti-suburbs, suburbia is less Democratic. How much of suburban political behavior is a product of class? Property values Holding property values constant reduced or reversed the differences in Democratic voting and registration in the low and middle ranges, but increased the differences in the high range. TABLE 100 PARTISAN DIFFERENCES BETWEEN SUBURBS AND ANTI SUBURBS HOLDING PROPERTY VALUE CONSTANT* Property Value Range* Political Variable Low Middle High Democratic registration September 1958 -1 .0 + 1 .0 -1 5.0 Partisan elections Brown, 1958 2.5 4.5 -11 .0 Betts, 1958 3.0 2.0 -13.5 Kennedy, 1960 -8.5 -2.5 -14.0 Brown, 1962 -8.5 -8.0 -15.5 Rose, 1962 -5.5 -3.0 -13.5 *See Table 85 for complete data. +A negative sign indicates that the suburban Demo- cratic value was lower than the anti-suburban • Occupation and income Unfortunately, there were no anti-suburban tracts which fell into the "high" range for either professional 283 occupation or family income. And, for family income there were no anti—suburban tracts which fell into the "middle" range. Thus, it was impossible to hold these two variables completely constant. However, results of the necessarily limited analysis indicate that much of the difference in political behavior between suburbs and anti-suburbs is rooted in occupation and income differences. The patterns of political behavior for the low family income and pro fessional occupation range, and for the middle range for professional occupation, are consistent with the patterns of political behavior observed in the previous table. Much of the difference in political behavior between suburbia and anti-suburbia appears to be rooted in class difference. TABLE 101 PARTISAN DIFFERENCES BETWEEN SUBURBS AND ANTI-SUBURBS HOLDING OCCUPATION AND FAMILY INCOME CONSTANT* ... . High Family pro^essional Occupation Political Variable Income r Low Range Low Range High Range Democratic registration 4.0 1.0 September 1960 b ■ ! ) Partisan elections Brown, 1958 9.5 8.5 8.5 Betts, 1958 10.5 7.5 7.5 Kennedy, 1960 2.5 -2.0+ -1 .0 Brown, 1962 2.5 -4.0 -4.0 Rose, 1958 6.5 1 . 5 -1 .5 *See tables 86 and 87 for complete data. ^The negative sign indicates that the suburban Democratic value was lower than the anti-suburban. 284 A second approach to the data, differences within typologies, further supports this conclusion. The differ ences between suburbs and anti-suburbs for Democratic registration and vote varied from 2.0 to 10.0 percentage points. But, within suburbia, holding value of property constant, the differences for Democratic vote and regis tration ranged from 15.0 to 23.5 percentage points. Thus, the differences within suburubia were greater than those between suburbia and anti-suburbia. Further, most of the difference was a product of the "high" range. Examination of Table 102 following illustrates the differences involved. The median difference between suburban and anti-suburban Democratic values was 7.0 percentage points. Within suburbia the median difference between high and low property value range for Democratic values was 19.0 per centage points; within anti-suburbia it was 9.0 percentage points. One can clearly find greater differences in partisan political behavior within suburbia than between suburbia and anti-suburbia. Residential Mobility A person who moves up in the social hierarchy will tend to change his friends, join new organizations, move to a new neighborhood; perhaps he will even change his religious affiliation; in some cases he will change his name; often he will alter his political a t t i t u d e s . 16 1 6 Seymour Martin Lipset and Reinhard Bendix, Social Mobility in Industrial Society (Berkeley and Los Angeles: University of California Press, 1959), p. 6. TABLE 102 A COMPARISON OF THE PARTISAN DIFFERENCES BETWEEN SUBURBIA AND ANTI-SUBURBIA AND THE DIFFERENCES BETWEEN THE TWO* Political Variables Between High and Low Range Between Suburbia for Property Value Within: and Anti-suburbia (Selected Data) Suburbia Anti-suburbia (In Percentage Points) Democratic registration September 1958 6.0 September 1960 7.0 20.0 6.0 December 1960 6.0 September 1962 9.5 December 1962 9.5 Partisan elections Brown, 1958 2.0 22.5 9.0 Betts, 1958 2.0 23.5 7.0 Kennedy, 1960 7.5 15.0 9.5 Brovn, 1962 10.0 17.0 10.0 Richards, 1962 7.0 Rees, 1962 9.0 Rose, 1962 7.0 17.0 9.0 Betts, 1962 7.0 Median: 7.0 19.0 9.0 Average: 6.9 19.2 8.4 *Data from tables 56 and 84. t o Oo u i 1 286 Residential and social mobility have been dominant themes in duscussions of suburbanism and, in particular, suburban politics. Both major causitive hypotheses assume either or both residential and social mobility. The chameleon hypothesis assumes that social mobility leads to residential mobility which, in turn, leads to an assumption of the perceived Republican norm of the new and desirable community. The segregation hypothesis assumes at least residential mobility and perhaps social mobility. But, in the latter case, one moves into a community which is composed of people like the mobile individual and no politi cal change follows. Lipset and Bendix assume that residen tial mobility follows social mobility (and that the mobile individual, in some cases, moves into an environment which impels him to change his politics): . . . the successful individual need only change his residential neighborhood to bring his economic and his social status into line.17 This intra-urban mobility should not be confused with the extra—urban mobility patterns discussed by Richard Scudder and C. Arnold Anderson, wherein they found social mobility 1 8 followed residential mobility. On balance, the segregation thesis seems to be more plausible in theory and to have received better empirical 17Ibid., pp. 66-67. 1 8 Richard Scudder and C. Arnold Anderson, "Migration and Vertical Occupational Mobility," American Sociological Review. IXX, 3 (June, 1954). s---- 287 support. Berger's blue collar and Greer's Catholic subur- 1 9 banites maintained their Democratic political behavior. On the other hand, Greenstein and Volfinger did find a 20 "suburban effect" in support of the chameleon hypothesis. Housing and mobility Since the age of housing relates to political behavior in Los Angeles County, and since age of housing was used as one of the criteria for defining suburbia, it is not used here as a measure of residential mobility. While, as a general rule, Democrats live in older housing than do Republicans (see Table 58), this is not proof that Democrats moved into older houses. Democrats may not be as mobile as Republicans; thus, they would be more likely to grow old with their houses. From the data in Tables 39 and 41, it appears that, areas with old housing (built before 1940) have a greater population stability than do areas with new housing. Residential mobility and partisan politics Turning to the specific measures of residential mobility used by the Bureau of the Census, it was found that the correlations between residential mobility and partisan 1 9 Berger, op. cit., and Greer, op. cit. 20 Greenstein and Wolfinger, op. cit. 288 political behavior were so low that in most cases they were not statistically significant. The measures of mobility were: residency in 1955 the same, residency in 1955 else where. Los Angeles, residency in 1955 elsewhere, California, and moved into house 1958-1960. TABLE 103 RESIDENTIAL MOBILITY AND PARTISAN BEHAVIOR IN LOS ANGELES COUNTY* Political Variables Residency in 1955 Same Elsewhere in Los Angeles Moved into House 1958-60 Democratic registra tion September 1958 -.018 .030 .140 + September 1960 -.017 .019 .127 + September 1962 -.006 .011 .116 Partisan elections Brown, 1958 -.050 .047 .153 + Betts, 1958 -.063 .060 .174 + Kennedy, 1960 -.040 .054 .133 + Brown, 1962 -.007 .015 .113 Rose, 1962 -.020 .033 .121 + *Based on tables 61, 62, and 63. +Statistically significant at p. < .05. Thus, while the signs were consistently negative for the correlation between measures of Democratic partisan behavior and residency in 1955 same, and consistently positive for residency in 1955 elsewhere in Los Angeles, in no case were these correlations statistically signifi cant. For the five-year time span (1955-1960), partisan 289 political behavior did. not correlate with residential mobility. However, by reducing the time span to two years (1958-1960), low positive and almost consistently statisti cally significant correlations were found between residen tial mobility and Democratic votes and registration. Such data suggests that to the extent residential mobility is a characteristic of /suburban areas, the new suburbs ought to be Democratic. However, this does not appear to be the case. While short-term residential mobility correlates positively to Democratic political behavior within the County as a whole, the relationship is reversed for the suburbs. But, in no case where the correlations statistically significant. In short, for suburbia, there is no significant correlation between residential mobility and partisan political behavior. Further examination of residential mobility corre lations, particularly to the three criteria variables, reveals patterns which support a general conclusion that for short periods of time suburban residents are less mobile than residents in anti-suburbia. But, for a longer period of time there is a slightly higher level of residen tial mobility in suburbia than in anti-suburbia. Since we have defined suburbia as recently built owner-occupied single unit structures, the low level of 290 TABLE 104 RESIDENTIAL MOBILITY AND PARTISAN BEHAVIOR IlSflLOS ANGELES SUBURBS* Political Variables Residency in 1955 „ Elsewhere in Same T * t Los Angeles Moved into House 1958-1960 Democratic registra tion September 1958 -.098 -.215 -.034 September 1960 .120 -.204 -.026 September 1962 .145 -.219 -.031 Partisan elections Brown, 1958 .089 -.221 -.053 Betts, 1958 .096 -.229 -.051 Kennedy, 1960 .156 -.226 -.045 Brown, 1962 .150 -.214 -.051 Rose, 1962 .117 -.182 -.034 *Based on tables 61, 62, and 63. TABLE 105 RESIDENTIAL AND MOBILITY BY SUBURB ANTI-SUBURB* Residency Suburb Suburb Stat. Sig. t Same in 1955 Elsewhere in Los Angeles in 1955 Moved into house 1958-1960 (Figures in Median Percentages) .37 .30 no .43 .42 yes .33 .55 yes *Source: Table 37 tStatistically significant at pi < .05. TABLE 106 RESIDENTIAL MOBILITY BY OWNER-OCCUPANCY* Residency- High Low Decile Decile Stat. Sig. t (Figures in Med. Percentages) Same in 1955 Elsewhere in Los Angeles, 1955 Moved into house, 1958-1960 .47 .31 .38 .42 .30 .54 yes yes yes ^Source: Table 34. tStatistically significant at p. < .05. TABLE 107 RESIDENTIAL MOBILITY BY SINGLE UNIT STRUCTURES* Residency High Low Decile Decile Stat. Sig. t (Figures in Med. Percentages) Same in 1955 Elsewhere in Los Angeles, 1955 Moved into house, 1958-1960 .37 .31 .43 .41 .33 .53 yes yes yes *Source: Table 36. tStatistically significant at p. < .05. TABLEI 108 RESIDENTIAL MOBILITY BY RECENTLY BUILT HOUSING* Residency High Low Decile Decile Stat. Sig. t (Figures in Med. Percentages) • Same in 1955 Elsewhere in Los Angeles, 1955 Moved into house, 1958-1960 .33 .41 .45 .39 .35 .39 yes yes no *Source: Table 35. tStatistically significant at p. < .05. 292 short-term residential mobility is not surprising. Both owner-occupied units and single unit structures would be expected to be occupied by people who do not move fre quently. On the other hand, use of new housing as a criterion for suburbia would be expected to produce a higher level of residential mobility over a somewhat longer period of time. These expectations were fulfilled. Two other variables often associated with suburbia are: (1) the professional and managerial occupations, and (2) middle and high income groups. With few exceptions, mobility did not correlate significantly with either occupation or income. Income Examining first the middle income family and resi dential mobility, the following tables summarize the general lack of any correlation between the two variables. In only two instances out of twelve was there a statistically sig nificant correlation. For the middle income families living in suburbia there was a low positive correlation with resi dence in California outside of Los Angeles in 1955. For the County as a whole there was a low positive correlation between residential mobility and middle income. For high income families, the correlation with mobility is exactly the reverse of that for the middle income families— where there is a statistically significant correlation. 293 TABLE 109 CORRELATION BETWEEN RESIDENTIAL MOBILITY AND FAMILY INCOME, $3,000-$10,000 BY THREE COMMUNITY TYPOLOGIES* Measure of Mobility County Sample Suburb Anti- Suburb Residence same as in 1955 -.058 -.124 -.124 Residence elsewhere in 1955, Los Angeles California -.019 -.049 -.016 .330 + -.057 .061 Moved into residence 1958-1960 .1 54f .067 .050 *Source: Table 48. Statistically significant at p. < .05. TABLE 110 CORRELATION BETWEEN RESIDENTIAL MOBILITY AND FAMILY $10,000 PLUS BY THREE COMMUNITY TYPOLOGIES* INCOME, Measure of Mobility County Sample Suburb Anti- Suburb Residence same as in 1955 .091 .095 .127 Residence elsewhere in 1955, Los Angeles California .008 .040 .054 -.349+ -.012 .300 Moved into residence 1958-1960 -,282t -.066 -.309 *Source: Table 49. +Statistically significant at p. < .05. 294 Occupation For occupation there is also little correlation with mobility. Aside from a low negative correlation in suburbia between residency elsewhere in California in 1955 and pro fessional occupation, and a low positive correlation for the same variables in the County, there is no statistically significant correlation between professional occupation and any of the measures of mobility. TABLE 111 CORRELATION BETWEEN RESIDENTIAL MOBILITY AND PROFESSIONAL OCCUPATION BY THREE COMMUNITY TYPOLOGIES* Measure of Mobility County Sample Suburb Anti- Suburb Residence same as in 1955 -.096 -.028 -.059 Residence elsewhere in 1955, Los Angeles California .012 .194 + .077 -.295t .115 -.119 Moved into residence 1958-1960 -.039 -.008 -.110 *Source: Table 50. +Statistically significant at p. < .05. Looking at managerial occupation there is a similar general lack of correlation with residential mobility. In no instance did mobility correlate to managerial occupation in suburbia. In the anti-suburbs there was a positive correlation to residential stability. For the County, the only 295 significant correlation was a low negative one to recent mobility. TABLE 112 CORRELATION BETWEEN RESIDENTIAL MOBILITY AND MANAGERIAL OCCUPATION BY THREE COMMUNITY TYPOLOGIES* Measure of Mobility County Sample Suburb Anti- Suburb Residence same as in 1955 Residence elsewhere in 1955, .079 .028 . 3371 Los Angeles .024 .116 .017 California .094 -.247 -.229 Moved into residence, 1958-1960 -.220+ -.030 -.3331 *Source: Table 51. tStatistically significant at p. < .05. Long-term stability and political behavior Examination of the variable, residence same as in 1955. produced a few statistically significant correlations with the several different measures of partisan political behavior. There was no correlation to either registration or voting, but there was some to voter turn-out. Long-term stability correlated positively with Democratic and Repub lican voter turn-out within the County. However, there was no correlation within suburbia. Short-term stability and political behavior Use of moved into house 1958-1960 as a measure of 296 short-term mobility did produce several statistically sig nificant correlations within the County but not for suburbia. As short-term mobility increases, both Republican and Democratic voter turn-out decreases. In short, residential mobility does not appear to correlate with partisan political behavior within the suburbs to any appreciable extent. This may be a function of the type of residential mobility operative within the suburbs, wherein such mobility is more "in" than "out." Summary Within the limits of time (1958-1962), the area (Los Angeles), the data (aggregate), the definition of suburbs (which did have political bias), and the analytic tools employed, the following conclusions appear to be justified. First, the suburbs in Los Angeles County are more Democratic than the whole of the County. This is the case whether one uses registration or vote data. Second, that even within the rather rigorous application of definitional criteria two suburbs emerged. One Democratic, with medium income families living in newer houses of less value, and the other Republican, with high income families living in older houses of greater value. Third, the existence of two suburbs suggests that class rather than anticipatory socialization plays the 297 key role in suburban political behavior. Holding class .constant generally reduced or even reversed the observed differences in political behavior between suburbs and anti-suburbso In fact, there appeared to be greater political differences within suburbia than between suburbia and anti—suburbia. Fourth, as a result of the third conclusion, it appears that the segregation thesis rather than the chameleon thesis is most credible. Fifth, residential mobility, as such, does not play a role in suburban political behavior. Needed Research While the focus of this dissertation has been upon the political behavior of suburbs, one of the major questions underlying much of the interest in suburban political behavior remains unanswered. How does social mobility relate to changes in political behavior? It has been shown that change in social status will sometimes be 21 related to changes in political behavior. Operationally, however, there are two major 21 Peter M. Blau, "Social Mobility and Interpersonal Relations," American Sociological Review, XXI, 3 (June, •1956), pp. 290-95; Elston F. Jackson, "Status Consistency and Symptoms of Stress." American Sociological Review. XXVII, 4 (August, 1962), pp. 469-80; and Gerhard E. Lenski, "Status Crys'tallization: A Norf-Vertical Dimension of Social Status," American Sociological Review. XIX, 4 (August, 1954), pp. 405-13. difficulties. First, the definitions of "change"— the change in social status and the change in political behavior. Social mobility, whether indexed by occupation, education, or some other variable, is usually measured over time in one of two ways, inter-generationally or intra- generationally. Natalie Rogoff found that between 60 and 75 per cent of the population are engaged in occupations which are different from those engaged in by their 22 fathers. However, much of this change has been to occupations which differ very little in status from the fathers1 occupations. Rogoff was also of the opinion that no more than half of the population has experienced intra- 23 generational occupational mobility. Seymour Lipset and Reinhard Bendix have stated that no more than one-fourth of the population has experienced intra-generational 24 mobility. These figures suggest that something between one-fourth and one—half of the population have experienced both inter- and intra-generational mobility. Not all of these, however, have experienced changes in political behavior. 22 Natalie Rogoff, Recent Trends in Occupational Mobility (Glencoe: Free Press, 1953), pp. 23-24. 23Ibid.. pp. 25-27. 24 Seymour Lipset and Reinhard Bendix, Social Mobility in Industrial Society (Berkeley and Los Angeles: University of California Press, 1959), p. 165. 299 The second variable, that of political change, offers even greater difficulties„ While the definition of change can be in terms of identification rather than vote or registration, the incidence of that change and its relation ship to social mobility are not at all clear. Angus Camp bell, et al., found 168 occupationally mobile respondents 25 who reported a change in political identification. But, as Campbell, et al.. said: . o „ even among the people who report both a change in partisanship and a change in their own occupational status there is no suggestion of a relationship between the two kinds of c h a n g e . 26 TABLE 113 RELATION OF REPORTED CHANGES IN OWN OCCUPATIONAL STATUS TO CHANGES IN SELF-IDENTIFICATION WITH PARTIES, 1956 Change in Party Identification Occupational Mobility Down Up Republican to Democratic ■ ' t C M 00 C M Republican to Independent 8 12 Democratic to Independent 32 29 Democratic to Republican 32 35 N: 47 121 While the data is inconclusive about the nature of the relationship, it does suggest that for a study to be 25 Angus Campbell, Philip E. Converse, Warren E. Miller, and Donald E. Stokes, The American Voter (New York: John Wiley and Sons, Inc., 1960), pT 459. 26Ibid.. p. 458 300 made of the relationship almost prohibitively large survey samples would be required. This is the second major problem, the validity of data. Assuming the usual SRC sample size of about 1,600 respondents, the 168 respondents tabulated in Table 113 represent about 10 per cent of the population. However, to adequately research the relationship between social mobility and political change would require at least 576 respondents 27 exhibiting such characteristics. Thus, to locate the 576 would require an initial cluster sample of almost 6,000 respondents. Since the essence of such a survey involves changes which may have occurred over a period of some twenty-five to thirty years (from political socialization as a child to political identification in the late thirties or early forties), memories may not prove to be very accurate. Worse, projection or identification could be expected to produce data concerning parental politics which would badly contaminate the data. Use of a panel survey over a period of twenty-five years would require an initial cluster sample 28 of teen-agers of about 60,000. This appears unlikely in the near future. 27 This assumes a tolerated error of 5 per cent at a 95 per cent confidence level. 28 Assuming an annual respondent mortality of 10 per cent, as was found by the SRC, an initial sample of 60,000 301 Perhaps, however, small sample survey techniques applied to unique clusters might prove useful. Such surveys might provide insight into some of the major dimensions of the relationship between social mobility and political change. Among these might well be the types of occupations which are more often associated with political change, or as an intervening variable, with occupational mobility. Perhaps the attitudes which one acquires as a child facilitate or hamper the ability to achieve social mobility and to change political behavior. The impact of secondary groups, particularly in urban areas, is not at all well understood. If such groups are not influential, then the very meaning of social mobility loses an inter personal dimension and would appear to be more of an internal personality characteristic. As long as those who experience both upward and downward occupational mobility can also cross party lines in both directions, there is indeed little to suggest a relationship between the two kinds of change. would have been reduced to about 6,000 twenty-five years later. And, of these 6,000, only about 10 per cent would be "changers." See: Angus Campbell, Gerald Gurin, and Varren E. Miller, The Voter Decides (Evanston: Row, Peterson, and Company, 1954), pT 230, for mortality data. A P P E N D I C E S 302 APPENDIX A MISCELLANEOUS DATA RELATING TO SUBURBAN POLITICAL BEHAVIOR 303 t TABLE 114 PEARSONIAN CORRELATION BETWEEN FOUR STATE-WIDE DEMOCRATIC CANDIDATES 1958, DEMOCRATIC REGISTRATION 1958, ALL LOS ANGELES COUNTY DEMOCRATIC ASSEMBLY CANDIDATES 1958, STEVENSON 1956, AND KENNEDY 1960. (BASED ON ALL THIRTY-ONE ASSEMBLY DISTRICTS IN LOS ANGELES COUNTY) Brown 1958 Ander son 1958 Betts 1958 Engle 1958 • A.D.*s 1958 Registra tion 1958 Steven son 1956 Kennedy 1960 Brown, 1958 .992* .982* .981* .295 .991* .974* .957* Anderson, 1958 .987* .985* .306 .987* .980* .957* Betts, 1958 .975* .316 .974* .975* .951* Engle, 1958 .345 .979* .953* .919* A.D.'s, 1958 .308 .275 .298 Registration, 1958 .956* .927* Stevenson, 1956 .983* *Statistically significant at p. < .05 by t test. u> o TABLE 115 PEARSONIAN CORRELATION BETWEEN FOUR STATE-WIDE DEMOCRATIC CANDIDATES 1958, DEMOCRATIC REGISTRATION 1958, AND FOURTEEN LOS ANGELES COUNTY ASSEMBLY CANDIDATES 1958, STEVENSON 1956, AND KENNEDY 1960 BY ASSEMBLY DISTRICTS. (BASED ON FOURTEEN LOS ANGELES COUNTY ASSEMBLY DISTRICTS WITH DEMOCRATIC INCUMBENTS SEEKING RE-ELECTION)* Brown 1958 Ander son 1958 Betts 1958 Engle 1958 A.D.1s 1958 Registra tion 1958 Steven son 1956 Kennedy 1960 Brown, 1958 .984* .924* .948* -.064 .975* .952* .888* Anderson, 1958 .953* .959* -.035 .971* .960* .873* Betts, 1958 .929* i ■ o v - n .894* .960* .852* Engle, 1958 .060 .955* .898* .775* A.D.'s, 1958 -.042 -.056 -.005 Registration, 1958 .896* .794* Stevenson, 1956 .950* ^Successful cross-filers included. *Statistically significant at p. < .05 by t test. TABLE 116 PEARSONIAN CORRELATION BETWEEN FOUR STATE-WIDE DEMOCRATIC CANDIDATES 1958, DEMOCRATIC REGISTRATION 195&, ELEVEN LOS ANGELES COUNTY DEMOCRATIC ASSEMBLY CANDIDATES 1958, STEVENSON 1956, AND KENNEDY 1960. (BASED ON THE ELEVEN LOS ANGELES COUNTY ASSEMBLY DISTRICTS IN WHICH THERE WAS A DEMOCRATIC INCUMBENT SEEKING RE-ELECTION WHO HAD NOT SUCCESSFULLY CROSS FILED IN THE PRIMARY) Brown 1958 Ander son 1958 Betts 1958 Engle 1958 A.D.*s 1958 Registra tion 1958 Steven son 1956 Kennedy 1960 Brown, 1958 .982* .911* .965* .943* .981* .951* .905* Anderson, 1958 .944* .983* .942* .983* .957* .871* Betts, 1958 .946* .830* .895* .956* .838* Engle, 1958 .921* .959* .935* .830* A.D.s, 1958 .955* .871* .791* Registration, 1958 .917* .841* Stevenson, 1956 .946* ♦Statistically significant at p. < .05 by t test. 306 TABLE 117 PEARSONIAN CORRELATION BETWEEN FOUR STATE-WIDE DEMOCRATIC CANDIDATES 1958, DEMOCRATIC REGISTRATION 1958, FIFTEEN LOS ANGELES COUNTY DEMOCRATIC ASSEMBLY CANDIDATES 1958, STEVENSON 1956, AND KENNEDY 1960. (BASED ON FIFTEEN LOS ANGELES COUNTY ASSEMBLY DISTRICTS WITH REPUBLICAN INCUMBENTS SEEKING RE-ELECTION) Brown 1958 Ander son 1958 Betts 1958 Engle 1958 A.D.'s 1958 Registra tion 1958 Steven son 1956 Kennedy 1960 Brown, 1958 .981* .988* .983* .957* .990* .954* .942* Anderson, 1958 .997* .982* .971* .974* .972* .963* Betts, 1958 .982* .976* .979* .967* .959* Engle, 1958 .943* .971* .934* .924* A.D.'s, 1958 .951* .971* .953* Registration, 1958 .945* .915* Stevenson, 1956 .985* *Statistically significant at p. < .05 by t test. 307 308 TABLE 118 COMPARISON OF CITT AND SUBURBAN PRESIDENTIAL VOTE IN THIRTEEN SELECTED AREAS, 1952-1960+ Area Democratic 1952 Percentage of 1956 Two-Party Vote 1960 Baltimore City 51 .7 44.1 64.0 Suburbs 37.5 32.6 49.7 Boston City 59.6 53.6 75.0 Suburbs 41 .4 38.2 59.1 Buffalo City 49.6 42.3 65.1 Suburbs 36.5 30.5 48.7 Chicago City 54.4 48.7 63.7 Suburbs 36.8 30.9 41 .1 Cleveland City 59.9 54.6 70.9 Suburbs 36.5 37.7 50.1 Detroit City 60.5 61 .8 66.0 Suburbs 45.6 46.5 52.0 Minneapolis-St. Paul City 52.6 50.8 55.5 Suburbs 43.9 43.3 45.3 ^Source: Research Division, Democratic National Committee, "The 1960 Election Report" (Washington, D.C.: Mimeograph, 1962), Appendix, Table 8. 309 TABLE 118— Continued Area Democratic 1952 Percentage of 1956 Two-Party Vote 1960 New York City 55.4 51 .0 62.9 Suburbs 37.6 31 .2 47.6 Philadelphia City 58.4 57.0 68.2 Suburbs 40.7 37.2 43.4 Pittsburgh City 56.1 52.3 67.0 Suburbs 51 .5 45.3 50.9 St. Louis City 62.0 60.9 66.7 Suburbs 51 .4 50.5 51 .1 San Francisco City 47.0 48.2 58.3 Suburbs 45.7 45.6 51 .1 Los Angeles City 47.9 49.1 54.0 Suburbs 39.5 40.1 41 .0 310 TABLE 119 RELATION OF TYPE OF COMMUNITY TO PARTY IDENTIFICATION* Party Identification Metropolitan Centers Suburbs City or Town Open Country Strong Democrat 22% 1 5 % 22 25?S* Weak Democrat 27 32 25 21 * Independent Democrat 11 9 7 10 Independent 7 5 8 7 Independent Republican 5 9 7 3 Weak Republican 11 9 15 18 * Strong Republican 12 * 20 1 2 10 * Apolitical, DK 5 1 4 6 Number of cases: 169 150 601 219 *Statistically significant at p. < *05 in comparison to the column "suburbs." ^Source: Campbell, Angus and Homer C. Cooper, Group Differences in Attitudes and Votes (Ann Arbor, Michigan: University of Michigan Press, 1956), p. 46. 311 TABLE 120 SUBURBAN VOTE 1948 BY AREA* Area Truman Dewey Didn* t Vote N* Northeast 33 < f o 00 25?S 81 Northwest 21 61 14 28 South 0 0 100 1 Vest 19 40 37 62 ■^Miscellaneous votes omitted from tables, but included in calculation of percentages. V TABLE 121 SUBURBAN VOTE 1952 BY AREA* Area Steven son Eisen hower Didn1t Vote N* Northeast 34 io 45 is ^ 77 Northwest 24 72 3 29 South 0 0 100 1 Vest 23 40 37 30 *Miscellaneous votes omitted from tables, but included in calculation of percentages. Source: Alfred de Grazia, The Western Public (Stanford: Stanford University Press'^ 1954), pT 165. 312 TABLE 122 SOUTHERN VOTING PATTERNS IN CENTRAL CITY AND SUBURBS, 1948-1956t Area Democratic Republican* ^Vote^ Central city 17# Presidential (1948) 5# 7 8fo Suburb 29 21 50 Central city 22 Presidential (1952) 21 58 Suburb 31 24 45 Central city 30 Congressional (1952) 3 68 Suburb 31 7 62 Central city 14 Presidential (1956) 33 53 Suburb 22 31 46 Central city 29 Congressional (1956) 13 58 Suburb 32 14 53 *Includes other minor party candidates. ^Source: Bernard Lazarvitz, "Some Characteristics of Presidential Belts in the Metropolitan Community," (unpub lished Ph.D, dissertation, University of Michigan, 1958), pp. 68-71. 313 TABLE 123 VOTER PARTI PREFERENCE, 1956 PRESIDENTIAL ELECTION* Party Preference: Catholics Central City Suburbs Democratic 82.5$ 7 4 . 2 % Republican 17.5 25.8 N: 131 341 y ? = 3.74* Party Preference: Non-Catholics Central City Suburbs Democratic 66.6$ 48.5$ Republican 33.4 51.5 N: 264 641 y 2 = 26.34* *No statistically significant difference between Suburbs and Central City for Catholic Democratic Party preference, but a very significant difference for non— Catholics. + Source: Scott Greer, "Catholic Voters and the Demo cratic Party," Public Opinion Quarterly. XXV (Vinter, 1961) 4, p. 616. Table data reorganized and modified due to error in original table. 314 TABLE 1 24 VOTER CANDIDATE PREFERENCE, 1956 PRESIDENTIAL ELECTION* Catholics Candidate Preference Central City Suburbs Democratic (Stevenson) 74.7% 54.8% Republican (Eisenhower) 23.3 45.2 N: 103 % 2 = 290 12.35* Candidate Preference Non-Catholics Central City Suburbs Democratic (Stevenson) 56.2% 41 .4% Republican (Eisenhower) 43.8 58.6 N: 185 * t o , I I 496 12.07* *Candidate preference statistically significant between Central City and Suburbs at p. < .001. 'Source: Scott Greer, "Catholic Voters and the Demo cratic Party," Public Opinion Quarterly. XXV (Vinter, 1961) 4, p. 616. " V 315 TABLE 125 OCCUPATION BY COMMUNITY OF ORIENTATION^ Parent Occupation Farm Rural Non- Farm and Urban to 250,000 Urban Pop. of 250,000 Plus Professional 5 fo 11^ 6fo Self-employed 16 1 2 16 Upper white collar 4 * 8 13 * Lower white collar 11 * 15 22 * Skilled 28 21 20 Semi-skilled 19 17 1 2 Unskilled 12 * 9 3 * Non-manual 41 * 53 * 65 * Manual 59 * 47 * 35 * N: 108 300 434 *Statistically significantly different for other values*in row at p. < .05. 4* 'Seymour M. Lipset and Reinhard Bendix, Social Mobility in Industrial Society (Berkeley and Los Angeles: University of California Press, 1959), p. 205. 316 TABLE 126 PERCENTAGE OF TIME SPENT IN PRESENT OCCUPATIONAL CATEGORYt Present Occupation N Per Cent of 80-1 Q O ’ fo 50-749$ Time Under 50?$ Professional 23 70 9 22 Semi-professional 19 47 32 21 Own business 105 11 31 57 Upper white collar 72 14 21 65 Lower white collar 67 18 33 49 Sales 42 26 24 50 Skilled worker 169 22 35 43 Semi-skilled 98 22 29 49 Unskilled 44 18 21 61 JSeymour M. Lipset and Reinhard Bendix, Social Mobility in Industrial Society (Berkeley and Los Angeles: University of California Press, 1959), p. 161. 317 TABLE 127 PERCENTAGE OF TIME SPENT IN OTHER THAN PRESENT OCCUPATIONAL CATEGORY? Present Occupation N Percentage of Time Manual Non-Manual Professional 23 6 0 Semi-professional 19 13 0 Own business 1 05 26 0 Upper white collar 72 10 0 Lower white collar 67 30 0 . Sales 42 21 0 Skilled 169 0 9 Semi-skilled 98 0 14 Unskilled 44 0 13 f Seymour M. Lipset and Reinhard Bendix, Social Mobility in Industrial Society (Berkeley and Los Angeles: University of California Press, 1959), p. 170. APPENDIX B MAP OF SUBURBAN AND ANTI-SUBURBAN TRACTS IN LOS ANGELES COUNTY 318 r Sa n t a Ca t a u n a j t , 1 . . j INSERT r O '* * M il N ‘ * I INSERT ® INSERT © |_!W— f • I ; * / V A * ' > / COUNTY |L ‘ 1 k f tT A O O D l ________I h.. 1 ft© .' * / < X 'V | V I# » T A M l t / M A lM 'H l » « V > i C « S rO * tO * T A " l cmatj«o*tn j t ' C M H I M U ' M u a N i ^ * mw h _.r a f l f i m u r T a» ro > o •* 1 » 7 N | M i • I V I O * !«■■■■» •M tM i ^ > A ? I C 0 l 21J V * f t O « k N « V t * vicToat £1 139 » % r i» A N b * -« D l A M v l N T u t A I c v o O P l i M | s V A A M l iI USJ IJ9* A a * . C > « 0 y ? 7imi I 4J1 • * |< C II( \ O A C A O S I O • I S 1 0 A A w n L * " » 1 4 I t ‘ " 7 7 3 * * A J J ) nTJM I L V O DUX CtNICN 14 it »4«»m» ^ L.. i Q n « n i i r U d l A L U f t l K I U * r l o i A ngelas Los Feliz 900 900 01-67 27 Agouro Lynwood 902 62-63 913 01 El Sagundo 902 45-46 Lugo 900 23 Alhambro 918 01-03 El Sarano 900 32 Main Office 900 12 Altadeno 910 01-03 Encino 9>13 16-17 Main Office 2 900 13 Arcadia 910 06.08 Flint 900 57 Malibu 902 65 A rtaiia 907 01 F lorence Branch 900 01 Manhattan Beach 902 66*68 A tcot 900 03 Foy 900 17 Market 900 2V A zuig 917 02-03 Gardena 902 47-49 Mar V ista 900 66 Baldwin Park 917 06-08 G lassell 900 65 Maywood 902 70 Barrington 900 49 G lendale 912 01-08 Metro 900 14 Ball 902 01-04 Glendoro 917 40 Monrovia 910 16-17 Bellflower 907 06 Green 900 37 Montebello 906 40-41 Beverly Hi 111 902 10-13 Greenmead 900 59 Monterey Park 917 54-55 Boy la 900 33 Griffith 900 39 Montrose 910 20 Braa 926 21 Hancock 900 44 North Hollywood 916 01-08 Briggs 900 48 Harbor City 907 10 Northridge 913 24-26 Burbonk 915 01-06 Hawthorne 902 50-52 Norwalk 906 50-52 C alo b asas 913 02 H aiard Branch 900 63 Oakwood 900 04 Canogo Park 913 03-06 Herm osa B each 902 54 P acific P a lisa d e s 902 72 Chatsworth 913 11 Highland Park 900 42 Pocoima 913 31 Cimorron 900 18 Hollywood 900 28 Palms 900 34 Claremont 917 11-18 Huntington Park 902 55-58 Palos Verdes P en in su la 902 74-75 Cola Branch 900 46 Inglewood 903 01-10 Paramount 907 23-24 Compton 902 20-22 Kearny 900 11 P asad en a 911 01-08 Cornell 913 15 La C anada 910 11 Pico Heights 900 06 Covina 917 22-24 La C rescento 910 14 Pico Rivera 906 60-62 Crenshaw 900 08 La Habra 906 31-33 Pomono 917 66-69 Culver City 902 30-34 Lakewood 907 12-14 P re u ss 900 35 Del V alle 900 15 La Mirada 906 38 Rancho Park 900 64 Dockwailer 900 07 La Puente 917 44-47 43 Redondo Beach 902 77-78 Downey 902 40-42 La Ti j era 900 R eseda 913 35-36 Duarte 910 10 La Verne 917 50 Rimpau 900 19 Eagle Rock 900 41 Law ndale 902 60 Rosemead 917 70-71 Eost L os Ange les Branch 900 22 Lincoln H eights 900 31 San Dimas 917 73 Edendale 900 26 Lomita 907 17 San Fernando 913 40-44 El Monte 917 31-34 Long Beach 908 01-15 Sanford 900 05 roatST / / / H'UNIO *1001 »i*l !»UC» TMil r , , *'IOO UNION no "I Mt \ s S U B U R B I A W n - S U W A w flV 1431 l a o M u r ur* |«4i (X-<\ • wmumi I \ w J ®vvflr\ : , S i Pi® IlL-MI / ] * ' i 5 ^ I *v\ \ J i i i % A m / 3 . > " k N •J 7 1 5 H rouWiJP s '\\4 ~ 2 n ,. Ji 8 K Toe h n U t l N O M . 1 M O O *U » 2»21 V c * « o i . < r o o c I H 4 1 0 2 ^ - r\ H tH O vti 21 i t j » ? H S ItMiWTON U t V t T U M 'J fKILrt' WN7J7H ••til M “ '//j ‘ 1 ,jV \ F/<tXS/\i>VJJ M r* ♦ / v . . a I # . X w v \ * * * fclnr'** <oj S* 2iH K2i«2 H« * « « * > 2 2 1 ) 1 2 2 ) 2 twuf»uw« \ w M ’ M 7 i f S M # & * ’ / * * * / y v f v t A V V ^ s i i f e S & i tut x m i 2U2 >* 4 \ / ✓ <s, J C ) A t nr \ * A A J*"W »N -L»~-lm n> l j “ * a \4 < V . m < ‘ ' & 3 & \ / W i y> , / \ X , :6nR rK *;r ’ C I J X / V A V ' q V' V ro“ ’ • *?JU kii) »uw twcnww 234) 4 2324 Mil It \ T 0 3 I L irik) X f - NOlMt lOK',--* 2341 W I I tI M 2173 2177SH* JHMHFI* J 1 . 2 . 3 ® / L u , m ! 6013*1 ecu im' imci Q ^ W 1 I 1 JlU J i ~ . lM f t « l» l 6001 I jmwiik i * : U 3W»M ll tt& U tl V > 0 1 4 I A K* V ' « ) f . i j j ! I .. .. I ,-Aww* ’ V J 1 , , . « o t 8 ! L j 4640 f*OLM6»> ini ' 1 T 1 j A.y~U ttUJffi. IAiJ11 - /<! 4ICi Jf ' * t M ' l i ' O I : A i»4*ri y.»«V\ L; 4 8 0 1 ■ m n ' ^ i s V \ f H J H , I N t 4 T 0 N 3 f >/ V- ' -• i _ ,. „ j \ J t j m l i W n c — V ^ ' T ' L 5 f f K C * * f r * I } * P"* f ««4 »fl«y i »/ jj A w \ * /[y y/^*L , l “ .vv"> * . * k > " J; ; m.^ i j i J ^; 4 \ „ • ' ' • $ ? . a , y M y . * / A x P l . M ^ ./=*!-«4L!.y. • si r * r s n jimilCJ LONOOIN 1)14 I !iu»5__ WcwiDt f" I l-*IT | I I 4041 \ \ 4051 J t*a»/^r»w " L f t M L 0 | 0 M * i f 1 * 2 ) W N t T N ( uAiruTui' mahoni Ah A O * t » L l U r r - . * ^ 1 | \ Van . M r i y lm h^ npC 5 ) 0 2 I H * W ( M L m . n j ..,, ^ h ■ 1 - F i r iB"V«"“ ? / \ v . t i 4 C M y f /«\i i " * IdimumnV IzApcf rr^Z \ I •• J"U3'3 * ’ f irT;/ / j ; l-UrA$ t : -J l » T V "A 7 r/ x-w . * 4071 MjJ S A N & * » •[, » * , HUDSONJ 5300 \ * 5 0 0 ) 1 * » w r * !-y / { . »J» 4 4064 6004 5 0 0 ) „ * 0 4 « M « N M I L L / ♦ , - / 6 jj . N *\ 6J * 1 , ' I \ NV i Il & T E K n . 6WQi«t« s ° ' o <r #j4 5 J ) 8 P t l T U l l IMlnol 6021 'C C N tU 4r *»*F* 6)61 0 1 ) V ivid/ ' * 4 0 4 I » u 6 ) 4 \i c k i i i m r\i / L< 0 *:v- ' V ^ ! M | T H O » O L I T » l l 0 M { 6 0 1 6 j | ! in n n n \ M U 4 V ! | T i 6016 j, -A0 1 4 I 5>, * V \ -A iVi N M * 4 V I 1 T * 6 0 0 2 5 0 1 6 0 2 6 r I lM I I t 504i 50)6 :! % * 0 0 0 'PttTHI-U » < - * r L 9 COuUI W A V 1*11 t Hi ItM LIN l V ' 0 A L T O N * A S H » O O T M H L C . L A O b T O N ( If 0 fjoTNILL A R R O W » I A 0 H C L L C M U > “\ - A ,v j • u O D l » C S T O « f J ' V X J » 4020 r»K*lr OSANCt < 4065 lima Oilton 4 0 t t > DOUlLCCHOVI ■ ' < - ^ r m K C i i f l u i r o 4025 « liCAOVt 4 0 3 2 ^-^7 m O S M A I unv"" ■ r < o lNAM(ltf PUt*H / ZIP CODES C o u n t y 9 0 0 0 1 - 6 7 L o s A n g e le s 7 2 P a c i f i c P a l i s a d e s 01 F l o r e n c e Bronch 7 4 - 7 5 P a l o s V e r d e s P e n i n s u l a 02 Watts 7 7 - 7 8 R e d o n d o B e a c h 0 3 A s c o t 8 0 - 8 3 S outh G a t e 04 O ak w ood 9 0 T o p a n g a 05 Sanford 9 1 - 9 3 V e n i c e 06 P i c o H e ig h ts 9 0 3 0 1 - 1 0 I n g le w o o d 07 D o e k w e ile r 9 0 4 0 1 - 0 6 Santa M o n ic a 08 C ren shaw 9 0 5 0 1 - 0 8 T o r r a n c e 11 Kearny 9 0 6 0 1 - 0 8 W hittier 12 Main O f f ic e 3 1 - 3 3 ' L a H abra 13 Main O f f i c e 2 38 L a Mirada 14 Metro 4 0 -4 1 M o n te b e llo 15 D el V a lle 5 0 -5 2 N o rw a lk 16 West A d am s 6 0 - 6 2 P i c o R iv e r a 17 Foy 70 S anto F e S prin gs 18 Cimarron 9 0 7 01 A r t e s ia 19 Rimpau 0 6 - 0 7 B e l l f l o w e r 21 Morkel 10 Harbor C ity KERN COUNTY L « i 4 f Q b i ' i h a m h O w h q a ‘ 1 ' EOWAROS AIR 9001 9003 L A N C A S T E R ; 9002 9012 9010 1004 9011 9 1 0 ) ' 9101 PAIMQALE QUARTZ H ILL 9102 . 1 1 0 1 9 1 0 0 , >1-1 )'■ l H i . X I H , ; it j— _ 1201 9107 ^ LITTLEROCK PEAF 9200 9109 9101 SAUGUS NEWHALL 9201 9300 N O R T H E R N P A R T OF LOS A N G E L E S C O U N T Y FO R C E s . i . 1 1 1 an.n i t inf 9001 tl!0 v,M02 ! 1 0 " * * < V £ JSltW SL-..J i l l l T H I " i • 1 0 0 * 0 0 I f •130*00J IIKH ■ HUM U C 4 S ' COmMoTI v 3 •* fiC H u uotj I * w 4 i tw i t j mooNoo | \s ^ ujjw- ^ ' t * ! I > 0 1 I . 6505 uu C H A N N E L rruuHfmL ... 1 r HWIMl W )uon AliiO ■iiiiu1 !!! I I I H 1 • ILIayrlK , r * IIHH1*’' »■»» j * ’ i - ■■■ in* mi i c w c x - ' 1 m ( fit IlllJn m D S N3 D )l»l " V I * ■ IV IB t mill 1 0 ion «>»’■ » 1 0 ’ afU ? ? / « < / / ■ ■ v r * k- om nart v, v o i i < : xjtallf 9 0 / ^ r * # f n * - — V H B 1 M } ^ — ; t m n I (TAG I iNpi. * l * d gS>b f ’ t ’ i S ^ I 0 f’ j s N O lO N IM H * R rtlHOTi ■ r M * | l Idonooii voianO; J S TRACT Boundaries and NUMBER^ OS ANGELES COUNTY COPYRIGHT 1965 by BREWSTIR MAPS LOS ANGELES MAP No. 65 0 107 19 21 22 23 24 25 26 27 28 29 3! 32 33 34 35 36 37 38 39 41 42 43 44 45 46 47 48 49 56 57 58 59 61 62 63 64 65 66 6 7 69 9 02 01-04 10-13 20-22 30-34 4 0 -4 2 4 5 -4 6 4 7 -4 9 5 0 -5 2 54 5 5 -5 8 60 6 2 -6 3 65 66-68 70 Rtnipau Moiktt East L, A. Branch Lugo Village W ait L. A. E d en d ale L o t F e liz Hollywood Vermont Avo. L in c o ln H e ig h ts El Soreno Boyle Palm s P r e u s s W tlshire-Lo Brea Green Wilcox Griffith E ag le Rock H ighland Pork York La T ijera Honcock W estch ester C ole Branch Wagner Briggi Barrington Flint Vernon Branch G reenm ead South Westvern H a ia r d Bronch R ancho Park G l a s s e d Mar V ista West Branch Bell Beverly H ills Compton C ulver City Downey El Segundo Gardena Haw thorne Herm osa Beach H untington Park L aw n d ale Lynwood Malibu Manhattan Beach Maywood 06-07 10 12-14 17 2 3 -2 4 3 1 -3 3 4 4 -4 6 90 8 0 1 -1 5 9 1 0 0 1 -0 3 0 6 -0 7 10 11 14 16 20 24 30-31 40 42 46 911 0 1 -0 8 9 1 2 0 1 -0 8 9 1 3 01 02 03-06 I I 15 16-17 2 4-26 31 35-36 40 -4 4 5 2 -5 3 56 64 -6 6 9 14 0 1 -0 9 9 1 5 0 1 -0 6 9 1 6 01-08 9 1 7 0 2 -0 3 0 6 -0 8 11-18 2 2 -2 4 3 1 -3 4 4 0 4 4 -4 7 5 0 5 4 -5 5 6 6 -6 9 70-71 73 7 5 -7 8 9 0 -9 2 9 1 8 0 1 -0 3 9 2 6 21 Bellflow er Harbor City Lakewood Lomita Param ount Son Pedro Wilmington Long Beach A ltadena A rcadia D uarte La C anada La C re s c e n ta Monrovia M ontrose Sierra Madre South P a s a d e n a Sun land T ujunga Verdugo C ity P a s a d e n a G lendale Agoura C a l o b a i o i C anoga Park C hatsw orth Cornell Encino N orthridge Pocoim a San Fernando Sun Valley T o rta n a Woodland H ills Van Nuys Burbank North Hollywood A zu sa Baldwin Park Claremont Covina El Monte Glendora L a P uente La Verne Monterey Park Pomona R osem ead San Dimas San Gabriel Tem ple C ity Walnut West C ovina Alhambra Brea 47 B I B L I O G R A P H Y 320 4 e BIBLIOGRAPHY A. Books Adrian, Charles R. Governing Urban America. New York: McGraw-Hill Book Company, Inc., 1961. Alford, Robert R. Party and Society. Chicago: Rand McNally and Company, 1963. Anderson, H. Dewey, and Davidson, Percy E. Ballots and the Democratic Class Struggle. Palo Alto: Stanford University Press, 1943. Baltzell, E. Digby. Philadelphia Gentlemen: The Making of a National Upper Class. Glencoe, 111.: The Free Press, 1958. Banfield, Edward C. Political Influence. Glencoe, 111.; The Free Press, 1961. Barber, Bernard. Social Stratification. New York: Harcourt Brace and Company, 1957. Bell, Daniel. The New American Right. New York: Criterion Books, 1955. __________. The End of Ideology. New York: The Free Press of Glencoe, 1960. Bendix, Reinhard, and Lipset, Seymour M. Class. Status and Power. Glencoe, 111.: Free Press, 1953. Benedict, Ruth. Patterns of Culture. Boston: Houghton and Mifflin Company, 1934. Bentley, Arthur F. The Process of Government. Chicago: University of Chicago Press, 1908. Berelson, Bernard R., Lazarsfeld, Paul F., and McPhee, Villiam N. Voting. Chicago: University of Chicago Press, 1954. 321 322 Berger, Bennett M. Working Class Suburbs. Berkeley and Los Angeles: University of California Press, 1960. A very useful study of the changes which took place in the attitudes and behavior of Ford motor car workers after they followed their plant to suburbia. The findings support the segregation thesis. Boskoff, Alvin, and Zeigler, Harmon. Voting Patterns in a Local Election. Philadelphia: J. P. Lippincott and Company, 1964. Brazer, Harvey E. Some Fiscal Implications of Metropoli- tanism. Reprint Series No. 61. Washington, D.C.: The Brookings Institution, 1962. Burdick, Eugene, and Brodbeck, Arthur. American Voting Behavior. Glencoe, 111.: The Free Press, 1959. Campbell, Angus, Gurin, Gerald, and Miller, Warren. The Voter Decides. Evanston, 111.: Row, Peterson and Company, 1954. _________, and Cooper, Homer C. Group Differences in Attitudes and Votes. Ann Arbor; University of Michigan Press, 1956. _________, Converse, Phillip E., Miller, Warren E., and Stokes, Donald E. The American Voter. New York: John Wiley and Sons, Inc., 1960. Cantril, Hadley. Public Opinion 1935-19 6. Princeton, New Jersey: Princeton University ?ress, 1951. Cartwright, Dorwin, and Zander, Alvin (t s.). Group Dynamics. Evanston, 111.: Row, eterson and Company, 1953. Centers, Richard. The Psychology of Social Class. Princeton, N.J.: The Princeton University Press, 1949. A pioneer work in the social-psychology and multidimensions of class. Paired with Eulau’s Class and Party in the Eisenhower Years, the two are must readings for an understanding of the contemporary approach to social class and political behavior. Chinoy, Ely. Automobile Workers and the American Dream. New York: Random House, 1955. 323 Cohen, Lillian. Statistical Methods for Social Scientists. New York: Prentice Hall, Inc., 1954. Cooley, Charles H. Social Organization. New York: Charles Scribner's Sons, 1909. Corey, Lewis. The Crisis of the Middle Class. New York: Covici-friede, 1935. Cox, Oliver. Caste, Class, and Race: A Study in Social Dynamics. New York: Doubleday and Company, 1948. Dahl, Robert A. ¥ho Governs? New Haven; Yale University Press, 1961. Dahrendorf, Ralf. Class and Class Conflict in Industrial Society. Stanford: Stanford University Press, 1959. [Originally published as: Soziale Klassen und Klassenkonflikt in de industriellen Gesellschaft in 1957. Translated and revised by the author.j Davidson, Percy E., and Anderson, H. Dewey, Occupational Mobility in an American Community. Stanford: Stanford University Press, 1937. De Grazia, Alfred. The Western Public: 1952 and Beyond. Stanford: Stanford University Press, 1954. Dewey, John. Human Nature and Conduct: An Introduction to Social Psychology. New York: Henry Holt and Company, 1922. __________. The Public and Its Problems. New York: Henry Holt Company and Inc., 1927. ' Douglas, Harlan Paul. The Suburban Trend. New York: The Century Company, 1925. Duncan, Otis D., Cuzzort, Raymond P., and Duncan, Beverly. Statistical Geography. Problems in Analyzing Areal Data. Glencoe, 111.: The Free Press, 1961. A comprehensive presentation of the methodology and limits of areal analysis. This is indispensible to anyone working with areal aggregates of data. Eulau, Heinz. Class and Party in the Eisenhower Years. New York: The Free Press of Glencoe, 1962. Following up Centers' examination of class and strata, Eulau attempts to assess the impact of objective and subjective class upon political attitudes and behavior. 324 Fenton, John H. The Catholic Vote. New Orleans: The Hauser Press, 1960. Festinger, Leon, Schacter, Stanley, and Back, Kurt. Social Pressures in Informal Groups. New York: Harper and Brothers, 1950. _________, and Kelley, Harold. Changing Attitudes Through Social Contact. Ann Arbor: Research Center for Group Dynamics, 1951. Fromm, Eric. Escape from Freedom. New York: Farrar and Rinehart, Inc., 1941. Gerth, Hans H., and Mills, C. Wright. From Max Weber’s Essays in Sociology. New York: Oxford University Press, 1945. Ginsberg, Morris. Studies in Sociology. London: Methven and Company, 1932. Gordon, Albert I. (Rabbi). Jews in Suburbia. Boston: Beacon Press, 1959. Gottman, Jean. Megalopolis. New York: Twentieth Century Fund, 1961. A major contribution to our understanding of the ultimate meaning of urbanism. Megalopolis is a continuous "city" which runs down the Atlantic sea coast from Boston to Charleston. _________. Economics« Esthetics, and Ethics in Modern Urbanization. New York: Twentieth Century Fund, 1962. Government Affairs Foundation. Metropolitan Surveys: A Digest. Chicago: Public Administration Service, 1958. Grayson, Henry. The Crisis of the Middle Class. New York: Rinehart and Company, Inc., 1955. Greer, Scott. The Emerging City. Glencoe, 111.: The Free Press, 1962. _________. Governing the Metropolis. New York: John Wiley and Sons, 1962. Gulick, Luther H. The Metropolitan Problem and American Ideas. New York: Alfred A. Knopf, 1962. 325 . Harris, Lou. Is There a Republican Majority? New York: Harper and Brothers, 1954. Hatt, Paul K. and Reiss, Albert J., Jr. (eds.). Reader in Urban Sociology. Glencoe, 111.: The Free Press, 1951 . Hauser, Philip M. Population Perspectives. New Brunswick, N.J.: Rutgers University Press, 1960. Hawley, Amos A. Human Ecology: A Theory of Community Structure. New York: Roland Press, 1950. Hays, William L. Statistics for Psychologists. New York: Holt, Rinehart, and Winston, 1963. Herring, Pendleton. Group Representation Before Congress. Baltimore: The Johns Hopkins Press, 1929. Holcomb, Arthur N. The Middle Class in American Politics. Cambridge, Mass.: Harvard University Press, 1940. _________. Our More Perfect Union. Cambridge, Mass.: Har vard University Press, 1950. Hollingshead, August D., and Redlich, Frederick C. Social Class and Mental Illness. New York: John Wiley and Sons, Inc., 1958. Horney, Karen. The Neurotic Personality of Our Time. New York: Norton, 1937. Hunter, Floyd. Community Power Structures. Chapel Hill: University of North Carolina Press, 1953. Janowitz, Morris (ed.). Community Political Systems. Glencoe, 111.: The Free Press, 1961. Jonas, Frank H. (ed.). Western Politics. Salt Lake City: University of Utah Press, 1961. Kahl, Joseph A. The American Class Structure. New York: Rinehart and Company, Inc., 1957. Katz, Elihu and Lazarsfeld, Paul F. Personal Influence. Glencoe, 111.: The Free Press, 1955. Key, V. 0., Jr. Politics, Parties, and Pressure Groups. 4th ed. New York: Thomas Y. Crowell Company, 1960. 326 Kornhauser, Arthur, Mayer, Albert J., and Sheppard, Harold. Vhen Labor Votes. New York: University Books, 1956. Lane, Robert. Political Life. Glencoe, 111.: The Free Press, 1959. * Lazarsfeld, Paul F., Berelson, Bernard, and Gaudet, Hazel. The People's Choice. New York: Columbia University Press, 1955. Linton, Ralph. The Study of Man. New York: Appleton- Century-Crofts, Inc., 1936. ________ , The Cultural Background of Personality. New York: Appleton-Century-Crofts, Inc., 1945. Lippman, Valter. Public Opinion. New York: Harcourt, Brace, and Company, 1922. Lipset, Seymour. Political Man. New York: Doubleday and Company, 1960. ________ , and Bendix, Reinhard. Social Mobility in Industrial Society. Berkeley and Los Angeles: University of California Press, 1959. An insightful use of data to analyze the impact of industrialization upon social mobility. % • Lubell, Samuel. Revolt of the Moderates. New York: Harper and Brothers, 1956. Lundberg, George A., Komarovsky, Mirra, and Mclnerny, Mary Alice. Leisure: A Suburban Study. New York: Columbia University Press, 1934. One of the first "contemporary" studies of suburban life. It is concerned with suburbia as a way of life. Maclver, Robert. Society. New York: Farrar and Rinehart, Inc., 1937. ________ , and Page, Charles H. Society, an Introductory Analysis. New York: Rinehart and Company, 1949. Published originally in 1937 by Maclver as Society. Mantoux, Paul. The Industrial Revolution in England in the Eighteenth Century. London: Jonathan Cape, 1928. Marshall, T. H. (ed.). Citizenship and Social Class. Cambridge, Mass.: Harvard University Press, 1950. 327 Marvick, Dwaine (ed.). Political Decision Makers. Glencoe, 111.: The Free Press, 1961. McDonald, Neil A. The Study of Political Parties. Gafden City: Doubleday and Company, Inc., 1955. McPhee, William, and Glaser, William. Public Opinion and Congressional Elections. 'New York: The Free Press of Glencoe, 1962. Mead, Margaret. Coming of Age in Samoa. New York: W. Morrow and Company, 1928. Mills, C. Wright. White Collar. New Yorkj Oxford Univer sity Press, 1956. An examination of that class of workers who are supposed to populate suburbia. Mumford, Lewis. The City in History. New York: Harcourt, Brace, and World, Inc., 1961. A modern tour de force which examines not just the physical growth of the city in history, but more significantly the place of the city in man’s life. Myers, Jerome K., and Roberts, Bertram H. Family and Class Dynamics in Mental Illness. New York: John Wiley and Sons, Inc., 1959. Newcomb, Theodore M. Social Psychology. New York: Dryden Press, 1950. _________, Eugene L. Hartley, and Maccoby, Nathan. Readings in Social Psychology. New York: Henry Holt and Company, 1947. Park, Robert E., Burgess, Ernest W., and McKenzie, Roderick D. The City. Chicago: University of Chicago Press, 1925. Presthus, Robert. The Organizational Society; An Analysis and a Theory. New York: Alfred A. Knopf, 1962. Queen, Stuart, and Carpenter, David B. The American City. New York: McGraw-Hill Book Company, Inc., 1953. Rischin, Moses. Our Own Kind. Santa Barbara: Fund for the Republic, 1960. 328 Roe, Anne. The Psychology of Occupations. New York: John Wiley and Sons, Inc., 1956. An interesting but unsuccessful attempt to evaluate the relationship between personality and occupation. Roethlisberger, F. J., and Dickson, William J. Management and the Worker. Cambridge, Mass.: Harvard Univer sity Press, 1939. Rogoff, Natalie. Recent Trends in Occupational Mobility. Glencoe, 111.: The Free Press, 1953. Rosenberg, Morris. Occupations and Values. Glencoe, 111.: The Free Press, 1957. Rossi, Peter. Why Families Move: A Study in the Social Psychology of Urban Residential Mobility. Glencoe, 111.: The Free Press, 1955. A study of why people leave one residence or community for another residence or community. It is concerned more with the physical comforts and/or advantages of moving than with aspiration or anti cipatory socialization. Sait, Edward McChesney. American Parties and Elections. New York: D. Appleton-Century Company, Inc., 1939. Scott, Stanley (ed.). Metropolitan Area Problems. Berkeley: Bureau of Public Administration and University Extension, 1960. Seeley, John R., Sim, R. Alexander, and Loosely, Elizabeth W. Crestwood Heights. New York: Basic Books, Inc., 1956. Similar to Lundberg's Leisure, Crestwood Heights is also largely concerned with suburbia as a way of life. Sherif, Muzafer. The Psychology of Social Norms. New York: Harper and Brothers, 1936. An indispensible work in the study of social norms, it is required in any attempt to understand socialization. Shevky, Eshref, and Bell, Wendell. Social Area Analysis: Theory. Illustrative Application, and Computational Procedures. Stanford: Stanford University Press, 1955. Shibutani, Tamotsu. Society and Personality. Englewood Cliffs: Prentice-Hal1, Inc., 1961. 329 Siegel, Sidney. Non Parametric Statistics. New York: McGraw-Hill Book Company, Inc., 1956. Smith, M. Brewster, Bruner, Jerome S., and White, Robert W. Opinions and Personality. New York: John Wiley and Sons, Inc., 1956. An attempt to understand how opinions are related to personality. However, the effect of the intervening variable of socialization needs more consideration. Sorokin, Pitirim. Social Mobility. New York: Harper and Brothers, 1927. Stouffer, Samuel A., Suchman, Edward A., DeVinney, Leland, Star, Shirley A., and Williams, Robin M., Jr. The American Soldier: Adjustment During Army Life. Yol. I. Princeton, N.J.: Princeton University Press, 1949. This book is recognized by now as one of the most important works in the field of social reality and norms. Taussig, Frank William, and Joslyn, Carl S. American Business Leaders. New York: The Macmillan Company, 1932. Truman, David B. The Governmental Process. New York: Alfred A. Knopf, 1951. Verba, Sidney. Small Groups and Political Behavior. Princeton, N.J.: Princeton University Press, 1961. Warner, W. Lloyd. Social Class in America. Chicago: Science Research Associates, 1949. White, Theodore H. The Making of the President. 1960. New York: Atheneum Publishers, 1961. Wood, Robert. Metropolis Against Itself. New York: Committee for Economic Development, 1959. _________. Suburbia, Its People and Their Politics. Boston: Houghton-Mifflin Company, 1959. Wood's book is one of the most recent attempts to come to grips with suburbia. More in response to the resurgence of concern over suburban behavior, it did not add as much as it could have to the earlier works of Lundberg and Seeley. 1400 Governments. Cambridge, Mass.: Harvard University Press, 1961. 330 B. Governmental Documents Background for Planning. Los Angeles: Los Angeles Velfare Planning Council, 1955. Business and Defense Services Administration. Changing Metropolitan Markets, 1950-1960. Washington: U. S. Government Printing Office, 1961. Hein, G. J. "The State of Rural People in Metropolitan Government." Farm Economics Division, United States Department of Agriculture. Miscellaneous Publication No. 869. Washington: U.S. Government Printing Office, 1961. "Recovery in the Labor Market." Federal Reserve Bulltin. Washington: Federal Reserve Board, May, 1959. U. S. Bureau of the Census. Statistical Abstract of the United States: 1961. 82d ed. Washington: 1961. _. U. S. Census of Population: 1950. Census Tract Statistics. Vol. III. Washington: U. S. Government Printing Office, 1952. _. U. S. Census of Population: I960. Number of Inhabitants, United States Summary. Final Report PC(1) - 1A. Washington: U"! S"! Government Printing Office, 1961. C. Media Crick, Bernard. "California's Democratic Clubs: A Revolt in the Suburbs," The Reporter. XIV (May 31, 1956), pp. 35-39. Feiffer, Jules. "Feiffer" cartoon, New Republic (February 12, 1962), p. 25. Harris, Lou. Life, May 11, 1962, p. 89. Lubell, Samuel. "Class, Party, Divide Suburban Voters," Los Angeles Times, October 24, 1962, p. 30. Miller, Helen Hill. "The City Vote and the Rural Monopoly," The Atlantic Monthly, CCX (October, 1962), pp. 61-65. Morris, William. "Texas Politics in Turmoil," Harper's Magazine, September, 1962, pp. 76-87. 331 New York Times, "Both Parties Court the Suburbs in Bid for New Residents' Votes," May 31, 1956, pp. 1 and 15. _________. "Democratic Gains in Suburbs Cited," June 8, 1957, p. 21. Zeckendorf, William. "Cities Versus Suburbs," The Atlantic Monthly. CC, 7 (July, 1952), pp. 24-28. D. Monographs and Articles in Professional Journals and Readers Adams, H. F. "Autokinetic Sensations," Psychological Monograph. No. 59, 1912. Alford, Robert R. "The Role of Social Class in American Voting Behavior," Western Political Quarterly. 16, 1 (March, 1963j"j pp. 180-94. ________ . "A Suggested Index of the Association of Social Class and Voting," Public Opinion Quarterly. 26, 3 (Fall, 1962), pp. 417-25. Allinsmith, Wesley and Beverly. "Religious Affiliation and Politico—Economic Attitude," Public Opinion Quarterly. 12, 3 (Fall, 1948), pp. 377-889. Anderson, C. Arnold. "A Skeptical Note on the Relation of Vertical Mobility to Education," American Journal of Sociology. 66, 6 (May, 1961), pp. 560-70. Anderson, Totton J. "California: Enigma of National Politics," Western Politics. Frank H. Jonas (ed.), Salt Lake City: University of Utah Press, 1961, pp. 69-112. Asch, Solomon E. "Effects of Group Pressure Upon the Modi fication and Distortion of Judgments," Group Dynamics, Dorwin Cartwright and Alvin Zander (eds.), Evanston: Row, Peterson, and Company, 1953, pp. 151-62. Axelrod, Morris. "Urban Structure and Social Participation," American Sociological Review. 21, 1 (February, 1956), pp. 13-18. Baggaley, Andrew. "White Collar Employment and Republican Vote." Public Opinion Quarterly, 20, 2 (Summer, 1956), pp. 471-73. 332 Banfield, Edward C. "The Politics of Metropolitan Area Organization,” Midwest Journal of Political Science. 1, 1 (May, 1957), pp. 77-91. Becker, Don. "The Fight for Seattle Metro,” Cases in State and Local Government. Richard T. Frost (ed.), Englewood Cliffs, N.J.: Prentice-Hall, Inc., 1962, pp. 292-301. Belknap, George, and Campbell, Angus. "Political Party Identification and Attitudes Toward Foreign Policy." Public Opinion Quarterly, 15. 4 (Winter. 1951-19527",' pp. 601-23. Bell, Daniel. "The Capitalism of the Proletariat," The End of Ideology. Daniel Bell (ed), Glencoe, 111.: The Free Press, 1960, pp. 208-221. Bell, Wendell, and Boat, Marion D. "Urban Neighborhoods and Informal Social Relations," American Journal of Sociology. 62, 4 (January, 1957), pp. 391-98. Blau, Peter M. "Social Mobility and Interpersonal Rela tions," American Sociological Review. 21, 3 (June, 1956), pp. 290-95. Bogue, Donald J. "Urbanism and Metropolitanism," The Suburban Community, William M. Dobriner (ed.), New York: G. P. Putnam's Sons, 1958, pp. 21-25. ________ . "Urbanism in the United States, 1950," American Journal of Sociology. 60, 5 (March, 1955), pp. 471-86. Bresard, Marcel. "La Mobilite Sociale: Le Choix d'un Echelle Sociologique," Transactions of the Second World Congress of Sociology, Vol. II, London: International Sociological Association, 1954, pp. 396-402. Campbell, Angus, Gurin, Gerald, and Miller, Warren E. "Political Issues and the Vote: November 1952," American Political Science Review, 47, 2 (June, 1953), pp. 359-85. Cantril, Hadley. "Identification with Social and Economic Class," Journal of Abnormal and Social Psychology. 38, 1 (January, 1943), pp. 74-80. Carney, Francis. "The Rise of Democratic Clubs in Cali fornia," Case Studies in Practical Politics. New York: Henry Holt and Company, Inc., 1958), pp. 1-16. 333 Case, Herman M. "Guttman Sealing Applied to Centers' Con- servatism-Radicalism Battery," American Journal of Sociology, 58, 6 (May, 1953), pp. 556-63. Centers, Richard. "The American Class Structure: A Psychological Analysis," Readings in Social Psychology, Theodore M. Newcomb and Eugene L. Hartley (eds.), New York: Henry Holt and Company, 1947, pp. 481-93. ________ . "Occupational Mobility of Urban Occupational Strata," American Sociological Review. 13, 2 (April, 1948), pp. 197-203. ________ , "Education and Occupational Mobility," American Sociological Review, 14, 1 (February, 1949), pp. 143-44. ________ . "Nominal Variation and Class Identification: The Working and Laboring Classes," Journal of Abnormal and Social Psychology, 45, 2 (April, 1950), pp. 195-215. ________ . "Social Class, Occupation and Imputed Belief." American Journal of Sociology, 58, 6 (May, 1953), pp. 543-555. Chinoy, Ely. "Research in Class Structure," Canadian Journal of Economics and Political Science, 16, 2 May," 1950), pp. 255-63. Clow, Frederick R. "Cooley's Doctrine of Primary Groups," American Journal of Sociology, 25, 3 (November, 1919), pp. 326-47. Collins, Orvis, Dalton, Melville, and Wray, Donald. "Restriction of Output and Social Cleavage in Industry," Applied Anthropology, 5, 3 (Summer, 1946), pp. 1-15. Cooper, Homer C. "Social Class Identification and Political Party Affiliation," Psychological Reports, 5, 2 (June, 1954), pp. 337-40. Cottrell, Edwin A., and Jones, Helen L. "Characteristics of the Metropolis," Metropolitan Los Angeles. Monograph No. 18, Los Angeles: The Hayes Foundation, 1962. Cox, Edward F. "The Measurement of Party Strength," Western Political Quarterly, 13, 4 (December, 1960), pp. 1022-427 334 Cuzzort, Raymond P. "Suburbanization of Service Industries Within Standard Metropolitan Areas," Studies in Population No. 10. Oxford Ohio: Scripps Foundation and Miami University, 1955. Daland, R. T. "Political Science and the Study of Urbanism," American Political Science Review, 51, 2 (June, 1957), pp. 491-509. “ Davis, Kingsley, and Moore, Wilbert E. "Some Principles of Stratification." American Sociological Review. 10 2 (April, 1945), pp. 242-49. Deeg, M. E., and Peterson, D. G. "Changes in Social Status of Occupations," Occupations. 25, 4 (January, 1947). Dotson, Floyd. "Patterns of Voluntary Association Among Urban Working-Class Families." American Sociological Review, 16, 5 (October, 1951), pp. 687-93. Douvan, Elizabeth, and Adelson, Joseph. "The Psychodynamics of Social Mobility in Adolescent Boys," The Journal of Abnormal and Social Psychology, 56 (January, 1958), pp. 31-44. Dublin, Louis I., and Vane, Robert J., Jr. "Shifting of Occupations Among Wage Earners as Determined by Occupational History of Industrial Policyholders," Monthly Labor Review. 28, 4 (April, 1924), pp. 732-40. Duncan, Otis D. "Research on Metropolitan Populations: Evaluation of Data," Journal of the American Statistical Association, 51, 276 (December, 1956), pp. 591-596. Duncan, Otis D., and Duncan, Beverly. "Residential Distribution of Occupational Stratification," American Journal of Sociology. 60 (March, 1955), 5, pp. 493-503. Eisenstadt, S. M. "Reference Group Behavior and Social Integration: An Explorative Study," American Sociological Review. 19, 2 (April, 1954), pp. 175-85. Eldridge, Hope T. "The Process of Urbanization," Demo graphic Analysis. Joseph Spengler and Otis D. Duncan (eds.), Glencoe, 111.: The Free Press, 1956, pp. 338-43. 335 Ennis, Philip H. "The Contextual Dimension in Voting," Public Opinion and Congressional Elections. William McPhee and William Glaser (eds.), New York: The Free Press of Glencoe, 1962, pp. 180-211. Epstein, Leon D. "Size of Place and the Division of the Two-Party Vote in Wisconsin," Western Political Quarterly, 9, 1 (March, 1956), pp. 138-50. Eulau, Heinz. "Perception of Class and Party in Voting Behavior: 1952," American Political Science Review, 49, 2 (June, 1955), pp. 364-84. _________. "Identification with Class and Political Perspec tive," Journal of Politics, 18, 2 (May, 1956), pp. 233-53. _________. "Identification with Class and Political Role Behavior," Public Opinion Quarterly. 20, 3 (Fall, 1956), pp. 515-29. _________, and Schneider, Peter. "Dimensions of Political Involvement," Public Opinion Quarterly. 20, 1 (Spring, 1956), pp. 128-42. Eysenck, H. J. "Social Attitude and Social Class," British Journal of Sociology, 1, 1 (March, 1950), pp. 56-66. Fava, Sylvia. "Suburbanism as a Way of Life," American Sociological Review. 21 (February, 19567~J pp. 34-37. Festinger, Leon. "Informal Social Communication," Group Dynamics. Dorwin Cartwright and Alvin Zander (eds.), Evanston: Row, Peterson and Company, 1953, pp. 190-203. Fornf, William H., and Miller, Delbert C. "Occupational Career Patterns as a Sociological Instrument," American Journal of Sociology, 49, 4 (January, 1944). _________, and Stone, Gregory. "Urbanism, Anonymity and Status Symbolism," American Journal of Sociology. 62, 5 (March, 1957), pp. 504-14. Fouriezos, Nicholas, Hunt, Max L., and Guetzkow, Harold. "Self-oriented Needs in Discussion Groups," Group Dynamics. Darwin Cartwright and Alvin Zander (eds.), Evanston: Row, Peterson and Company, 1953, pp. 354-60. 336 Freeman, Howard E., and Shovel, Morris. "Differential Political Influence of Voluntary Associations." Public Opinion Quarterly, 15, 4 (Vinter, 1951), pp. 703-14. Fuchs, Lawrence H. "American Jews and the Presidential -Vote," American Political Science Review. 49, 2 (June, 1955), pp. 385-401. Gladfelter, David D. "Vater for Vauwatosa," Cases in State and. Local Government, Richard T. Frost {ed.), Englewood Cliffs: Prentice-Hall, Inc., 1962, pp. 280-291. Glantz, Oscar. "Class Consciousness and Political Solidarity," American Sociological Review. 23, 4 (August, 1958), pp. 375-83. Glaser, Villiam A., and Kadushin, Charles. "Political Behavior in Midterm Elections," Public Opinion and Congressional Elections, Villiam McPhee and Villiam Glaser, (eds.), New York: The Free Press of Glencoe, 1962, pp. 251-72. Goldschmidt, Valter R. "Social Class in America— A Critical Review," American Anthropologist. 52, 4 (October, 1950), pp. 483-98. Gordon, Milton M. "The Concept of the Sub-culture and Its Application," Social Forces, 26, 1 (October, 1947), pp. 40-42. ________ . "Kitty Foyle and the Concept of Class as Culture," American Journal of Sociology, 53, 3 (November, 1947), pp. 210-17. ________ . "Social Class in American Sociology," American Journal of Sociology. 55, 3 (November, 1949) , pp. 262-68. Grant, Dan R. "Urban and Suburban Nashville: A Case Study in Metropolitanism," Journal of Politics. 17, 1 (February, 1955), pp. 82-99. Greenstein, Fred L., and Volfinger, Raymond E. "The Suburbs and Shifting Party Loyalties," Public Opinion Quarterly, 22, 4 (Vinter, 1958), pp. 473-82. Greer, Scott. "Urbanism Reconsidered: A Comparative Study of Local Areas in a Metropolis," American Sociologi cal Review, 21, 1 (February, 19567^ pp. 19-25. 337 _________. "Catholic Voters and the Democratic Party," Public Opinion Quarterly, 25, 4 (Vinter, 1961), pp. 611-25. ________ , and Kube, Ella. "Urbanism and Social Structure: A Los Angeles Study," Community Structure and Analysis, Marvin B. Sussman (ed.), New York: Thomas Y. Crowell Company, 1959, pp. 93-112. Gross, Llewellyn. "The Use of Class Concepts in Sociologi cal Research." American Journal of Sociology. 54, 3 (March, 1949), pp. 409-21. Hastings, Philip. "The Non-Voter in 1952: A Study of Pittsfield, Massachusetts," Journal of Psychology. 38, 1st half (July, 1954), pjT. 301-312. Hastorf, Albert H., and Cantril, Hadley. "They Saw a Game: A Case Study," Journal of Abnormal and Social Psychology. 49,” 1 (January, 1954), pp. 129-34. Hauser, Philip M. "Some Political Influence of Urbaniza tion," Reader in Urban Sociology. Paul K. Hatt and Albert J\ Reiss, Jr. (eds.), Glencoe, 111.: The Free Press, 1951, pp. 459-69. Hawley, Amos, and Zimmer, Basil G. "Resistance to Unifica tion in a Metropolitan Area," Community Political Systems. Morris Janowitz (ed.), Glencoe, 111,: The Free Press, 1961, pp. 146-84. Hofstadter, Richard. "The Pseudo-Conservative Revolt," The New American Right. Daniel Bell (ed.), New York: Criterion Books, 1955, pp. 33-55. Hollingshead, August B. "Selected Characteristics of Classes in a Middle Vestern Community." Arnerican Sociological Review. 12 (August, 1947), pp. 385-95. Hyman, Herbert H. "The Value Systems of Different Classes: A Social Psychological Contribution to the .Analysis of Stratification," Class. Status, and Power. Villiam Bendix and Seymour M. Lipset (eds.), Glen coe, 111.: The Pree Press, 1953, pp. 426-42. Inkeles, Alex, and Rossi, Peter. "National Comparisons of Occupational Prestige," American Journal of Sociology. 61, 4 (January, 1956), pp. 329-39. Jackson, Elton F. "Status Consistency and Symptoms of ? tress," American Sociological Review. 27. 4 August, 19b2), pp. 469-8U. ----------- 338 Janosik, G. Edward. "The New Suburbia: Political Signifi cance," Current History. 31, 185 (August, 1956), pp. 91-95. Janowitz, Morris. "Some- Consequences of Social Mobility in the United States," Transactions of the Third Congress of Sociology, Vol. 3, London: International Sociological Association, 1956, pp. 191-201. _________, and Miller, Villiam E. "The Index of Political Predisposition in the 1948 Election," Journal of Politics, 14, 4 (November, 1952), pp. 710-27. Johnson, Benton. "Ascetic Protestantism and Political Preference," Public Opinion Quarterly, 26, 1 (Spring, 1962), pp. 35-46. Jones, Victor. "Economic Classification of Cities and Metropolitan Areas," Municipal Yearbook. 1954, Chicago: International City Managers Association, 1955, pp. 62-70. Kahl, Joseph A., and Davis, James A. "A Comparison of Indexes of Socio-Economic Status," American Sociological Review. 20, 3 (June, 1 955), pp. 317-25. Katz, Daniel. "The Functional Approach to the Study of Attitudes," Public Opinion Quarterly, 24, 12 (Summer, 1960), pp. 163-2057 Kaufman, Valter C., and Greer, Scott. "Voting in a Metro politan Community: An Application of Social Area Analysis," Social Forces. 38, 3 (March, 1960), pp. 196-204. Key, V. 0., and Munger, Frank. "Social Determinism and Electoral Decision," American Voting Behavior. Eugene Burdick and Arthur Brodbeck (eds.), Glencoe, 111.: The Free Press, 1959, pp. 281-99. Kish, Leslie. "Differentiation in Metropolitan Areas," American Sociological Review. 19, 4 (August, 1954), pp. 388-98. Kitt, Alice S., and Gleicher, David B. "Determinants of Voting Behavior," Public Opinion Quarterly. 14, 3 (Fall, 1950), pp. 393-412. Kornhauser, Arthur V. "Attitudes of Economic Groups," Public Opinion Quarterly. 2, 2 (April, 1938), pp. 260-68. 339 ________ . "Public Opinion and Social Class," American Journal of Sociology, 55, 4 (January, 1950), pp. 333-345. Lassvell, Harold. "The Moral Vocation of the Middle-Income Skill Group," International Journal of Ethics. No. 45 (1935), pp. 127-37. Latham, Earl. "The Group Basis of Politics: Notes for a Theory," American Political Science Review, XLVI, 2 (June, 1952), pp. 376-97. Lazarsfeld, Paul, and Menton, R. K. "Friendship and Social Process," Freedom and Control in Modern Society. Morroe Berger, T. Abel, and C. H. Page (eds.), New York: D. Van Nostrand and Company, 1954, pp. 18-66. Lazerwitz, Bernard. "Metropolitan Community Residential Belts," American Sociological Review, 25, 2 (April, 1960), pp. 245-52. Lemasters, E. E. "Social Class Mobility and Family Inte gration," Marriage and Family Living, 16, 3 (August, 1954), pp. 226-32. Lenski, Gerhard E. "American Social Classes: Statistical Strata or Social Groups?" American Journal of Sociology. 58, 2 (Septembe^ 1952), pp. 139-44. _________. "Status Crystallization: A Non-Vertical Dimension of Social Status," American Sociological Review. 19, 4 (August, 1954), pp. 405-13. ________ . "Trends in Inter-Generational Occupational Mobility in the United States," American Sociologi cal Review. 23, 5 (October, 19587"J pp. 514-23. Lewin, Kurt, and Grabbe, Paul. "Conduct, Knowledge, and Acceptance of New Values," Journal of Social Issues. 11, 3 (August, 1945), pp. 53-64. Liebman, Charles S. "Functional Differentiation and Political Characteristics of Suburbs," American Journal of Sociology. 66, 5 (March, 196T) , pp. 485-90. Lipset, Seymour M. "Social Mobility and Urbanization," Rural Sociology. 20, 3-4 (September-December, 1955), pp. 220-28. 340 ________ , Lazarsfeld, Paul, Barton, Allen, and Linz, Juan. "The Psychology of Voting: An Analysis of Political Behavior," Handbook of Social Psychology. Vol. II, Gardner Lindzey (ed.), Cambridge, Mass.: Addison- Vesley Publishing Company, Inc., 1954), pp. 606- 1226. ________ , and Zetterberg, Hans L. "A Theory of Social Mobility," Transactions of the Third World Congress of Sociology. Vol. Ill, London: International Sociological Association, 1956, pp. 155-77. Maccoby, Eleanor E., Mathews, Richard E., and Morton, Anton S. "Youth and Political Change," Public Opinion Quarterly, 18, 1 (Spring, 1954), pp. 23-39. MacRae, Duncan, Jr. "Occupation and the Congressional Vote, 1940-1950," American Sociological Review. 20, 3 (June, 1955), pp. 332-44. Manis, Jerome G., and Stine, Leo C. "Suburban Residence and Political Behavior," Public Opinion Quarterly. 22, 4 (Vinter, 1958), pp. 483-89. Mayer, Kurt. "The Theory of Social Classes," Transactions of the Second Vorid Congress of Sociology. Vol. II, London: International Sociological Association, 1954, pp. 321-35. McClosky, Herbert, and Dahlgren, Harold. "Primary Group Influence on Party Loyalty," American Political Science Review, 53, 3 (September, 1959), ppl 757-76. Merriam, Charles E. "Progress Report of the Committee on Political Research: Recent Advances in Political Methods," American Political Science Review. 17, 2 (May, 1923), pp. 274-95. Meyer, John V. "A Reformulation of the 'Coattails' Problem," Public Opinion and Congressional Elections. Villiam McPhee and Villiam A. Glaser (eds.), New York: The Free Press of Glencoe, 1962, pp. 52-64. Moneypenny, Phillip. "Political Science and the Study of Groups: Notes to Guide a Research Project," Vestern Political Quarterly. 7, 2 (June, 1954), pp. 183-201. __________, and Steiner, Gilbert Y. "Merger? The Illinois Consolidation Case," Cases in State and Local Government. Richard T~ Frost (ed.), Englewood Cliffs, N.J.: Prentice-Hall, Inc., 1962, pp. 267-79. 341 Monk, Mary, and Newcomb, Theodore M. "Perceived Consensus Within and Among Occupational Classes," American Sociological Review. 21, 1 (February, 1956), pp. 71-79. Moore, Joan V. "Social Deprivation and Advantage as Sources of Political Values," Western Political Quarterly. 15, 2 (June, 1962), pp. 217-26. Murphy, Raymond J., and Morris, Richard T. "Occupational Situs, Subjective Class Identification, and Political Affiliation," American Sociological Review. 26, 3 (June, 1961), pp. 383-92. National Opinion Research Center. Opinion News, reprintedin Class. Status and Power. Reinhard Bendix and Seymour Lipset (eds.), Glencoe, 111.: The Free Press, 1953, pp. 411-26. Nixon, Charles R. "The Coming Electorate: 1965-1970," Western Political Ouarterly, 13, 3 (September, I960), pp. 620-35.-------- Parsons, Talcott. "Social Classes and Class Conflict in the Light of Present Sociological Theory," American Economic Review, 39, 3 (May, 1949), pp. 16-26. ________ . "Social Strains in America," The New American Right, Daniel Bell (ed.), New York: Criterion Books, 1955. Perlin, Leonard I. "Shifting Group Attachments and Attitudes Toward Negroes," Social Forces, 33, 1 (October, 1954), pp. 47-50. Pfautz, Harold W. "The Current Literature on Social Stratification: Critique and Bibliography," American Journal of Sociology. 58, 4 (January, 1953), pp. 391-418. Pool, Ithiel de Sola, and Abelson, Robert P. "The Simula tion Project," Public Opinion Quarterly, 25, 2 (Summer, 1961), pp. 167-183. Powell, Elwin H. "Occupation, Status, and Suicide," American Sociological Review, 23, 2 (April, 1958), pp. 131-139. Precker, Joseph A. "Similarity of Valuings as a Factor in Selection of Peers and Near—Authority Figures," Journal of Abnormal and Social Psychology, 47, 2 (April, 1952), pp. 406-14. 342 Redick, Richard V.* "Population Growth and Distribution in Central Cities, 1940-1950," American Sociological Review. 21, 1 (February, 1956), pp. 38-43. Reiss, Albert J., Jr. "Research Problems in Metropolitan Population Redistribution," American Sociological Review. 21, 5 (October, 1956), pp. 571-77. Riecken, Henry ¥. "Primary Groups and Political Party Choice," American Voting Behavior, Eugene Burdick and Arthur Brodbeck (eds.), Glencoe, 111.: The Free Press, 1959, pp. 162-83. Riesman, David. "The Suburban Dislocation," The Annals of the American Academy of Political and Social Sciences. 314 (November, 1957), pp. T23-46. Robinson, ¥. S. "Ecological Correlations and Behavior of Individuals." American Sociological Review. 15, 3 (June, 1950), pp. 351-57. Roper, Elmo. "Classifying Respondents by Economic Status," Public Opinion Quarterly. 4, 2 (Summer, 1940), pp. 270-72. Rose, Peter I. "Student Opinion on the 1956 Presidential Election," Public Opinion Quarterly, 21, 3 (Fall, 1957), pp. 371-76. Rosen, Bernard C. "Conflicting Group Pressure: A Study of Parent-Peer Group Cross Pressures," American Sociological Review. 20, 2 (April, 1955), pp. 155-61. ________ . "The Achievement Syndrome: A Psychocultural Dimension of Social Stratification," American Sociological Review. 21, 2 (April, 1956), pp. 203-11. Rossi, Peter H. "Four Landmarks in Voting Research," American Voting Behavior. Eugene Burdick and Arthur Brodbeck (eds.), Glencoe, 111.: The Free Press, 1959, pp. 5-54. Scammon, Richard M. "Voting for President in the Larger Metropolitan Areas 1952-1956," Midwest Journal of Political Science. 1, 3-4 (November, 1957), pp. 330-33. Schnore, Leo F. "The Function of Metropolitan Suburbs," American Journal of Sociology. 61, 5 (March, 1956), pp. 453-58. 343 _________. "Social Mobility in Demographic Perspective," American Sociological Review, 26, 3 (June, 1961), pp. 407-23. _________. "Municipal Annexations and the Growth of Metro politan Suburbs, 1950-1960," American Journal of Sociology. 67, 4 (January, 1962), pp. 406—17. _________, and Varley, David. "Some Concomitants of Metro politan Size," American Sociological Review. 20, 4 (August, 1955), pp. 408-14. _________, and Alford, Robert R. "Forms of Government and Socioeconomic Characteristics of Suburbs," Administrative Science Quarterly. 8, 1 (June, 1963), pp. 1-17. Scudder, Richard, and Anderson, C. Arnold. "Migration and Vertical Mobility." American Sociological Review. 19, 3 (June, 1954), pp. 329-34. Sherif, Muzafer. "Group Influences Upon the Formation of Norms and Attitudes," Readings in Social Psychology. Theodore M. Newcomb and Eugene L. Hartley (eds.), New York: Henry Holt and Company, 1947), pp. 77-90. Shibutani, Tamotsu. "Reference Groups and Social Control," Human Behavior and Social Process. Arnold Rose (ed.), Boston: Houghton Mifflin Company, 1962, pp. 128-47. Shills, Edward A. "The Study of the Primary Group," The Policy Sciences. Daniel Lerner and Harold D. Lasswell (eds.), Stanford: Stanford University Press, 1951, pp. 45-51. Sibley, Elbridge. "Some Demographic Clues to Stratifica tion." American Sociological Review. 7, 3 (June, 1942), pp. 322-330. Sjoberg, Gideon. "Are Social Classes in America Becoming More Rigid?" American Sociological Review. 16, 6 (December, 1951), pp. 775-83. Steiner, Ivan D. "Some Social Values Associated with Objectively and Subjectively Defined Social Class Membership," Social Forces. 31, 4 (May, 1953), pp. 327-32. Stokes, Donald E., and Miller, Varren E. "Party Government and the Saliency of Congress," Public Opinion Quarterly. 26, 4 (Vinter, 1962), pp. 531-46. 344 Strunk, Mildred. "The Quarter's Polls." Public Opinion Quarterly. 13, 1 (Spring, 1949), pp. 155-76. ________ . "The Quarter's Polls," Public Opinion Quarterly. 13, 4 (Winter, 1949-1950), pp. 709-32. ________ . "The Quarter's Polls," Public Opinion Quarterly, 14, 1 (Spring, 1950), pp. 174-92. Thompson, Wayne E., and Horton, John E. "Political Alienation As a Force in Political Action," Social Forces. 38, 3 (March, 1960), pp. 190-95. Westoff, Charles F., Bressler, Marvin, and Sagi, Philip C., "The Concept of Social Mobility: An Empirical Inquiry," American Sociological Review. 25, 3 (June, 1960), pp. 375-85. Wirth, Louis. "Urbanism As a Way of Life," American Journal of Sociology. 44, 1 (July, 1938), pp. 1-24. E. Unpublished Reports and Materials Belknap, George M. "Motivational Research in the 1960 Presidential Campaign," a paper delivered at the 1961 Convention of the American Political Science Association. "The Challenge of Metropolitan Government." The American University, Washington, D.C., 1958. (Mimeographed.) Converse, Phillip E. "Group Influence in Voting Behavior." Unpublished Ph.D. dissertation, Department of Social Psychology, University of Michigan, 1958. (Microfilm.) Cooper, Homer C. "Interaction of Socioeconomic Character istics and Reference Groups vith Respect to Voting Behavior," Unpublished Ph.D. dissertation, Department of Social Psychology, University of Michigan, 1957. (Mic.rofilm.) Dye, Thomas. "Certain Political Correlates of the Areal Differentiation and Stratification of Subpopulations in a Metropolitan Area." Unpublished Ph.D. disser tation, Department of Political Science, Pennsylvania University, 1961. (Microfilm.) _________. "Leadership and Constituency in Fifteen Suburban Communities." A paper delivered at the 1963 Conven tion of the American Political Science Association, 345 Holden, Mathew. "The Governance of the Metropolis As a Problem in Diplomacy." A paper delivered at the 1962 Convention of the American Political Science Association. (Mimeographed.) Kaplan, Norman. "Reference Group Theory and Voting Behavior." Unpublished Ph.D. dissertation, Department of Political Science, Columbia Univer sity, 1955. (Microfilm.) Lazerwitz, Bernard. "Some Characteristics of Residential Belts in the Metropolitan Community, 1950-1956." Unpublished Ph.D. dissertation, University of Michigan, 1958. (Microfilm.) Mitchell, Robert E. "Three Theories of Status Crystalliza tion: Uses and Misuses of Indicators of Social Stratification." A paper delivered at the third annual conference of the Pacific Coast Chapter of American Association for Public Opinion Research, January, 1963. (Mimeographed.) Research Division, Democratic National Committee. "The 1960 Election Report," Washington, D.C., 1962. (Mimeographed.) Research Division, Republican National Committee. "The 1960 Elections," Washington, D.C., 1961. (Mimeographed.) ________ . "The 1962 Elections," Washington, D.C., 1963. (Mimeographed.) Stokes, Donald E. "1960 and the Problem of Deviating Elections." A paper delivered at the 1961 Convention of the American Political Science Association. Wallace, David. "The Sociology of Stability and Change in One Suburb's Voting." Unpublished Ph.D. disserta tion, Department of Political Science, Columbia University, 1962. (Microfilm.) Warren, Robert. "A Municipal Services Market Model of Metropolitan Organization." A paper delivered at the 1962 Convention of the American Political Science Association. (Mimeographed.) Williams, Oliver P., Dye, Thomas R., Herman, Harold, and Liebman, Charles S. "Urban Differentiation and Political Integration." A paper delivered at the 1962 Convention of the American Political Science Association. (Mimeographed.) 346 Wirt, Frederick. "Suburban Patterns in American Politics." A paper delivered at the 1960 Convention of the American Political Science Association. (Mimeographed.)

Abstract (if available)

Abstract This dissertation examines the changing spatial and functional patterns of metropolitan living which, in both professional literature and popular media, are alleged to have had a direct influence upon the political system and political beliefs.

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

The Federal theatre in the Los Angeles area

PDF

The French in Los Angeles: A study of a transplanted culture

PDF

Special library collections in the Southern California area as an approach to historical studies

PDF

The social status of the negro in Pasadena, California

PDF

A study of juvenile gangs in the Hollenbeck area of East Los Angeles

PDF

The unattached Negro woman on relief: A study of fifty unattached Negro women on relief in the Compton district office of the State Relief Administration of California in Los Angeles

PDF

Experimental analysis of the instructional materials in behavioral college instruction using a marketing management course

PDF

A study of the transient boy in Los Angeles County in 1939

PDF

A sociological study of public opinion concerning certain police practices in Los Angeles

PDF

Population pressure and its effect on the land use pattern and land valuations in Los Angeles County from 1940 to 1953

PDF

Adjustment of large downtown and boulevard churches in Los Angeles to socio-cultural factors in the community

PDF

Essays on bundling and discounts

PDF

Stereotypes of selected white college students concerning Negroes

PDF

Social media's role, utility, and future in video game public relations

PDF

Sociological survey of Main Street, Los Angeles, California

PDF

Informal consent: the complexities of public participation in post-civil war Lebanon

PDF

Perceptions of dementia among Latinos

PDF

Archiving the absence: female infanticide in nineteenth-century British India

PDF

The significance of zones and boundaries in the urban community with special reference to the Los Angeles Area

PDF

Empress Wu of the Tang dynasty: becoming the only female emperor in China

Asset Metadata

Creator Bell, Charles Gordon (author)

Core Title Suburban Political Behavior In Los Angeles County, California

School Graduate School

Degree Doctor of Philosophy

Degree Program Political Science

Publisher University of Southern California (original), University of Southern California. Libraries (digital)

Tag aspiration,attitude stability,ballot propositions,class politics,conformity,cross pressures,Democrat,election participation,group membership,home ownership,household,Immigrants,Income,non-partisan voting,OAI-PMH Harvest,occupational mobility,partisan voting,political behavior,political science, general,reference groups,Republican,residential mobility,social mobility,social norms,social reality,Suburbia,urbanization

Place Name California (states), Los Angeles (counties), USA (countries)

Format xvi, 346 leaves : ill., col. map ; 29 cm. (aacr2)

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Anderson, Totton J. (committee chair), Kooker, Arthur R. (committee member), Krinsky, Fred (committee member), Kurth, Jasumasa (committee member), Rigby, Gerald (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-m10

Unique identifier UC11359809

Identifier 6611561.pdf (filename),usctheses-m40 (legacy collection record id),usctheses-c18-217050 (legacy record id),usctheses-m10 (legacy record id)

Legacy Identifier 6611561.pdf

Dmrecord 217050

Document Type Dissertation

Format xvi, 346 leaves : ill., col. map ; 29 cm. (aacr2)

Rights Bell, Charles Gordon

Type texts

Source University of Southern California (contributing entity), University of Southern California Dissertations and Theses (collection)

Access Conditions The author retains rights to his/her dissertation, thesis or other graduate work according to U.S. copyright law. Electronic access is being provided by the USC Libraries in agreement with the au...

Repository Name University of Southern California Digital Library

Repository Location USC Digital Library, University of Southern California, University Park Campus, Los Angeles, California 90089, USA