University of Southern California Dissertations and Theses

An Investigation Of Complementary Needs Between Marital Partners

(USC Thesis Other)

An Investigation Of Complementary Needs Between Marital Partners

PDF

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content AN INVESTIGATION OF COMPLEMENTARY NEEDS BETWEEN MARITAL PARTNERS by Charles Raymond Fowler 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 (Sociology) June 1971 71- 27,921 FOWLER, Charles Raymond, 1928- AN INVESTIGATION OF COMPLEMENTARY NEEDS BETWEEN MARITAL PARTNERS. University of Southern California, Ph.D., 1971 Sociology, family University Microfilms, A X E R O X Company, Ann Arbor, Michigan UNIVERSITY OF SOUTHERN CALIFORNIA THE GRADUATE SCHOOL UNIVERSITY PARK LOS ANGELES. CALIFORNIA S 0 0 0 7 This dissertation, written by .Charlesi. Raynond__Fowler........... under the direction of Dissertation Com mittee, and approved by all its members, has been presented to and accepted by The Gradu ate School, in partial fulfillment of require ments of the degree of D O C T O R OF P H I L O S O P H Y Dean Date. DISSERTATION COMMITTEE / Vi u Chairman PLEASE NOTE: Some pages have small and indistinct type. Filmed as received, University Microfilms ACKNOWLEDGEMENTS This exercise is affectionately dedicated to Clarence and Edith, my parents, and to Barbara, with whom I entered into a relationship based on thoroughly complementary needs, and to Jerry Nims by whose art, as well as science, we were encouraged to transform our pattern of need expression into one of greater homogamy. I wish to acknowledge my considerable appreciation to my friend, mentor, and guide, Dr. James A. Peterson - friend for nearly three decades, mentor in the discipline of my academic field, and guide in those larger issues of life and work which elude definition. Also I wish to express my sincere thanks to Dr. Thomas E. Lasswell and Dr. Wm. Fawcett Hill, who, as members of my committee bore patiently the vicissitudes of service. And finally, to Mr. Reuben Snipper, to whom I owe a debt of statistical significance such as I cannot enumerate by any method adequate to the test, I express my thanks for his indis pensable assistance in the design and execution of the analysis of the data on which this research is based. This expression of indebtedness and appreciation is not to disclaim responsibility in anyway - whatever errors, aberra tions, and distortions may be found in this work are entirely my own. ii TABLE OF CONTENTS ACKNOWLEDGMENTS ............................ , LIST OF TABLES............................ . LIST OF FIGURES ................ ....... Chapter I. INTRODUCTION ...................... Statement of the Problem Definitions of Terms Used Hypotheses to be Tested Research Methods Limitations of the Study Order of Presentation II. REVIEW OF THE LITERATURE .......... Mate Selection as Need Gratification III. THEORETICAL STATEMENT ............. Axiomatic Statements Propositions Hypotheses IV. RESEARCH METHODS .................. Sample Instruments Statistical Procedures THE FINDINGS OF THE Hypothesis I Hypothesis II Hypothesis III Hypothesis IV Hypothesis V Hypothesis VI Emergent Findings VI. SUMMARY AND CONCLUSIONS Summary iii iv Chapter Conclusions Implications For Future Research APPENDIX A - The Questionnaire...................... 129 APPENDIX B - The Tattles and a Figure From the Miller Study.............. * ........... 135 APPENDIX C - The Circumplex Schema of the ICL............ 1^9 BIBLIOGRAPHY . . . .................. 15^ LIST OP TABLES Table Page 1. Comparison of Two Tests For Pre-Nuptial Partners and Remarried Partners . . • • • • • • • • « • . . 9 2. Age Distribution of Sample, by Percentage........ 57 3. Social Class Positions of Males and Females for the Selected Sample, by Percentages 62 Occupational Positions of the Sample by Sex • . • • 65 5. Educational Positions of the Sample by Sex • ■ • . 66 6. Length of Present Marriage • •.....••••*• 68 7. Religious Preferences of the Sample by Percentages 69 8. Mixed Religious Marriages, by Religion ■•••>•. 70 9. Marital Role Typology by Sex ....•••.•••• 95 10. Accuracy of Perception by Sex .............. 97 11. DOM Need Difference................ 99 12. LOV Need Difference . . . . . . . . . . . . . . . . 100 13* Total Need Difference 101 1^. Correlations Between the Males and Females Scores Among Subsamples and Total Sample............ 105 15. Correlations ............ ............. 108 16. Correlations Between the Males and Females Accuracy of Perception Scores Among Subsamples and the Total Sample ....... ...... Ill v LIST OF FIGURES Figure 1. Layout of a Two-Way Analysis of Variance Design ■ vi Page . 88 CHAPTER X INTRODUCTION Mate selection is the basis for the marital interaction which will take place between the husband and wife during their years of marriage. Therefore, it is important to study mate selection as a basis for understanding marital Interaction processes. The purpose of this research is to consider the two principal theories of mate selection and the variables related to them. Each of these two theories of mate selection attempts to answer the question of how a person selects a mate. Does he select one whose characteristics are like his own, or one whose characteristics are different from his own? This is an old question. One which is present in the folk wisdom of marriage through the two phrases "like marries like" and "opposites attract." In the research literature on marriage one also finds much regarding this question with impressive evidence being pre sented to support each of these two contradictory theories of mate selection. A recent study of the question pertaining to mate selec tion was completed by Dr. Louis Irving Miller.^ The theoretical explanation forwarded by Dr. Miller for the contradictory ^Louis I. Miller* Familial Role Typology. Accuracy of Perception, and Mutual Needs Among Pre-Nuptial Partners, un published dissertation, University of Southern California, 1966, 1 role typology, accuracy of perception, and interpersonal needs* While Miller used the term "familial" to identify the role typology he investigated, the present researcher prefers "marital*" Since the role interaction is between marital partners as husband and wife (i.e., marital) rather than as father and mother (i.e., familial), the term "marital role typology" seems more accurate than "familial role typology," The instrument which Miller used to measure this variable remains the same in this replication study. Definition of Terms Used It is necessary to define the major terms to be encoun tered throughout the remainder of this research* Marital Bole Typology As it is used in this research, marital role typology refers to the cluster of concepts which differentiate the two major types of husband-wife interaction: the traditional, and the equalltarian, Burgess and Locke use the terms "institutional i f and "companionship" for these two types of marital relationship* Parsonis and Bales describe the role-allocations within the tra ditional pattern as "instrumental" for the husband and "expres- sive" for the wife. While admitting that in certain respects, ^Ernest W, Burgess and Harvey J. Locke, The Family (2nd ed,; New York: American Book Company, 1953)» P* 22, ^Talcott Parsons and Robert F, Bales, Family. Socialiga- t-lon and Interaction Process (New York: The Free Press, 1955) p* 339« Chapter 6, Morris Zelditch, Jr,, "Role Differentiation in the Nuclear Family: A Comparative Study," American middle class marital role patterns appear equalitarian, or at least "equivocal," that is, not clearly and rigidly allo cated as husband-instrumental and wife-expressive, Zelditch asserts that there is an underlying residual role-expectation supporting this traditional pattern. Blood and Wolfe report from their Detroit Area Study sample of 1957 that most middle class families tended to make decisions jointly (equalitarian) 7 and a few were one-partner dominant (traditional). For this research a typology of marital roles was constructed which con forms in the main to the traditional-equalitarian typology just described. Accuracy of Perception In this study, accuracy of perception of the partner's role attitudes (and therefore, expectations) will be a measure of the ability of one partner to predict correctly how his Bpouse will respond on the scale of marital role attitudes. Interpersonal Needs Interpersonal needs are defined as needs of a person for responses from others based on internal motives or drives to achieve a specified goal. Timothy Leary, whose pioneering work has been utilized in this research, devised a matrix for the 6Ibid. 7 rRobert 0. Blood, Jr. and Donald M. Wolfe, Husbands and Wives (Olencoe, 111.: The Free Press, I960), p. 3^. analysis of interpersonal needs by dichotomizing them into those concerned with ascendancy (dooinance-submission) on the horizon tal axis, and affiliation (love-hate) on the vertical axis. Two types of organization of needs as a basis for mate selection will be considered. They are defined in the following paragraphs* Complementary Needs An organization of needs in such a way that the person perceives his mate as having needs and responses in marriage which are different from his own, is termed complementary needs. Homogamous Needs An organization of needs in such a way that the person perceives his mate as having needs and responses in marriage which are Bimilar to his own, is termed homogamous needs. The Hypotheses The following hypotheses are tested in this study: I. The difference between the interpersonal needs of marital partners will vary as a function of their marital role attitudes. XX. The difference between the interpersonal needs of marital partners will vary as a function of the inter actional effects of their marital role attitudes and their accuracy of perception of the mate's marital role attitude. 5 III. There will be a negative correlation between the interpersonal needs of traditional type marital partners. IV. There will be a positive correlation between the interpersonal needs of equalitarian type marital partners. V. There will be a positive correlation between the difference of interpersonal needs and accuracy of perception of mate's marital role attitude for marital partners of the traditional type. VI. There will be a negative correlation between the difference of interpersonal needs and accuracy of perception of mate's marital role attitude for marital partners of the equalitarian type. Research Methods The sample of sixty-three couples was obtained through the adult education classes taught by the author for the Extension Division of the University of California. In each of the couples one or both of the partners had been previously married. As this is a replication study, the criteria for inclusion in the sample were the same as those of the exploratory study done by Miller. There were three social characteristics: Caucasian race, of non-Jewish religion, with the husband rated as middle or upper middle class on Hollingshead's "Two Factor Index of Social Position." The measurement of the variables was operationalized through the use of the Marital Role Typology Scale, the Accuracy of Perception Scale, and the Interpersonal Check List. A two-way analysis of variance design, as well as Pearson's product-moment coefficient of correlation techniques were used to test the six hypotheses at the .05 level of significance. Limitations of the Study The first factor to be considered as a limitation of the study is the representativeness of the sample. Just as Miller's study was limited by his sampling procedure, so the sampling method of the present study undoubtedly introduced some kind of a bias. Whereas his was a known loss of approximately ten percent according to his estimate, since the present study was based on a sample partially gathered by others, there is no satisfactory method to estimate the amount of bias introduced by the unknown refusal rate. As was true of Miller's research, as well as others in survey research, whenever the target population refuses to participate in the survey, it introduces the specter of un representativeness of the results. Lid those couples who refused to participate in this research have traditional, equali tarian, or mixed marital role typologies? It is not known. Moreover, those couples who did agree to participate in ^Miller, "Familial Role Typology," p. 109* the study by returning a completed questionnaire did so at least in part out of loyalty to the class instructor, or else to a friend who asked them to assist him in thiB for a class he was taking. Whatever the precise composition of the motives, this selection procedure introduced a bias in the direction of over representing public school teachers (the occupation of a majority of the class members in the extension classes where the forms were distributed) or else friends of these teachers. Another consideration which tends to limit the power of the study for generalization is the validity of the subject's responses to the survey instruments. This issue has two parts, the first of which is the question of the psychic state of the couples at the time they completed the test items. One of the possible explanations offered by Miller for his lack of significant findings was the influence of euphoria on the pre- 9 nuptial couples. It may very well be that even after they are married, couples continue to retain some residual euphoria or lack of reality-testing in their assessment of themselves and of each other. If such iB the case, it could logically be expected to bias their responses. The second part of this issue of the validity of subject responses is the possible violation of the instructions to the couples to work separately on the questionnaire. Since these test forms were not filled out in the presence of the researcher, ^Miller, "Familial Role Typology," p. 110. it was not possible to control for this source of bias. A comparison of responses between partners may have occurred as an artifact of either subordinate or superordinate behavior - i.e., a subordinate spouse may have copied the mate's responses in order to "give" compliance, or else a superordinate spouBe may have directed the mate to reply in certain ways in order to "receive" compliance. These two possible cases of contamination are congruent to the traditional-authoritarian marital role typology. It is more difficult to conceptualize a source of such bias from the autonomous-democratic portion of the sample, since our theoretical formulation by definition predicts such couples to be more highly individuated, as well as autonomous, and therefore more likely to refuse to participate at all than to feel compelled to participate, yet feel constrained to violate the terms of the contract for participation presented in the test instructions. Since the Marital Hole Typology test and the Accuracy of Perception test are especially susceptible to comparison-of and cooperation-on answers, it is worthy of note that there is only slight difference between the Miller study results and the results of the present study on these two measures. The mean score for the total samples on the Marital Hole Typology test is presented in Table 1* The mean percent of accuracy for the total samples on the Accuracy of Perception test is also pre sented in Table 1* Finally, there is a factor to be considered in regard to TABLE 1 COMPARISON OF TWO TESTS FOR PRE-NUPTIAL PARTNERS AND REMARRIED PARTNERS SAMPLES Pre-nuptial Remarried TEST Partners N = *4-20 Partners N = 126 Marital Role Typology (Mean Score) 57 55 Accuracy of Perception (percent accuracy) 68 71 io subject responses not related to suphoria, nor to Intentional duplication of the spouse's answerst but rather to the method ological issue of determining the source of accuracy of percep tion. That is, for those persons who were accurate in their predictions of their mate's marital role attitudes, no assurance is available that this accuracy resulted from true knowledge of the mate's attitudes, rather than from the spurious coincidence between his own projected attitudes and the mate's actual responses. Generalization from the findings of this study may be subject to certain limitations of a methodological nature. The population from which the sample subjects were recruited is not sufficiently representative to avoid the limitation of selection bias. No satisfactory measure of euphoria was administered to the sample couples, so that their responses may have been dis torted by their state of mind. Further distortion in the array of couples' responses may have occurred either as a result of intentional copying between spouses, or the fortuitous coincidence of mates' responses merely as a projection of their own attitudes, rather than as an accurate perception of the other's attitudes. Order of Presentation Chapter II reviews the literature pertaining to mate selection as need gratification, especially as related to the Winch hypothesis of need complementarity. Chapter III presents the theoretical framework within which this study is oriented. It concludes with a statement of this theory in axiomatic form, followed by the research proposi tions derived from the axiomatic summary, as well as the hypo theses to be tested. Chapter IV describes in detail the research methods used to test the hypotheses. Chapter V reports the findings of the study. Chapter VI contains the summary, conclusions, and implica tions for future research. CHAPTER II REVIEW OF THE LITERATURE Studies of Mate Selection as Need Gratification With the removal of traditional parental control of the mate selection process, social researchers have turned their attention to the question of why persons choose a marriage partner as they do# For social factors the evidence indicates conclusively that hoaogamy is the rule#'*' One of the most impor- 2 tant social characteristics is age# A second hasic social • 5 characteristic is race*^ The third limiting social character- L istic is social position# Religion is also a major social 5 characteristic influencing mate selection# The final social ^For a general discussion, see Ernest W. Burgess and Paul Wallin, "Homogamy in Social Characteristics,1 1 American Journal of Sociology. XLIX (19^3)» 109-12^. 2 P* C# Gliclc and E« Landau, "Age as a Factor in Marriage," American Sociological Review# XV (1950), 517-529* 3 "X# Freeman, "Homogamy in Interethnic Mate Selection," Sociology and Social Research. XXIX (1955)i 369-377* 1 ^ S# Dinitz, F. Banks, and B# Pasamanick, "Mate Selection and Social Class: Changes During the Past Quarter Century," Marriage and Family Living# XXII (i960), 3^8-351* 5 Paul Glick, "Intermarriage and Fertility Patterns Among Persons in Major Religious Groups," Eugenics Quarterly# VII (I960), 31-38. 12 13 characteristic of importance is residential propinquity*^ From the manifold variety of factors posited, two models have been arranged to attempt to answer the question of mate selection as need gratification* The first model makes the assumption that persons with different (complementary) needs will be attracted to each other* This model has been forwarded by Robert F. Winch and others* Winch delineated four main combinations of need gratification from an early study of twenty-five couples. These four styles are as follows: 1* Mother-Son, in which a capable, dominant wife cares for a leaning, dependent husband; 2. Father-Daughter, in which a strong, capable husband protects and provides for a passive, compliant wife; 5* Master-Servant girl, in which a superior husband is served by a catering wife; and km Amazon-Milquetoast, in which a powerful wife dominates a frightened and frustrated husband* Winch's theory of complementarity in its latest revision is stated as follows: Where love is regarded as a desirable precondition to marriage, it is theorized that mate-selection will take place on the basiB of complementary needs* The central hypothesis of the theory of complementary needs states that within the field of eligibles, persons whose need ^A. M. Katz and R. Hill, "Residential Propinquity and Marital Selection: A Review of TheoryMarriage and Family Living. XX (1958), 27-3^* patterns provide mutual gratification will tend to choose each other as marriage partners.7 This is the most generalized statement of the theoryT and does not specify the two types of complementarity which Winch posited, and tested, and for which he claimed to find some support. These two types of complementarity are Type I comple mentarity (i.e., different) on the same need, as for example when one of the pair has a high need to dominate, the partner will have a low need to dominate; Type XI complementarity is when one of the pair has a need to nurture, the partner will complement that need by a need for succorance (i.e., need to be nurtured). These two types of complementarity can be seen as differences of intensity and kind, respectively. Type I complementarity 1b of the same need at very different levels of intensity (high and low), while Type XI is of different needs at some unspecified Q level of intensity. The operational difficulty of confirming the presence or absence of complementary mate selection processes is the lack of tightness of the design at this point. That is, Type I and Type II may be existing within the same pair, and even within the same needs* For instance, one of the pair may have high dominance needs, while hie partner may have either low dominance needs (Type I). Or more complementary (in the Bense of different) than that, she may have moderate to high submissive ^Robert F. Winch, The Modern Family (revised edition; New York: Holt, Rinehart and Winston, Inc., 19&3)» P* &07* 8. 15 needs* So that the design which seems to he dichotomous, is in operation subject to interpretation as several points along a continuum ranging, in this case, from high dominance to high submission* It is this problem which Winch attempted to solve operationally by the introduction of the two types of need complementarity, Intensity and kind* In order to test his theory, Winch chose a sample of twenty-five married couples from the student body of Northwestern University. He and his co-workers then studied this Bample intensively through the administration and analysis of an extended interview composed of a need segment, a case history, and an eight card Thematic Apperception Test. Each element of the battery was analyzed and interpreted and then the investigating 9 team pooled their results for a summary rating of all the data* Winch organized his study along the lines of "needs" originally explicated by Henry A. Murray. These needs are reported as follows: g Robert E* Winch, Thomas Ktsanes, and Virginia Ktsanes, "Empirical Elaboration of the Theory of Complementary Needs in Mate Selection," Journal of Abnormal and Social Psychology. LI (1953), 508-513. ^Henry A* Murray, Explorations in Personality (New York: Oxford University Press, 1938), p. 1^2. 16 Abasement Achievement Approach Autonomy Deference Dominance Hostility Nurturance Recognition Sex Needs To accept or invite blame, criticism or punishment. To blame or harm the self. To work diligently to create something and/or to emulate others. To draw near and enjoy interaction with another person or persons. To get rid of the constraint of other persons. To avoid or escape from domination. To be unattached and independent. To admire and praise a person. To influence and control the behavior of others. To fight, injure, or kill others. To give sympathy and aid to a weak, helpless, ill, or dejected person or animal. To excite the admiration and approval of others. To develop an erotic relationship and engage in sexual relations. 17 Status Aspiration To desire a socio-economic status considerably higher than one has. (A special case of achievement.) Status Striving To work diligently to alter one's socio-economic status. (A special case of achievement.) Succorance To be helped by a sympathetic person. To be nursed, loved, protected, • ^ i -a 11 indulged. To these needs Winch added three general traits which are recorded in the following form. General Traits Anxiety Emotionality Vicariousness Fear, conscious or unconscious, of harm or misfortune arising from hostility of others and/or social reaction to one’s own behavior. The show of affect in behavior. The gratification of a need derived from the perception that another person is deriving gratification. 12 Thomas and Virginia Ktsanes, nThe Theory of Complemen tary Needs in Mate-Selection," original article in Robert F. Winch and Robert McGinnis, eds., Selected Studies in Marriage and the Family (New York: Holt, 1953)» PP* 522-523* 12Ibid. 18 The first operation performed on the data according to the research reports published by the Winch team was to group four of these needs into a cluster labelled "assertive." These needs were Achievement, Autonomy, Dominance, and Hostility. The accompanying hypothesis was that high assertives tend to pair with high receptives. This hypothesis was supported as reported 13 in the earliest publication of this group on complementarity. From the original data, Winch and his associates ulti mately constructed a design matrix of forty-four sub-variables accounting for the computation of over nineteen hundred inter personal correlations, testing for both Type I and Type II complementarity. From this array, Winch made hypotheses regarding three hundred eighty-eight of the correlations. He expected three hundred forty-four to be positive since these were between different interspousal needs, Type II. He pre dicted on the other hand, that forty-four of the total correla tions between spouses would be negative, since they were for the Ik same need, Type I. For example, if one of the pair had high needs to dominate, then the partner would be expected to have a high need to be submissive, giving a positive sign to the inter spousal correlation. Whereas, on some needs, such as achievement, it would be expected that the husband who rates high would select 13Ibid. lh Robert F. Winch, "The Theory of Complementary Needs in Mate Selection: Final Results on the Test of the General Hypo thesis," American Sociological Review. XX (1955)* 552-555* a wife who is low on this need, giving a negative interspousal IK correlation. This elaborate array of correlations was achieved by subjecting the data to five measures of association. Included were two ratings on the need interview, one on the case history, one on the TAT protocols, as well as one on the summary confer ence, In order to test support for the complementarity hypo thesis, Winch used the proportion of correlations found to be in the predicted direction, A greater than chance frequency in that direction would therefore support the hypothesis. The findings were that on the two need interview ratings, as well as the one summary conference rating, the directionality of the correlations supported the thoery of need complementarity at the one per cent level. However, the remaining two tests were dis appointing, as they failed to support the hypothesis of greater than chance directionality. Nonetheless, since the non-support- ive findings tended to cluster near zero, Winch put forward modest claims for some support of his theory from these findings. He emphasized that whatever failure for support had been achieved, these results also did not support horaogamy of needs,^ Winch and his collaborators continued to subject the basic data from the small sample of marital pairs to exhaustive IS ^Robert F. Winch, Thomas Ktsanes and Virginia Ktsanes, "The Theory of Complementary Needs in Mate Selection: An Analytic and Descriptive Study," American Sociological Review, XIX (195*0, 2^1-2^. l6Winch, (1955), p. 552-555. 20 analysis and interpretation for a number of yearB. The efforts to compare husband-wife correlations on six hundred ratings of the need interview schedule, as well as the summary conference, With a matched' control group of twenty-five men and women who were not marital partners, did support the complementarity hypothesis for spouses as compared to non-spouses. The results on both the need interview data and summary conference data were in the expected direction, but only the former achieved signi- 17 ficance at the five per cent level. As the result of the exhaustive examination and manipulation of his data over a period of several years Winch finally concluded in his volume on mate selection that "... the bulk of the evidence supports the general hypothesis of complementary needs in mate selection, and I accept the hypothesis with the tentativenesB usual in any 18 scientific conclusion*" Winch’s theory of complementary needs in mate selection is an inherently attractive one, with considerable prima facie support from intuitive sources within the individual in our culture. Hence, it is not surprising that just as Winch spent years analyzing and interpreting the extensive data generated from his original sample of twenty-five couples, other researchers have joined the lists in the heterogamy-homogamy controversy. A ^Winch, Ktsanes, and Ktsanes, (1953), P* 508-513* O Robert F, Winch, Mate Selection (New York: Harper and Brothers, 1958), p. 119* 21 number of these are important to review here as their findings bear directly on the problem of contradictory evidence supporting both heterogamy and homogamy. Several studies antedated the beginning of Winch's work. We shall mention those which focused to some extent on some kind of matching or comparison of personality types or traits as related to mate selection. The first such study of mate selection which deserves mention here is that of Kirkpatrick who published the results of his statistical investigation of the psychoanalytic theory of mate selection. He held that personality complementarity was an indicator of neurosiB in the parties, and hence of 19 neurotic interaction in the marriage relationship. An important study by E. Lowell Kelley, using a very elaborate battery of instruments and a panel design for longi tudinal follow-up for a period of seven years left Kelly with the conclusion that to the extent that selective mating does occur, it follows the rubric of homogamy more than heterogamy. Among the instruments he administered to the three hundred engaged couples were the Bell, Bernreuter, Strong, Otis, and Allport-Vernon scales. He also asked them a host of questions 19 Clifford Kirkpatrick, "Statistical Investigation of the Psychoanalytic Theory of Mate Selection," Journal of Abnormal Psychology. XXXII (1937), **27-430. 22 about attitudes and activities in relation to hone, home life, 20 marital roles and decision making, religion, and Bocial life* Meanwhile, Strauss was gathering data from a sample of three hundred seventy-three engaged or recently married persons to test hypotheses concerning the relevance of childhood family 21 and school relationships and mate selection* Specifically, Strauss wondered if personality needs are not a direct effect of early affectional patterns with the parents so that a mate is selected with reference to these needs. Strauss further hypo thesized that these parent-related needB might be either the same needs as those previously filled by a parent, or those left unfulfilled by a parent. Moreover, since needs may not be merely an effect of early parent-child relationships according to Strauss, he also may select a mate on the basis of needs originating in the experiences of his early school years. The former hypothesis is rooted in Freudian psychoanalytic theory, while the latter is derived,from the school of the developmental psychologists. Strauss found statistical support for his primary hypothesis that mate selection is based on personality needs. He identifies these needB in descriptive, rather than analytic terms; For someone to love me; to confide in; who shows me a lot of affection; who appreciates what I have to achieve; 20 E. Lowell Kelley, "Psychological Factors in Assortstive Mating," Psychological Bulletin. XXXVII (1940), 473* 21 Anselm Strauss, "Personality Needs and Marital Choice," Social Forces. XXV (March, 1947), 332-335* 23 who understands my moods; who helps me in making important decisions; to stimulate ray ambition; whom I can look up to very much; who gives me self-confidence in my relations with people.22 During the early ^O's Christensen twice polled his students at Brigham Young University regarding their views on mate selection. He found that while males and females consider personality characteristics such as emotional maturity and dependability, males emphasize physical qualities and social skills, whereas femaleB put greater weight on achievement potential (financial, education, ambition), similarity of back- 2^ ground, chastity, and prospects for normal children. The next important study chronologically i3 that conducted 2*5 by Gray. The importance of Gray's work is twofold: first, it utilized a definite typology (Jungian), and second, it evidently attracted Winch's interest in the whole issue of complementary 26 needs. The Jungian typology employed by Gray posited a series of dichotomous variables as personality types: introversion- extroversion, thinking-feeling, sensation-intuition. Gray sum- 22 Ibid. 2^H. T. Christensen, "Student Views on Mate Selection," Marriage and Family Living. IX (19*+?), 85-88* 2l fIbid. 2^Horace Gray* "Psychological Types in Married People," Journal of Social Psychology. XXIX (19^9), 189-200. 26 Robert F. Winch, Mate Selection: A Study of Comple mentary Needs (New York: Harper and Brothers, 1938) 1 P* 151* 2k » raarized his find ngs as four combinations of these variables among the two hundred seventy-one couples who served as his subjects: (a) Three complementary powers and no like power for fifteen percent of the sample. (b) Two complementary powers and one like power for thirty- two percent. (c) One complementary power and two like powers for thirty- eight percent. 27 (d) All three powers alike for fifteen percent. From this evidence Gray concluded that there is an un conscious attraction between partners based on complementary 28 personality factors. Gray's published data were examined by Winch who failed to find statistical justification for Gray's po conclusions. The work of Winch and his collaborators has been discussed already in this chapter. Included among these publi cations was the publication in 1955 of the doctoral dissertation of Ktsanes using factor analysis as a means of measuring com- 30 piementarity of needB on the Winch sample. He hypothesized that among the forty-four variables on the needs interview* persons who recorded high on a given factor would be mated with persons who were low or negative on the same factor* Through factor analysis he derived clusters of factors he could type as 2^Gray, "Psychological Types," p. 198. 28Ibid. 2^Winch, Hate Selection (1958), p. 131* ^Thomas Ktsanes, "Mate Selection on the Basis of Person ality Type: A Study Utilizing an Empirical Typology of Person ality," American Sociological Review. XX (1955)* 55?-551» 25 follows: hostile dominance, mature nurturance, yielding depend ence, and neurotic self-depreciation. Eighty-eight percent of the sample fitted some combination of these types, leading Ktsanes to the conclusion that needs complementarity was supported Another doctoral dissertation study from the Winch sample 32 was produced by Roos* He also factor analyzed the data, claiming to find some limited support for the theory by a process of clustering separately tested factors. In these efforts, both Hoos and Ktsanes were following out the recommendations of Rosow which were made in a very thoughtful critique of the problems encountered to that time in the testing of the theory of comple- 33 mentary needs* As the Winch studies began to appear, the first researchers to attempt a re-test of the complementarity theory were Bowerman and Day*^ Using the Edwards Personal Preference Schedule (EPPS) they correlated personality variables in a sample of sixty 31Ibid. Donald E. Roos, "Complementary Needs in Mate Selection: a Study Based on R-Type Factor Analysis," (Unpublished doctoral dissertation, Northwestern University, 1956), discussed in Winch, Mate Selection, pp, 127-128. 33 ■"^Irving Rosow, "Issues in the Concept of Need-Comple- mentarity," Sociometry. XX (1957), 216-235. 34 Charles E. Bowerman and Barbara R* Day, "A Test of the Theory of Complementary Needs as Applied to Couples During Courtship," American Sociological Review* XXI (1956), 602-605* couples from the University of Washington who were either engaged or dating steadily. Since they found that less than three percent (2.2) of the possible correlation were significant in the hypothesized direction, they concluded no support for heter ogamy (complementarity of needs). However, their findings also failed to significantly support homogamous mate selection. Therefore, they concluded that whatever the true facts might be, a general theory of mate selection must be much more compre hensive than a simple dichotomy between needs for heterogamy and 35 needs for homogamy. A somewhat different emphasis was added by Kernodle in his analysis of complementary needs. He stressed social facts rather than intrapsychic needs, maintaining that the "configura tion of social and cultural phenomena1 ' is more deeply related to mate selection than the peculiar life experience of the individ- 37 ual. He drew these implications from a study of college students using Strauss' list of personality needs* From the findings that for both men and women, the most often reported need was "for someone to love me," as well as that more than half the listed needs were checked by both males and females, J Wayne Kernodle, "Some Implications of the Homogamy- Complementary Needs Theories of Mate Selection for Sociological Research," Social Forces. XXXVIII (1959)» 1^5-152. 57Ibid. ^Strauss, "Personality Needs," p. 333* 27 Kernodle concluded as follows: Persons in our culture seek a mate who will fulfill the need to have someone love them, not because they have been unloved themselves by parents, or do not have the capacity to love others, but primarily because marriage is defined in the culture as the relationship which one cannot have without being in love.39 Meanwhile Schellenberg and Bee were at work attempting to reconcile the differences in the Winch findings and those of 40 Bowerman and Day. Since Winch's need interview had been closely based on Murray's need list, and since the Edwards Personal Preference Schedule (EPPS) was also based on that list, Bowerman and Day had argued for some comparability of measurement. Actually, they claimed the EPPS was a more objective instrument than those used by Winch. But they used this instrument on an engaged or steady-dating sample, so that Winch could with some justification claim non-comparability of the sample when they came in with as much or more support for homogamy than for 4l complementarity. It was this issue of comparability which Schellenberg and Bee attacked. They did it in two ways. They chose the EPPS as their instrument, thereby maintaining test- comparability with Bowerman and Day. Then they selected a ^Kernodle, "Some Implications," p. 150* 40 James A. Schellenberg and Lawrence S. Bee, "A Re-examin- ation of the Theory of Complementary Needs in Mate Selection," Marriage and Family Living, XXII (i960), 227-252* 41 Robert F. Winch, "Comment on 'A Test of the Theory of Complementary Needs as Applied to Couples During Courtship' by Bowerman and Day," American Sociological Review. XXII (1957)1 236* 28 sample of one hundred college couples, sixty-four of whom had been recently married, eighteen of whom were engaged, and eighteen going steady. This sample provided comparability with both the Winch and the Bowerman and Day samples. They .found support for homogamy from results for the married portion as well as the unmarried portion of the total sample. However, the latter was not statistically significant, though when joined to the former, the total sample was statistically significantly 42 associated with homogamy. Since the EPPS had first been noticed as an instrument having some conceptual comparability to the constructs of Winch based on Murray's enumeration of needs, it is not surprising that researchers continued to approach the problem of need complementarity through the use of .this measurement technique. The study by Katz, Qlucksberg and Krauss utilized a modified version of the EPPS to test needs complementarity on a sample k-? of fifty-six couples. This study, though not entirely com parable on grounds of both the sample and the test, found no support for the Winch theory. Correlations of pairs of needs for couples resulted in only one in the direction of comple mentarity, whereas three were reported in the direction of homogamy. Taken separately, husbandB and wives correlated in kp Ibid. i j - 3 Irwin Katz, Sam Glucksberg, and Robert Krauss, "Need Satisfaction and Edwards PPS Scores in Married Couples," Journal of Consulting Psychology. XXIV (i960), 205-208. 29 opposite directions on several of the variables. Another study using the EPPS was done by Day who had collaborated with Bowerman on the first test of Winch's hypo- 1*5 thesis by UBe of the EPPS. In this study Day selected a sample of sixty courting couples and their same-sex friends from among volunteers in a university population. She administered the EPPS to assess need patterns of courtship couples, as well as between members of these pairs and their respective same-sex friends. Her concern was to see if she could discover comparisons between mate selection and friend selection. Day found no systematic pattern either for homogamous needs or for heterogamous needs as between mates nor between same-sex friends. Nonetheless, she did find a significant difference between the specific needs and combination of needs related to mates as compared to same-sex friends. This latter finding suggests that there is an important difference between the needs-patterns of these two intimacy relationships, courtship and same-sex friends. Murstein introduced another type of comparison group 1*7 design to his study of the heterogamy-homogamy question. 1 * 1 * Ibid. ^Barbara Day, "A Comparison of Personality Needs of Courtship Couples and Sarae-Sex Friendships," Sociology and Social Research. XXXXV (196l), k6t. . . Ibid. 1*7 'Bernard I. Murstein, "The Complementarity Need Hypothesis in Newlyweds and Middle-Aged Couples," Journal of Abnormal and Social Psychology. LXIII (1961), 19^-197. 30 Taking hie cue from Winch's assertion that unless couples were complementarily matched, they would not likely survive many years of marriage together, Murstein selected a sample of twenty recently married couples and forty-eight marriages of ten or more years duration. He administered several instruments includ ing the EPPS. His results were inconclusive, but supported homogamy on the middle-aged sample; whereas, for the newlyweds, neither homogamy nor heterogamy was supported. Murstein concluded from these results, as others had before him, that the theory of complementarity in mate selection as propounded by Winch was conceptually inadequate to encompass the variety of effects to be encountered. He asserted that such issues as the strength of needs, their variance over time, the impact of interpersonal association, and the variation in complexity of needs among individuals, cannot be circumscribed by a limited dichotomous framework such as heterogamy-homogamy. In a later paper, Murstein follows up these criticisms of Winch's theory of complementarity of needs with a more 50 complex explanatory framework he calls "SVR Theory.” He builds on the findings of Kerckhoff and Davis, who concluded that mate selection is a rather multi-faceted process of filtering in 48Ibid. 49Ibid. 50 Bernard I. Murstein, "Stimulus— Value— Role: A Theory of Marital Choice," Journal of Marriage and the Fanri.lv. XXXII (1970), Jf65-Wl. 31 51 acceptables, as well as filtering-out unacceptables. Kerckhoff and Davis, in the first major advance in theory beyond Winch, based their formulation on the results of a longitudinal study of Duke University sorority women and the partners to whom they 52 were engaged, pinned, or at least "seriously attached*" This panel of pairs was polled at the beginning of the school year and again near the end of it to see if the partners thought the relationship had progressed in the intervening seven months. A battery of tests was also administered to provide both a values profile and a personality needs scale score for each person. These constituted the independent variables as measured by 55 Farber's Value Consensus Index and Schutz's FIRO-B test* The issue of "progress" in the relationship, that is, advancement toward marriage, was the dependent variable. Kerckhoff and Davis also asked each couple to indicate the length of time they had been seriously involved with each other* And this proved to be a most rewarding criterion on which to manipulate the data* For they found significantly different patterns in the short term (less than 18 months) as compared to the long-term couples ^Alan C* Kerckhoff and Keith E. Davis, "Value Consensus and Need Complementarity in Hate Selection," American Sociological Review. XXVII (1962), 295-303* g2Ibid.. p. 295* 55 ^Bernard Farber, "An Index of Marital Integration," Socio met ry. XX (195?)t 117-13^* For additional data on validation studies of this index, see Murray A, Strauss, Family Measurement Techniques (Minneapolis: University of Minnesota PresB, 1969), pp. 98-102 and William C, Schutz, FIRO: A Three Dimensional Theory of Interpersonal Behavior (New York: Rinehart and Co.,1958)* (more than 18 months duration). They found that though their main hypothesis regarding a positive relationship between value consensus and "progress' 1 was only partially supported, it was when they turned their attention to the question of the longevity of the relationship that their most impressive results emerged. For what they discovered was that though both long and short term couples reporting progress toward permanence displayed leadings toward value consensus, the significant difference appeared bn need complementarity. For long term couples, two of the three measured of the needs variable were statistically significant, while the third, though not significant, also supported the hypothesis of an association between needs complementarity and progress toward permanence. Whereas, for short term couples, no such relationship was shown. Therefore, Kerckhoff and Davis concluded that there was support for the theory of complementary needs, but not throughout the courtship period. That on the contrary, there seemed to be. blocking of needs awareness in the earlier stages of the relationship. This led then to posit the theory of mate selection as a filtering process. This model was devised to conform to the complexities of their findings. It consists of a three stage filtering system using both homogamous and heterogamous modalities. The mate selection process begins with a homogamous filter which limitb the dating partners from among all possible candidates, to those of similar social back ground (social homogamy). Involvement increases with a given partner based to some degree on the operation of a second filter, 33 also homogamous, of value consensus. Though their data base is not as conclusive for this function, they nonetheless believe that those couples with high consensus tend to progress toward permanence more than do those couples with low consensus* There after, those couples who have passed successively through the first and second stages of the filtration provess based on homogamous criteria, are more likely to report further progress toward permanence if they are heterogamously matched on person ality needs* Kerckhoff and Davis account for the filtering process, as well as the absence of findings of needs complementarity in the earlier stages of courtship by reference to the concept of 5b romantic idealization first explicated by Waller. As Kerckhoff and Davis describe it: Short-term couples were likely to be responding to an idealized version of the love object which would make the effectiveness of any personality complementarity less pro bable. They were responding to a stylized role relationship rather than to another personality* Not until the idealiza tion is destroyed, and only then can need complementarity "make a difference" in the relationship.55 The conclusion was also supported by their findings on Farber's Value Consensus Index that short term couples were not as likely to attribute negative characteristics to their mates as were long term couples. This was confirmed by internal com- ^Sfillard Waller and Reuben Hill, The Family. A Dynamic Interpretation (New York: The Dryden Press, 195li, PP- 120-12^. ^Kerckhoff and DaviB, "Value Consensus," pp. 302-303* parisons of the shift toward negative evaluations between the 56 first and second interviews* Another pair of researchers who concerned themselves with a comparison of complementarity in two groups of couples, was 57 Hobart and Lindholm* They studied a sample of one hundred couples on the basis of need scale they generated themselves, dichotomizing for a homogamy score and a heterogamy score. They found that these couples were more similar than complementary, thereby adding evidence contrary to Winch's theory* For a com parison group, they randomly matched each spousal partner from the first sample with a non-spousal partner to test whether these random partners would be more or less complementary than the former sample. Their findings were that on all the need criteria they measured, the marital pairs were found to be less complementary than the random pairs. Though none of their differ ences were significant, Hobart and Lindholm concluded that their data suggested that married pairs are less complementary than 58 randomly matched non-marital pairs* One of the limitations of the Hobart and Lindholm study was that they did not use an already standardized instrument in their measurement of the sample couples, but instead chose to 56Ibid. 57 Charles W. Hobart and Lauralee Lindholm, "The Theory of Complementary Heeds, A Re-Examination," The Pacific Sociological Review. VI (1 9 6 3), 73-79* 58Ibid. 35 devise one of their own, a practice which*though it occasionally reeultB in the introduction of a new and powerful instrument, more often hampers scientific research in that it fragments measurement efforts, impedes comparability, and obstructs -extra polation. The choice of the EPPS by Heiss and Gordon represents the alternative practice in which these researchers chose to utilize in their study of complementarity in mate selection, an instrument which had already achieved wide use in the examination 59 of this theory. This instrument had been used earlier by the Bowerman and Day study, as well as by the Schellenberg and Bee study. In their research, HelsB and Gordon administered the EPPS to a sample of sixty-two couples who were dating, going steady, or engaged in order to test the relationship between needs complementarity and mutual satisfaction. They found only one of the fifteen needs measured by the EPPS to evoke a statis tically significant correlation between complementarity and mutual satisfaction. Therefore, since it was statistically possible that this might have happened by chance, they concluded that their results failed to support the theory of complementary needs, even though eight additional needs produced non-significant loadings for complementarity.^^ The last of the published studies to be reviewed before 59 Jerold S. Heiss and Michael Gordon, "Need Patterns and The Mutual Satisfaction of Dating and Engaged Couples," Journal of Marriage and the Family. XXVI (1964), 337-339« 6°Ibid. 36 we conclude this review of the literature, is that of Murstein's most recent formulation.^ Building on the work of his pre decessors as reviewed above, Murstein presents a three stage theory of marital choice. This is the latest in the series of efforts to expand upon the simplicity of Winch's theory of complementary needs, so as to achieve a degree of complexity of the model adequate to the comprehension of the diversity of the findings. Murstein’s theoretical design is tripartite and sequential. The three stages of this model of mate selection are stimulus, value, and role. The process of mate selection begins with the "stimulus" stage in which the basic satisfaction gained is sensory and non-interactional. The second stage, "value" is characterized by involvement in verbal interaction. And the third and final stage, "role," requires the partners to interact 62 through mutually assigned roles* Murstein describes each of these three stages in greater detail. In the "stimulus" stage: An individual may be drawn to another based on his per ception of the other's physical, social, mental, or reputa tional attributes and his perception of his own qualities that might be attractive to the other person. Because ini tial movement is due primarily to non-interactional cues not dependent on interpersonal interaction, these are cate gorized as "stimulus" values. ^3 ^Hiurstein, "Stimulus— Value— Role," pp. 465-481, ^Ibid.. p. 465? ^Ibid.. p. 466. 37 The second stage is called the "Value" stage, and is described by Murstein as follows: Assume for the moment that mutual "stimulus" attraction has occurred between a young man and woman at a "mixer" dance, and that they sit down and talk to each other. They are now entering the second stage, that of "value comparison." Un like the "stimulus" stage in which attributes of the partner are evaluated without any necessary interpersonal contact, the value comparison stage involves the appraisal of value compatibility through verbal interaction. The kinds of values explored through discussion are apt to be much more varied than those possible in the "stimulus" stage* The couple may compare their attitudes towards life, politics, religion, sex, and the role of men and women in society and marriage. The fact that the couple is now interacting also permits more continuous and closer scrutiny of physical appearance, as well as other important factors such as temperament, "style" of perceiving the world, and ability to relate to others. The third stage is called the "role" stage, and is des cribed by Murstein as follows: There are many tasks which face the couple in the "role” stage before they move into marriage . . . the author's conceptual framework and data collection dictate limiting the analysis to three broad areas; perceived role "fit," personal adequacy, and sexual compatibility. Concerning role "fit," . . . as the couple's relationship ripens, the members increasingly confide in each other and, thus, become aware of a broader range of each other's be havior than heretofore. A second task is to . . . measure . . . personal adequacy • » ., moodiness, inability to make decisions, dislike of the self, and neuroticism. ... The third task in the "role" stage involves the neces sity of attaining sexual compatibility whether by achieving a good sexual relationship in practice or by agreement as to the degree of sexuality which will be expressed during the "role" stage prior to marriage. 64 Ibid.. p. 468. 65Ibid., pp. 469-470 Having described the three stages in the development of courtship, Murstein then tested a series of nineteen hypotheses related to his theory* He found that all of them received at least moderate support, concluding that; Although the data offer considerable support for SVR theory, the author is aware that consistency of the theory *.»*! 1 1 * m « <9 • > ! m + a w o l * ? ^ o 4 * a A 4+ O © with existing data is not sufficient to validate it. With reBpect to the issues of homogamy and heterogamy, Murstein's research supported homogamy on the first and second stages, "stimulus" and "value," respectively. However, in the third stage, "role," the criterion becomes more complex, includ ing role perceptions of the self, the ideal self, the partner, and the ideal partner. For our purposes his salient hypotheses on this issue are as follows: Couples with high self-acceptance view their partner as significantly more similar to themselves than couples with low self-acceptance. Individuals tend to choose partners whose level of self-acceptance is similar to their own.®? These two hypotheses of Murstein, confirmed by his data, lead to the following set of propositions underlying Miller's and this author's basic hypothesis regarding discrepant findings for both homogamy and heterogamy. These propositions are: 1. High self-acceptance persons will choose high self acceptance partners. 2. Low self-acceptance persons will choose low self acceptance partners. 66Ibid.. p. 479. 6?Ibid.. pp. 471-473. 3. High self-acceptance couples, seeing themselves as similar to each other, will be homogamous. k. Low self-esteem couples, seeing themselves as dis similar, will be heterogamous. Though Miller did not include a measure of self-acceptance in hiB study, and therefore these propositions will not be tested directly in this replication, they will be seen to form a part of the propositional complex surrounding and supporting the theoretical framework to be developed in the following chapter. Since these most recent findings did not appear until last year (1970), the comment with which we conclude this review of the literature was not meant to encompass it in the estimate of the state of our knowledge about the theory of complementarity in mate selection. However, it seems not unfair to summarize the field, even including the latest efforts, as Bolton did writing as far back as 1961* We know very little, scientifically, about mate selection either as a process or as a relationship as such — that is, as a love or intimate relationship. After decades of cor relation studies the theoretical dispute between the theories of homogamy and heterogamy (or theory of complementary needs) is unresolved. While, on the surface, homogamy appears to have the stronger case, the evidence is actually indecisive. Even taking the contributions of both the homogamy and complementarity theories, much if not most of the variance in mate selection continues to be unexplained. 68 Charles D. Bolton, "Mate Selection as the Development of a Relationship," Marriage and Family Living, XXIII (1961), 23^-240. CHAPTER XII THEORETICAL FRAMEWORK The Ideal Type as a Sociological Construct Before considering specifically a typology of marital roles it may be appropriate to define the term "typology" as it is used in this and most sociological research. The term "ideal type" was introduced into sociological thought by Max Weber. It is a conceptualization of a phenomenon in terms of a finite number of defined variables, explicitly stated. "Real" phenomena are ordinarily so complex that an unmanageable number of vari ables and their interrelationship would have to be used for an accurate description of them. Therefore an ideal type of the phenomenon is constructed. As described by Martindale, Ideal types are not general or abstract concepts, but hypo thetical individuals consisting of a selection of itemB which could appear in reality. . . . ^hey are hypothet ically concrete individuals (personalities, social situations, changes, revolutions, institutions, classes, and so on), con structed out of their relevant components by the researcher for the purpose of instituting precise comparisons. . . . The ideal type is a strategy in empirical explanation. It is framed in terms of the scientific knowledge available to the researcher at the time of his study and in terms of the empirical situations he is trying to understand. There are two criteria for constructing an ideal type: objective possibility and adequate causation. • . ^Don Martindale, The Nature and Types of Sociological Theory (Boston: Houghton Mifflin, l'SMjQ), pp. 3&1-383 passim. 40 *fl A* more dictionary-like definition of the term is given by Hose: Ideal type: an abstract concept of a class of phenomena arrived at by specifying what are thought to be the key characteristics common to these phenomena and exaggerating these characteristics to the extreme. Thus an ideal type indicates clearly the main characteristics of a class of objects| but such ideal types as unities are too exaggerated to be found in real life.^ The phenomena to be studied as ideal types in this research are "marital roles." A typology of marital roles refers to the cluster of variables which differentiate the two major types of husband-wife interaction: the traditional- institutional , and the democratic-companionship. A Typology of Marital Roles Ernest W. Burgess' definition of the family as "a unity of interacting persons. * . ."is the best base from which to launch our theoretical statement, for it succinctly alludes to both framework and the process, the structure and the function, of the marital dyad (and later, those additions which render it a family).^ Expanding from this basic definition Burgess Arnold M. Bose, Sociology (2nd ed*, revised; New York: Alfred A. Knopf, 1965)* p* 72^. Burgess and Locke, The Family, p. 7* bz Locke identify two (ideal) typee of marital or family inter action. The Traditional-Institutional Role The older form is the "institutional family" in which the husband as head of the household wields undisputed authority over the other members and unity is maintained by the willingness to do one's duty, fulfill one's obligations and help carry out the essential functions of the family. This form is still a major style of family organization in America today. The unity of the traditional institutional family is based on control by the father and by the social matrix of obligation and duty which supports him in his role of controller* A detailed look at the traditional form of family organi zation will show clearly the social sources of the psychological needs met by this type of pattern* The traditional institutional family is organized on a vertical axis with the husband-father at the top of the system and the wife-mother and all subsequent members of the family subordinated to him in some one or another of several possible inferior positions* The husband-father ful- i f . fills the roles described by Parsons as "instrumental*" That is, he functions as the bread-winner, authority-figure, disciplinarian, and spokesman-contact for the larger social system of the external world* This has been the traditional Parsons and Bales, Family* p* b?. ^3 pattern of family organization for most of recorded history. It is based on the biological differences between male and female, 5 as well as on age differences within the family. Founded on these age and sex differences, the form of the family has been supported by certain social patterns of the larger society. Laws regarding property, voting, and inheritance have until rather recently supported the husband-father's claim to dominance- control over both his wife and his children. According to a widely used text on the law of domestic relations published in 1905, the situation was thusi In general, marriage imposes, at common law, no sidabilities whatever upon the husband except that he cannot deal directly with his wife he may not dispatch her into the next life, nor inflict unreasonable punishment upon her nor testify as a witness for or against her. He may contract, sue or be sued, and acquire, hold, and, in general dispose of, property in the same manner and to the same extent as if sole single . The marriage does not affect his status, identity, or capacity before the law. In contrast to the husband's status, the wife is affected as follows: At common law a woman, by marriage, loses, to a considerable extent, the legal capacity which she possessed as a feme sole. Her disability in some cases, as, for example, in respect to the making of contracts, is practically complete; in other cases, as in respect to making wills of personality, it is partial only. These disabilities are the result, in part, of the legal merger of wife's existence into that of her husband, and in part of the presumed control exercised over her by him. Her disabilities in some cases, perhaps, work a 5Ibid., p. 313. ^Joseph H. Long, A Treatise on the Law of Domestic Relations (St. Paul: Keefe-Davidson Company, 1905!), p. 170, hh hardship upon her, but in most cases they result simply in inconvenience, and, on the whole, their general effect is to protect the wife. Moreover, for these disabilities she is liberally compensated by the obligations which the marriage imposes upon the husband to provide for her support during the coverture. . . . She has also many exemptions from civil and criminal process to which he alone is liable, although both may have participated in the benefit of the contract or com mission of the crime, during the continuance of the matri monial connection. . .. These personal disabilities the common law imposed partly for that of the wife.7 The picture emerges then of a traditional institutional family in which the husband is the provider, and the wife's place is in the home; in which the husband is the legal agent for the family, and the wife must rely on him for legal status; in which the husband is the decision-maker, and the wife may not oppose him; in which the husband is protector, and the wife and children dependent upon his protection; and in which the husband is the disciplinarian, and the wife and children subject to his 8 commands and punishment. In spite of the fact that major revisions of the legal framework governing the family have occurred in the past generation, and indeed are occurring momentarily, marital role expectations are still heavily influenced by the traditional family pattern just described. There seems to be a residual 9 role-expectation supporting this system. 7Ibid.. p. 171. g Parsons and Gales, Family, p. 313* 9Ibid., p. 339. The Democratic-Companionship Role The newer form of family pattern is the one which Burgess and Locke call the "companionship" family.^ It is characterized by an assumption of equality of husband and wife, democracy in family decisions, and a maximum of freedom for self expression 11 and personality development. This family form is one in which the unity iB based not on autocratic control, but on consensus. How the shift from institutional to companionship forms of family system has come about, and is coming about is perhaps best described by Evelyn Duvall: As the family ceased being the center of industry, the father left his role as family foreman and took his place beside other workers outside the family. More recently women, stripped of a large part of their domestic duties, have followed their husbands and children to work. Home no longer has a single dominant head who keeps tabs on the doings of every member. Discipline is through self-control. Children are brought up with the freedom to make their own choices. Family decisions are jointly made.^-2 In the democratic family a whole different set of roles is essential for effective functioning within the family. These roles are based on a set of behavioral norms just as are those of the traditional-institutional family system. Norms will be outlined in ideal typological manner so as to facilitate a comparison between the two types of family system. 10 Burgess and Locke, The Family, p. 18. n ibid. ^Evelyn M. Duvall, "Growing Edges in Family Life Educa tion." Marriage and Family Living. VI (19^*0, 21. bS Norms Supporting the Democratic-Companionship Role Non—deetruction ±8 the first and most basic norm* It is a recognition by both the husband and wife that neither partner will — nor in fact can — destroy the other* nor himself. This means that threats* actual or implied, of massive aggression, whether it be physical, psychological, or social, against the other, or against the self, are not effective. This is in contrast to the traditional-authoritarian family where the husband is generally considered to hold the power of life and death over his wife and children. (That is not to say that wives do not protect themselves from this threat by effective measures of an indirect, passive-aggressive nature, such as talcing to their bed with protracted illnesses, or at least enjoying 'delicate' health, eto. These are extreme examples. The more moderate and common ones consist of withholding certain domestic and conjugal pleasures as bargaining points for their own advantage.) The wider social system now supports the individual, at least minimally, in his claim to separate existence - economically, politically, psychologically, etc. Therefore, the individual may interact in marriage with a certain confident autonomay, knowing that though divorce or separation may not be a very pleasant alternative, he does not need to be destroyed by the marriage simply in order to escape destruction in its termination. Limited Contract is a norm based on the acceptance of the former norm, non-destruction. Since each partner recognizes that he can neither destroy nor be destroyed by the other, there is no psychic nor social necessity to demand am unlimited contract* It iB unnecessary for each partner to be "everything" to his mate* Together they may choose what they wish to include in their contract, as well as what they wish to leave out. Since a satisfactory marriage is possible between a toothpaste tube squeezer and a toothpaste tube roller, they can agree not to include in their contract of marital behavior the one-and-only right-and-true method of extracting toothpaste from the tube* And in the final analysis, should it come to that, they are able to terminate the contract. That is, it is limited in a temporal sense if they choose it to be so, knowing that this decision to terminate, painful though it may be, will not "destroy" in any legitimate sense, either partner. By contrast, the traditional- authoritarian family system relies heavily on the psychic images of potential destruction to require compliance. When the macro system surrounding the family no longer supports the vertical distribution of this marital power, then, as is the case today, such power-based families may become increasingly unstable. Disagreement is acceptable because, since it is not a harbinger of destruction nor evidence of disrespect or lack of love, it can be acknowledged without fear of disaster. And since the contract between the two partners is limited, they may choose to leave outside the contract certain items or areas of life on which they disagree. For instance they may limit their marital contract to exclude religion or politics, upon kS either of which they may acceptably disagree. However, there is one stipulation regarding the acceptability of disagreement, and that is that when they discover a disagreement, they must decide together what they wish to do regarding this disagreement. Attempting to compel the other partner to change his basic position on the disagreement is not one of the options, as it carries the implication of a vertical (superior-inferior) relationship between them. Reliance on historical causes is relinquished. That is, the partners in an autonomous democratic-companionship marriage do not claim that their behavior is "caused" by the antecedent behavior of the other. Reference to historical incidents in the courtship and marriage is utilized only for elucidation, never justification. By contrast, in the authoritarian traditional- institutional family system, justification of one's own blame worthy behavior is customarily sought in the defective behavior of others which is claimed to have "caused" it. When recourse to this kind of historical causation is common, it is seldom possible to focus on the here-and-now of the present in a truly cooperative and problem-Bolving manner. A here-and-now focus is the final essential norm for be havior in the democratic-companionship marriage. Since the indisputable status and role of "head" of the household is abolished, and in fact, irrelevant in the democratic-companion ship family, the chief integrative force is a commitment to the above norms including a present-oriented willingness to work creatively and constructively on the issues of the Moment - not in a vacuum, but with a recognition of the impact of the past and the prospect of the future* The five behavioral norms which form the conceptual frame work of the democratic-companionship family, as well as their opposites which comprise the authoritarian-traditional-instutional family framework are operationalized in the test instruments. (This operationalization will be discussed in detail in Chapter IV, p. 78)* To recapitulate the main thesis in reference to the ideal typologies just presented of the authoritarian traditional- institutional family and the autonomous democratic-companionship family! it is assumed that persons socialized to the traditional type family will seek a partner with complementary needs, whereas persons socialized to the democratic family will seek a partner with homogamous needs. It is the contention, as it was of the Miller research on pre-nuptial partners, that the interpersonal needs of those marrying for institutional or control motives will be comple mentary, whereas those marrying for companionship or consensus motives will disclose homogamous interpersonal needs as measured by the instruments to be used. To place the theoretical statement in more axiomatic form it would be stated as follows: 1. The self responds to another in terms of his percep tions of that other, rather than in terms of objective reality. 2. The self selects a mate whom he perceives to have the greatest potential for providing him with maximum need gratification# 3* When the self is able to predict another’s percep tions accurately, he will be better able to select a mate whose need gratification pattern is reciprocal. To summarize the theoretical statement, the following propositions are submitted: 1. Among couples with attitudes toward their marital roles of the traditional type will be found all those who express complementary interpersonal needs. 2. Among couples with attitudes toward their marital roles of the equalitarian type will be found all those who express homogamous or similar interpersonal needs. 3# Among couples with attitudes toward their marital roles which are mixed (one traditional and the other equalitarian) will be found all those who express mixed interpersonal needs. Among couples with traditional attitudes toward marital roles, those in which both partners accurately perceive the other's role attitudes will have the greatest interpersonal need difference or complementarity. Among couples with traditional attitudes toward marital roles, those in which neither partner accurately perceives the other's role attitudes will have the least need difference or complementarity of any couples in the traditional group. Among couples with traditional attitudes toward marital roles, those in which one partner only accurately perceives the other's role attitudes will have less need differences than those traditional couples where both partners' perceptions of the other are accurate, but more than those traditional couples where neither is accurate. Among couples with equalitarian attitudes toward their marital roles, those in which both partners accurately perceive the other's role attitudes will have the least need difference (homogamy). Among couples with equalitarian attitudes toward their marital roles, those in which neither partner accurately perceives the other's role attitudes will have the greatest need difference (or least homogamy) of any couples in the equalitarian group. Among couples with equalitarian attitudes toward marital roles, those in which one partner only accurately perceives the other's role attitudes will have more need difference than those equalitarian 52 couples where both partners* perceptions of the other are accurate, but less than those equalitarian couples where neither is accurate. The Hypotheses I, The difference between the interpersonal needs of marital partners will vary as a function of their marital role attitudes. II. The difference between the interpersonal needs of marital partners will vary as a function of the interactional effects of their marital role attitudes and their accuracy of perception of the mate's marital role attitude. III. There will be a negative correlation between the inter personal needs of traditional type marital partners. IV. There will be a positive correlation between the inter personal needs of equalitarian type marital partners. V. There will be a positive correlation between the differ ence of interpersonal needs and accuracy of perception of mate's marital role attitude for marital partners of the traditional type. VI. There will be a negative correlation between the differ ence of interpersonal needB and accuracy of perception of mate's marital role attitude for marital partners of the equalitarian type. CHAPTER IV RESEARCH METHODS The Sample The sample was drawn from the population of metropolitan areas of southern California, primarily southeastern Los Angeles County, Orange County, and a few from San Diego County. Following the criterion of comparability, the same socio-economic and ethnic measures were applied in this study as were applied by Miller. That is, the sample is limited to middle-class Caucasians of non- Jewish religion.^ The socio-economic criterion of middle class ness was presumed to be measured by Hollingshead*s "Two Factor p Index of Class Position. The two factors of class - education and occupation - were weighted and combined according to a derived formula developed by Hollingshead. The sample was gathered through the University of Cali fornia Extension classes for adults in the field of divorce and remarriage taught by this author at the Irvine and San Diego campuses of the University. A few of the respondents were class ^Miller ruled out those of Jewish ethnicity on grounds that the "strong familial allegiances of JewiBh people" might contaminate the sample. Miller, "Familial Roly Typology," p. 75. 2Ibid.. p. 78. 53 members. Many of the class members who were not themselves eligible to answer the questionnaire since they were not currently a partner in a remarriage, volunteered to ask friends of theirs who did meet this criterion to participate in the-study. This . created a methodological problem in that subjects were then not in direct contact with the researcher, but were asked to mail in the completed forms to the author. A pre-stamped self-addressed envelope was provided for this purpose with each questionnaire. Of the total of one hundred four pairs of questionnaires given out to class members to complete themselves or distribute to an eligible couple of their acquaintance, a total of seventy-eight were represented in the returns. Of those returned, two pairs were entirely blank, signifying refusal of the target couple to participate in the study. In addition, we may infer from non return of twenty-six pairs of forms that a total of twenty-eight couples probably refused to participate. However, this parameter Is somewhat unclear as some class members may have approached more than one eligible couple before gaining a participant pair, while other class members or target couples themselves may have had second thoughts after examining the questionnaire, and simply discarded it instead of sending it back blank or completed. Since there was no academic coercion (grade-in-the-course, completion- of-requirements, etc.) attached to the distribution of the forms, a lower rate of return than otherwise would be expected was achieved. Incomplete units occurred in ten instances in which only 55 one partner responded (ten units). Of the ten partial units only the wife returned hers in eight cases, and only the husband in two. Since anonymity was maintained, no direct follow-up could be undertaken with the incomplete units. However, an indirect follow-up was done through a personal postcard mailed to all class members one month after distribution to request that they re-contact their target couple to ask that the forms, if not already mailed in, be returned to the author, completed or not. This probably accounts for the higher than average ratio of return for mail questionnaires. A number of important factors affect the rate of return of a questionnaire such as this: sponsorship, format, length, ease of response and return, and target population. In this caBe, the brief request for coopera tion of the respondent in the study "to learn more about the attitudes and values of people who are married," indicated i f sponsorship of the University of Southern California. The format of the questionnaire was as simple, short, and attractive as it could be made without sacrificing content. All answers on tlie four instruments required nothing more than a simple check mark in the appropriate box, except on the biographical data page where a few dates were requested. As for the ease of return, a 3 Claire Selltiz, Marie Jahoda, Morton DeutBch, and Stuart W. Cook, Research Methods in Social delations, revised in one volume, (New York: Henry Holt and Co., Inc., i960), p. 2^1f* See specimen, Appendix A, p. 129* self-addressed envelope with an individual stamp (not metered, etc.) was provided with, each questionnaire. Such evidences of individual personal attention it was hoped would enhance motiva tion of the respondents to complete their task. The target population in this case was middle-class Caucasian and could therefore be expected to have some acquaintance and facility with forms, blanks, and questionnaires. Also, each target couple was approached by a personal friend with the request to participate in this project. Therefore, it is not surprising that this project obtained a mailed return of seventy-five per cent as compared to the more characteristic rate of less than 5 fifty percent. The social characteristics of the sample to be discussed in this section are age, class, education, religious affiliation, and previous marital status. Age The distribution of ages of the males and females is presented in Table 2. While the mean age of the males is slightly older the difference is not significant (t = 1.^5). In sin attempt to obtain accurate ages, the respondents were 5 Fred N. Kerlinger, Foundations of Behavioral Research (New York: Holt, Rinehart and Winston, Inc., 196£ i) , p. 397. For a more detailed discussion of mail questionnaires, see M. B. Parten, Surveys. Polls, and Samples (New York: Harper and Row, 1950), Chapter 11, pp. 391-^02; see also W. Goode and P. Hatt, Methods in Social Research (New York: McGraw-Hill, 1952), Chapter 12. TABLE 2 AGE DISTRIBUTION OF SAMPLE, BY PERCENTAGES AGE Males N = 63 Females N = 63 20-24 6.3 ll.l 25-29 12.7 23.7 30-34 20.6 ll.l 35-39 17.5 15.9 I f 0-44 11.1 17-5 45-49 14.3 ll.l 50-54 12.7 3-2 55-59 1.6 3.2 60-64 1.6 3.2 65-69 1.6 - 100.0 100.0 Mean 39.5 36*8 Standard deviation 10.8 10.6 Median 38.0 35*8 Not significant (t = 1.4) p .1 This table corresponds to Miller's TABLE 4 in Appendix B. 58 queried indirectly. The subject was requested to fill in his birthdate. Age was then determined as of December 31* 1970, figuring to the nearest tenth of a year. Thus a birthdate of July 1, 1940 corresponds to 3°*5 years of age. In addition to providing more accurate ages, the extra information yields more reliable estimates of the difference in ages of each couple. The mean age difference is 2.67 years. To assess the comparability of this sample to the Miller sample, the age at first marriage is significant. In the Miller sample, the median age for the males was 22.9, while median age for those males in the sample for whom it was a first marriage was reported as 22.6. In the present study, the median age for males at first marriage was 25*4. For females in the Miller sample the median age was 20.9, while the median age for those females in the sample for whom it was a first marriage was reported as n 20. In. the present study, the median age for females at first marriage was 22*5* Considering the fact that the present sample is of a somewhat older age cohort than the Miller sample and therefore contracted their first marriages at a time when the median age at first marriage was somewhat higher than it is currently, it would appear that on this criterion the two samples are comparable. ^Miller, "Familial Role Typology," p. 84. ' ' ' ibid., p. 87. 59 Social Position The social position of each marital unit in the sample waB obtained by combining an educational factor and an occupational 8 factor to indicate the social position of the male. This is the method used by Miller following Hollingshead's "Two Factor Index of Social Position." According to Hollingshead's formula, educational and occupational levels are each scored on a seven point scale with low values indicating high positions, and high values indicating low positions. A factor weight of four is assigned the educational scale score and a factor weight of seven is the occupational scale score. The two scores are multiplied by their respective factor weights and then combined to provide a single score for a position on this index of social class. The following example is Hollingshead's: John Smith is the manager of a chain supermarket. He completed high school and one year of business college. His Index of Social Position is computed as follows: Factor Scale Score Factor Weight Score X Weight Occupation 3 7 21 Education 3 ^ 12 Index of Social Position 10 Score 33 8 Were this not a replication study, we would have utilized the social position scores of both the husband and wife to pro vide a combined score for the marital unit on the assumption that each had a status to be considered. q . August B. Hollingshead, Two-Factor Index of Social Posi tion (New Haven, Conn.: Privately published by the author, 1957)• ^Ibid., p. 10. 60 On. the occupational scale Mr. Smith receives a score of 3t since his job is found in Hollingshead1s list of occupations under section Three, "Administrative Personnel, Small Independent Businesses, and Minor Professionals," under the title "Store Managers (Chain)• On the educational scale he receives a score of 3 also since Hollingshead ranks educational achievement in the following way: Educational Attainment Score Graduate Professional Training 1 Standard College or University Graduate 2 Partial College Training 3 High School Graduate A - Partial High School 5 Junior High School 6^ less than Seven Years of School 7 Then by multiplying these scores of 3 each by their respective factor weights of 7 (for occupation) and b (for education), we derive a combined score for his Index of Social Position: 33« By consulting Hollingshead's paper again we find that he assigns the following groupings to total scores: Social Class Range of Computed Scores I (Upper) 11-lk II (Upper-Middle) 15-2? Ill (Middle) 28-if3 IV (Working) V (Lower) 61-77 ^ ^Ibid., p. 12Ibid., p. 9. 1; 5Ibid,, p. 10. 6l Therefore, in the example cited by Hollingshead, Mr. Smith rates as in Social Class 3 (28-4-3) with a score of 33* In the sample drawn for this study, as in the Miller sample, only those couples in which the husband score.d in either Class 2 or Class 3 were eligible for inclusion. Table 3 presents lif the social class positions of the sample dyads. In the present study, sixty-three coupleB were found to be middle- or upper- middle class based on husbandfs score. It is worthy of note that there are a number of females in the sample with social class positions higher than their mates. Twenty-eight females (4-3 percent) had social class ratings of #2 when their mate was #3* whereas only thirteen males (21 percent) were higher than their mate. Thus nearly one-half the sample (43 percent) of couples reflected dyads in which the female partner was higher on social class than the male. Occupation The previous section examined the class composition of the sample but did not present the two factors which comprise the basis for that class position. This section delineates the occupational factor and the following section the educational. Hollingshead uses a seven point scale to classify a subject's occupation with low values indicating a higher occupational status. In his paper can be found detailed lk The table of Social Class Positions for the Miller sample is placed in Appendix B for purposes of comparison, p. 148. 62 TABLE 3 SOCIAL CLASS POSITIONS OF MALES AND FEMALES FOR THE SELECTED SAMPLE, BY PERCENTAGES Males Females Class N = 63 N = 63 1 2 39.6 57.1 3 60.4 24.0 k - 6.3 5 - NR - 12.6 100.0 100.0 Median 2.7 2.1 NR = no response or females who designated housewife as an occupation This table corresponds to Miller's TABLE 1 in Appendix B. 63 descriptions and numerous examples for each of the seven posi tions., however only the major classifications will be given here. Position 1 includes a. Higher executives b. Proprietors of large concerns (value over $100,000) c. Major professionals Position 2 includes a. Business managers in large concerns b. Proprietors of medium concerns ($35,000-$100,000) c. Lesser professionals Position 3 includes a. Administrative personnel b. Small business owners ($6,000-335,000) c. Semi and minor professionals d. Farm owners ($25,000-335,000) Position k includes a. Clerical and sales workers b* Technicians c. Owners of little businesses ($3,000-36,000) d. Farm owners ($10,000-320,000) Position 3 includes a. Skilled manual employees b. Small farmers Position 6 includes a. Machine operators b. Semi-skilled employees c. Farm tenants Position 7 includes a. Unskilled employees b. Share croppers c. Unemployed (no occupation) 6^ Table k presents the distribution of occupational scale scores for the sample, by sex."^ It is of interest that while the males appear evenly represented in the middle and upper middle class occupational positions, the females appear heavily concentrated in the second and fourth positions. The femaleB who were assigned occupational scale score 2 are mostly teachers (89 percent of the females who scored 2 are teachers). Those who were assigned scale score sales the other large group were mostly secretaries and sales workers. Education According to Hollingshead's formula, education is scored on a seven point scale. The scale Is based on the number of years of schooling completed by the subject, the smaller the scale score the greater the number of years of education. Scale Years of School Completed 1 Professional (M.A., M.S., M.D., Ph.D., L.L.B) 2 Graduate of four year college (B.A., B.S., A.B., B.M.) 3 1-3 years college and business schools Graduate of high school 3 10-11 years of school 6 7-9 years of school 7 Less than 7 years completed The distribution of the sample, by sex, on the education 16 scale is presented in Table 5. Approximately 76 percent of 15 ^The table of Occupational Positions of Males in the Miller sample is placed in Appendix B for purposes of comparison, P« See Appendix B, p..137 , for the Miller sample table of Educational Positions. 65 TABLE h OCCUPATIONAL POSITIONS OF THE SAMPLE, BY SEX Occupational Males Females Position N = 63 Percent N = 63 I 9.5 2 33. 5^.0 3 20.6 if.8 k . 22.2 23.8 5 l*f.3 1.6 6 - 3.2 7 - - NR - 12.6 100.0 100.0 Mean 3*0 5*0 Median 3*3 1*8 NR = no response or women who wrote housewife as their occupation This table corresponds to Miller's TABLE 2 in Appendix B, 66 TABLE 5 EDUCATIONAL POSITIONS OF THE SAMPLE, BY SEX Educational Position Males N = 63 Percent Females N s 63 1 31.8 38.0 2 22.2 2 8.6 3 22.2 14.3 4 20.6 15.9 5 3.2 3.2 6 - - 7 - - 100.0 100.0 Mean 2.4 2.2 Median 2.8 2.4 This table corresponds to Miller's TABLE 3 in Appendix B. the males and 8l percent of the females had one or more years of college. As on the occupational measure for social position, the females in the sample rate higher than the males. Length of Present Marriage The results from the sample on the length of present marriage are shown in Table 6. The median length of marriage was nearly eighteen months. Responses range from less than one month to months, but the majority, 77*3 percent of the couples have been married less than three years. Religious Affiliation Though religious affiliation was not one of the hypothe sized variables in this study, nor in the exploratory study by Hiller, it was utilized as a control factor to limit eligibility of couples gathered for the sample. Tables 7 and 8 display the religious preference and mixture of the sample. On this measure, the overwhelming majority of both males and females identified themselves as Protestant: approximately sixty-four percent for males, and seventy percent for females, giving a combined total of approximately sixty-seven percent. In this matter of religious affiliation the sample of the present study and the Hiller sample are not as closely comparable as they might have been, since Hiller had a much higher representation of Catholics in his sample, as well as many fewer respondents who reported themselves as "other" or "none." The Table of Religious Preferences for the Hiller sample is placed in the Appendix for purposes of TABLE 6 LENGTH OF PRESENT MARRIAGE IN YEARS, BY PERCENTAGE Length in Years Per Cent Less than 1 39.7 Less than 2 23.8 Less than 3 23-8 Less than k 12.7 100.0 Median 17*6 months Cl.^7 year) 69 TABLE 7 RELIGIOUS PREFERENCES OF THE SAMPLE BY PERCENTAGES Religious Preference Male N = 63 Percent Female N = 63 Total Catholic 7.9 11.1 9.5 Protestant 63.6 69.8 66.6 Other 3-2 3.2 3.2 None 25.3 15.9 20.7 Jewish - - - 100.0 100.0 100.0 This table corresponds to Miller's TABLE 5 in Appendix B TABLE 8 MIXED RELIGIOUS MARRIAGES, BY RELIGION Type Number of Couples Protestant - Catholic 5 Protestant - Other 3 Protestant - None 6 Catholic - None 1 Mixed Marriages 15 Recent Mixed Marriages 2k 71 17 comparison. The Instruments Used Each member of the marital pair completed the following items; Cl) a face sheett giving basic statistical and biograph ical information about the person, as well as the marriage; (2) an inventory of motives, the Leary Interpersonal Check List; (3) a marital roles scale, The Marital Role Typology Scale, on himself, and C^f) an estimate of his partner*s self-scoring of l8 the same marital roles scale. The Face Sheet Purpose. The biographic and demographic information necessary to identify the partner, as well as the unit, as eligible for inclusion in the sample was obtained in a series of twenty-four items. These were designed to be short answer type responses, easily converted to IBM punch cards. This information included sex, age, occupation, income, education, ethnic origin, and religious preference. Data regarding the marriage were also gathered. These included date of present marriage, dates and duration of previous marriages, length of engagement to present spouse and number of children in the home* Finally, a few questions relating to friendship and acquaintance 17See Appendix B, p. 1 ^See Appendix A, p. 129* 72 19 patterns were included. The Interpersonal Check List Purpose. The Interpersonal Check List, or ICL, as it is commonly known, was developed by Timothy Leary, a psychologist who headed a research team of psychologists, psychiatrists, and other mental health professionals at the Kaiser Foundation PO Hospital, Oakland, California. The purpose of the ICL is the assessment of personality and diagnosis of psychic phenomena through the use of an interpersonal scheme of measurement. Based on the system of clinical social psychiatry germinated by Harry Stack Sullivan, and developed by Karen Horney, Eric Fromm, and others, the ICL has its conceptual roots in the work of 21 social theorists Charles Horton Cooley and George Herbert Mead. Such a frame of reference evokes a description of personality in terms of interpersonal operations (interpsychic processes) rather 19 Since the survey research process is a costly and time consuming one, it seemed worthwhile to collect a small additional amount of data from the sample couples even though there was no intention of analyzing it in the course of this project as it did not relate directly to the main hypotheses being re-tested here. 20 Timothy Leary, The Interpersonal Diagnosis of Personality (New York: The Ronald Press, 1957^- 21 Charles Horton Cooley, Social Organization (New York: Scribners, 1915)* Eric Fromm, Escape From Freedom (New York: Rinehart, 19*H)« Karen Horney, Neurosis and Human Growth (New York: W. W. Norton and Co., 1950)* George Herbert Mead, Mind. Self, and Society (Chicago: University of Chicago Press, 193^). Harry Stack Sullivan, The Interpersonal Theory of Psychiatry (New York: W. W. Norton and Co., 1953)* 73 than in terms of strictly organic and intrapersonal conditions (intrapsychic states)* The utility of such a formulation is at once obvious, since it leads to emphasis on readily identifiable sets of behavior for measurement of "personality." According to this system, personality is an organization of sets of behavior developed in response to interactions with others, and integrated according to conformity with a self- concept forged from these interpersonal stimuli. In short, it is the "dynamic interaction of the self with the social environ- 22 ment. The person understands himself to "be" the sum of his "behaviors." That is, he comprehends himself as the result of the cumulative effect of the roles he takes in response to cues (expectations) from others, as well as the expectations he has of himself and others. These role expectations are understood as role perceptions and attitudes, and may be identified as discrete entities through the linguistic device of descriptions of interpersonal behavior. Leary's account of this process is as follows: After extensive informal surveys of the many varieties of data, a list of several hundred terms for describing inter personal behavior was assembled. The next task was to sort through the long lists of terms and to determine the generic interpersonal motives. . . . The gradually developing lists of generic terms were then combined to eliminate overlaps and repetitions until a list of sixteen generic interpersonal motivations resulted. All of the original terms - which 22 Franz Alexander, "Psychoanalysis in Western Culture," American Journal of Psychiatry. CXII (March, 1956), p. 699* ?k numbered several hundred - could be expressed as differen- tiated varieties of the sixteen basic interpersonal themes. Next Leary and his team organized these sixteen themes into a two dimensional scheme related to power and affiliation. The eventual result of this surveying and sifting process was the construction of a circumplex schema of personality having two principal axes and eight major sectors, each of the latter 2k incorporating two of the sixteen themes or motivations. The two axes dichotomize the circle of personality (basic inter personal behaviors) into hemispheric dimensions of dominance- submission (top half-bottom half) and affiliative-hostile (right half-left half). Such a division of the circle provides four quarter-circle constructs of interpersonal role attitudes. In Figure 2, Appendix D, beginning with the upper right quadrant, the theme is "affectionate dominance," followed clock-wise by "hostile-dominance," then "hostile weakness," and finally, "affectionate-submissiveness*" Each of these quadrants contains two octants and four of the sixteen types of role descriptions. An analysis of the "affectionate-dominance" quadrant will provide an example of the entire theoretical construct. The four types of role descriptions in this quadrant 2*5 Leary, Interpersonal Theory, pp. 63-6^. oh See Appendix D, p. 16j. 25 The term "affectionate" is used in this discussion instead of the more generic term, "affiliatlve" because it con notes the intended dimension more effectively and less ponder ously. 75 are managerial, autocratic, responsible, and overgenerous. Com mencing at the center of the circle and moving upward toward the zenith of dominance, we move from the sector in which behavior is "managerial" to the upper extremes of this theme where it is more "autocratic." This is noted in Figure 2, Appendix D. It is worth noting in this regard that the octant "AP— Managerial- Autocratic" is not confined to the north-northeast half of the upper right quadrant known as "affectionate dominance." The closer it approaches the outer Cupper) limits of the circle of behavioral themes, the more it encroaches on the left hemisphere of hostile dominance. This is a schematic method of describing the progressively less affectionate and more hostile the role descriptions become as they move from the inner core of the circle to the outer perimeter. Leary conceptualizes this as an adaptive-maladaptive dimensions of the total scheme. That is, behaviors nearer the center, regardless of the octant in which they lie, are more adaptive than those nearer the perimeter of the circle in that same octant. Consulting Figure 3 in. Appendix D, we see that for example, the eight role descriptions in the left half of octant AP, directly above the letter "A" progress upward as fallows: Intensity Location Description 1 inner core (generic theme) able to give orders 2 second ring (adaptive) likes responsibility good leader forceful Intensity 3 Location Description third ring (less adaptive- aa ladaptive) manages others dominating bossy outer ring (maladaptive) dictatorial The location of a particular descriptive adjective or phrase at a given ring of intensity was determined by the pro portion of subjects checking that item in the standardization and validation studies done on the ICL during its composition, as well as subsequently. According to these findings, ninety per cent of the respondent population checked intensity level one items, sixty-seven percent checked intensity level two items, and thirty-three percent checked intensity level three items, 26 while only ten percent checked intensity level four items. In the present study, as well as in the work of Miller which it replicates, the purpose of the ICL is to identify the subject's scores on the two dimensions of dominance - submission and affiliation - hostility. These scores are then compared to the partner's scores and a "total difference of needs" score is computed in the following series of steps: 1. The test is scored by counting the number of items checked for each of the octants, AP, BC, DE, FG, HI, JK, LM, and NO. 2. These raw scores are converted to standard scores and a dominance score (DOM) and an affiliation score (LOV) derived according to the following formulae: 26 Timothy Leary and H. S. Coffey, "Interpersonal Diagnosis Some Problems of Methodology and Validation," Journal of Abnormal Social Psychology. L (January, 1955)1 110-12^. 77 DOM = AP - HI + .7 (NO + BC - FG - JK) LOV = LM - DE + .7 (NO - BC - FG + JK) 3. The difference (disregarding arithmetic signs) between the partners' DOM scores is added to the difference (d.a.s.) between their LOV scores in order to obtain a "total difference of needs" score for the couple. The third step provides a dyadic score which is the operationalized measure of complementarity or homogamy of needs in our research design. While this latter step in scoring is not a part of. the original ICL format as developed by Leary, it has been used in subsequent studies, including those of King and Kotlar.27 Validity. The validity of the ICL has been established by comparing diagnoses of subjects on the ICL and on the 28 Minnesota Multiphasic Personality Inventory. These validation checks included several hundred subjects, patients as well as "normalB." 2^ The reliability of the ICL has been established under test-retest conditions in which correlations ranging from .73 to Elsie V. King, "Personality Characteristics— Ideal and Perceived in Relation to Mate Selection," (unpublished Ph.D. dissertation, University of Southern California, 1961). Sally Lee Kotlar, "Middle-Class Marital Roles— Ideal and Perceived in Relation to Adjustment in Marriage," (unpublished Ph.D. disserta tion, University of Southern California, 1961). 28 Leary, Interpersonal Theory, p. 287. 29 Leary and Coffey, "Interpersonal Diagnosis," p. 122, 78 .78 have been found.^ A study of the reliability of the ICL by Armstrong indicates that it achieves high internal consistency, 31 regardless of population and rating category. Procedure. Each partner in the sample of marital pairs completed the one hundred twenty-eight item check list, answering i as it applied to himself. The instructions read: Please check those items which describe YOURSELF. Write on each line: + if the statement describes you 0 if the statement does not describe you The Marital Role Typology Scale Purpose. In order to differentiate between traditional and equalitarian couples an inventory of attitude toward authority decision-making, and personal development in marriage was com- 32 piled. This compilation was done by Miller. A series of attitudinal statements regarding marriage was obtained from several sources. These were mostly items from existing inven tories or from reports of previous research done on marital role attitudes, conflict, and idealogy, each of which seemed to have ^R* LaForge, and R. F, Suczek, "The Interpersonal Dim ension of Personality: III. An Interpersonal Check List," Journal of Personality. XIV (1955) t 9^-112. ^R. G. Armstrong, "The Leary Interpersonal Check List: A Reliability Study," Journal of Clinical Psychology. XIV (1958), 393-39^. 32Miller, "Familial Role," p. 99. 79 some bearing on the problem of differentiating marital role attitudes on a traditional-equalitarian basis. Validity. Of the thirty-five items compiled by Miller, which he then submitted to an expert panel, twenty-three items received their unanimous approval as discriminating between tra ditional and equalitarian attitudes. Those scale items which re ceived this judgment were obtained by Miller from three sources. 33 The first eighteen statements are from Jacobson. The nineteenth 3lf and twentieth statements are from Levinson and Huffman. And the last three items, twenty-one, twenty-two, and twenty-three, he obtained from Motz. These statements and their identifying typology (T = traditional, E - equalitarian) are as follows: E 1. The husband should help with the housework. E 2. The wife should take a job if she wants to. T 3« If the husband insists, the wife should quit a needed job. E km If a husband runs around, so can his wife. T 5. The husband should decide how to spend any extra money. A* H. Jacobson, "Conflict in Attitudes Toward the Mari tal Roles of Husband and Wife," Research Studies of the State College of Washington. XIX (June, 1951^1 103-106. J. Levinson and Phyllis E. Huffman, "Traditional Family Ideology and Its Relation to Personality," Journal of Personality. XXIII (March, 1955), 251-273- -^Annabelle B. Motz, "The Role Conception Inventory: A Tool for Research in Social Psychology," American Sociological Review. XVII (August, 1952), ^65-^71. 8o T 6. Husbands should be more strict with their wives. T 7. A married woman should not work outside of the home. T 8. What a husband does in his spare time is his own business. T 9* The husband should decide where to live. T 10. Woman's place is in the home. T 11. The wife should fit her life to her husband's. T 12. The husband's wishes should come first in most things. T 13. Marriage is the best career for a woman. T 14. The husband should wear the pants. T 13« Marriage should be a full time job for the wife. E 16. It is okay for the wife to earn as much as her husband. T 17. A wife should let her husband decide most things. T 18. Almost all money matters should be decided by the husband. T 19. Women who want to remove the word "obey" from the marriage service do not understand what it means to be a wife. E 20. Women have as much right as men to sow wild oats. T 21. The wife should help support the family only 36 when it is absolutely necessary. E 22. Marriage iB a partnership in which the wife should share the responsibility of supporting the family with the husband whenever possible. E 23- A married woman should work if she is able to and enjoys work. Thus there are sixteen statements of a traditional type and seven statements of an equalitarion type. Internal Consistency. 37 Following the method, outlined in Selltiz, was used to determine whether each item was consistent with the responses of those subjects with the highest and lowest total score on the entire set of items. First, the total (summated) score was calculated for each subject, on the Marital Role Typology Scale. Second, two groups were separated. One group was composed of the top twenty-five percent scorers and the other of the bottom twenty-five percent scorers. Next, the means and standard deviations for each item for both groups were calculated. Lastly, difference of means tests were run on each and every item comparing the means from the two groups. This represented Through a typographical error in the original study, the word "only" was omitted in the actual questionnaire, thus invalidating this item. 37 Selltiz, Research Methods, p. l85« 82 the "discriminating power" of each item* While not all the items produced statistically significant results, each did differentiate strongly in the expected direction, with one exception* Item 21 on the Marital Role Typology Scale reads as follows in the text of Miller’s dissertation, "The wife should 38 help support the family only when it is absolutely necessary* This is faithful to the source, Motz, from which the item was 39 taken. However, when Miller transcribed the twenty-three items in the Marital Role Typology onto the form he administered, i*0 he somehow dropped the word "only" out of the sentence. So that the form as it was actually tested was incorrectly worded* Unfortunately, in the effort to make the replication as absolutely comparable as possible, the test form from the Appendix of Miller's dissertation was photocopied so that exactly the same format was reproduced. Hence the crucial limiting adverb "only" was omitted* This error rendered the item non-discriminative aB all respondents marked either "agree" or "strongly agree," Therefore the item did not differentiate effectively the tradi tional from the equalitarian subjects and was excluded from all subsequent analysis including the Accuracy of Perception calcula tion. ^Miller, "Familial Role," p. 101. The italics are mine. ^Motz, "The Role Conception Inventory," p. 469. Miller, "Familial Role," p. 170. 8? Procedure For the completion of the Marital Role Typology Scale, the respondent was directed to, "CHECK THE COLUMN WHICH MOST CLOSELY EXPRESSES YOUR FEELINGS ABOUT EACH STATEMENT." ^ There were four columns provided categorized as follows: "strongly agree," "agree," "disagree," and "strongly disagree." Scoring Following Miller's procedure, low weights were assigned to those responses defined as expressing an equalitarian attitude, while high were assigned the traditional attitudes. Since each of the statements on the form was defined as either traditional or equalitarian, a response "strongly disagreeing" with a traditional statement would be scored as one point, as would a response "strongly agreeing" with an equalitarian state ment. So also, a response "strongly agreeing" with a traditional statement would be scored as five points, as would a response "strongly disagreeing" with an equalitarian statement. With a total of twenty-two items (item twenty-one omitted) scored on a scale of five points, the theoretical range of scores would be from twenty-two for the most extremely equalitarian score, to one hundred ten for the most extremely traditional score. For purposes of dichotomizing the sample into an equali tarian and traditional segment, a respondent is seen as being "equalitarian" if his score is below the median and he is seen lfl See Appendix A, p. 130. as "traditional" if his score is above the median. Accuracy of Perception Scale Procedure. For the completion of the Accuracy of Per ception Scale, the respondent was directed to a second copy of the Marital Role Typology form, with the following change of orientation. He was instructed to mark it this time, not accord ing to his own attitudes, but rather to reply as he thought his mate would if she were taking the test. The directions were as follows: "PRETEND THAT YOU ARE YOUR PRESENT MATE, AND CHECK THE COLUMN WHICH WOULD MOST CLEARLY EXPRESS YOUR MATE'S FEELINGS FOR EACH STATEMENT. REMEMBER, ATTEMPT TO GIVE YOUR MATE'S FEELINGS, NOT YOUR OWN.**2 Scoring. Following the procedure used by Miller, the score an individual received was derived from the number of correct predictions he made in proportion to the total number of usable responses. For this instrument, both responses of agreement were combined as a single prediction, as were both the disagreement responses. So that if a person predicted his mate's response as either "agree" or "strongly agree," when in fact the mate had responded with either, he was credited with an accurate "perception." However, if the mate in this case had actually marked either "disagree" or "strongly disagree," then this person would be rated as an inaccurate perceiver on this item. See Appendix A, p. 131. If a person correctly predicted fifteen of his mate's responses, and incorrectly predicted seven of them his accuracy score would be 15/22, or seventy percent. Any statements which were left unmarked by either partner on his own Marital Role Typology form were not included in the computation of the accuracy of perception test. Statistical Procedures The design of the study, the type of data collected and the nature of the hypotheses to be tested, logically suggest use of an analysis of variance technique as the method of choice for i f 3 the statistical analysis of the study. In contrast to other parametric techniques for making statistical inferences, such as the t - test or a - test, the advantages of an analysis of 44- variance procedure are considerable. The most important reason for chosing an analysis of variance design is that when multiple tests of significance are employed there is a probability that the t - test or z - test might by chance reach a level of probability which would lead to rejection of the null hypothesis when in fact it should be accepted. Thus committing a Type I error. For example, if 4 - 5 This is also the statistical design selected by Miller. 44- William L, Hays, Statistics for Psychologists (New York: Holt, Rinehart and Winston, 19^3), pp» 356-357« 86 five tests are carried out on the same data then 5 * (.05) = -25 of these tests will show statistical significance at the *05 level by chance alone* So at least one (actually 1)0 of these five tests would be expected to show significance by chance ilK alone* Thus a single test is to be preferred. The analysis of variance technique also tests whether the entire distribution of sampling statistics could have occurred by chance alone* The actual calculations were done by computer UBing the special extended precision of modern computers (over ten significant digits). Because of the unequal distribution of data in the cells of the three by three tables* a special computer program had to be written to do the analysis of 46 variance* The special program allowed use of all the data available and did not require the discarding of cases from those cells which had too many couples as would have been the case with conventional computer programs* A check was made to see if the data met the assumptions underlying the analysis of variance technique: normal bivariate populations* equality of variance for each cell and independence of errors across all pairs of observations. An examination of the frequency distributions of the data revealed that the ^Hays* Statistics* pp. 410* 46 Written by Mr. Reuben Snipper and processed at the Computer Center* University of Southern California on an IBM 360 model 65# assumption of normality was met. Tests for the homogeneity of variance were run and were within the acceptable limits of the rather "robust" latitude of the analysis of variance. Finally, the distribution of cell sizes was tested by an appropriate chi- square test and found to be within the ninety-five percent limits. Thus the various cell sizes were the product of chance alone and not some systematic influence. The two-way analysis of variance allows testing of three main types of hypotheses: unequal row means Crow effects), unequal column means (column effects) and the interactional effects of specified cell meanB. Using the theoretical frame work no hypothesis is possible for row effects; Hypothesis I predicts the column effects and Hypothesis II, the interactional effects. Figure 1 will help clarify the various hypotheses. Hypothesis I is that X.^ > X.^ > X.^ Hypothesis II has three parts 1. H*1 H V X12 » *13 2. H V *21 » *31 3* xi3< X23 « *33 A careful examination of the hypotheses and the figure suggest the reason no row effects can be predicted: the first row is hypothesized to contain both the highest score (in cell 11) and the lowest score (in cell 1?) of all the cells. When the number of cases in each cell is unequal, two— ways analysis of variance is not straightforward. Blalock 88 FIGURE 1 LAYOUT OF A TWO-WAY ANALYSIS OF VARIANCE DESIGN Both Accurate Mixed Both Inaccurate Both Traditional Mixed Both Equalitarian 11 12 13 21 22 23 31 32 33 V Note: In the simplified notation used here corresponds to the mean and the subscripts to the row and oolumn in that order. The dot -2 indicates that the summation was over all the rows down column 2. This figure corresponds to Miller's FIGURE 2 in Appendix B* 89 47 suggests two procedures. One uses logarithms but is really an approximation method. The other, described in detail in 48 Walker and lev, is conceptually very simple. The means of the individual cells are treated as though they were single cases and the sums of squares and estimates of variance are obtained for the between-rows, between-columns and interaction terms by assuming that the mean is the only case in each cell. The error sum of squares is obtained as in the ordinary two-way analysis of variance from the difference between the total sum of squares and the between-cell sum of squares using the total number of cases rather than the mean of each cell. The error estimate is found by dividing by the degrees of freedom, as usual, but then this estimate is divided by the harmonic mean of the number of cases in each cell. This last operation is necessary to make the error estimate, based on the total number of cases, compar able to the estimates based only on the cell means treated as single cases. F tests are made in the usual way. Hypotheses V and VI are tested by calculation of the Pearson product-moment correlation coefficient. The assumptions underlying this parametric test were also met. The critical region for rejection of the null hypothesis was set at the .05 level for all of the hypotheses tested. ^Hubert M. Blalock, Jr., Social Statistics (New York: McGraw-Hill Book Co., i960), p. 264. 48 Helen M. Walker and J. Lev., Statistical Inference (New York: Holt and Co., Inc., 1953)* PP» 381-382. 90 Limitations of the Study The first factor to be considered as a limitation of the study is the representativeness of the sample* Just as Miller*s study was limited by his sampling procedure, so our sampling k9 method undoubtedly introduced some kind of bias. Whereas his was a known loss of approximately ten percent according to his estimate, since the present study was based cm a sample partially gathered by others, there is no clearly satisfactory method to estimate the amount of bias introduced by the unknown refusal rate. As was true of Miller's research, as well as others in survey research, whenever the target population refuses to participate in the survey, it introduces the specter of Inrepre sentativeness of the results. Did those couples who refused to participate in this research have traditional, equalitarian, or mixed marital role typologies? We do not know. Moreover, those couples who did agree to participate in the study by returning a completed questionnaire did so at least in part out of loyalty to the class instructor, or else to a friend who asked them to assist him in this for a class he was taking. Whatever the precise composition of the motives, this selection procedure introduced a bias in the direction of over- ^Miller, "Familial Role Typology," p. 109 91 representing public school teachers (the occupation of a majority of the class members in the extension classes where the forms were distributed) or else friends of these teachers. Another consideration which tends to limit the power of the study for generalization is the validity of the subject's responses to the survey instruments. This issue has two parts, the first of which is the question of the psychic state of the couples at the time they completed the test items. One of the possible explanations offered by Miller for his lack of significant findings was the influence of euphoria on the pre- 50 nuptial couples. It may very well be that even after they are married, couples continue to retain some residual euphoria or lack of reality-testing in their assessment of themselves and of each other. If such is the case, it could logically be expected to bias their responses. The second part of this issue of the validity of subject responses is the possible violation of the instructions to the couples to work separately on the questionnaire. Since these test forms were not filled out in the presence of the researcher, it was not possible to control for this source of bias. A comparison of responses between partners may have occurred as an artifact of either subordinate or superordinate behavior - i.e., a subordinate spouse may have copied the mate's responses in order to "give" compliance, or else a superordinate spouse may ^**M±ller, "Familial Role Typology," p. 110, have directed the mate to reply in certain ways in order to "receive" compliance. These two possible cases of contamination are congruent to the traditional-authoritarian marital role typology. It is more difficult to conceptualize a source of such bias from the autonomous-democratic portion of the sample, since our theoretical formulation by definition predicts such couples to be more highly individuated, as well as autonomous, and therefore more likely to refuse to participate at all than to feel compelled to participate, yet feel constrained to violate the terms of the contract for participation presented in the test instructions. Since the Marital Role Typology test and the Accuracy of Perception test are especially susceptibel to comparison-of and cooperation-on answers, it is worthy of note that there is only slight difference between the Miller study results and the results of the present study on these two measures. The mean score for the total samples on the Marital Role Typology test is presented in Table 8. The mean percent of accuracy for the total samples on the Accuracy of Perception test is also presented in Table 8, Finally, there is a factor to be considered in regard to subject responses not related to euphoria, nor to intentional duplication of the spouse’s answers, but rather to the method ological issue of determining the source of accuracy of percep tion. That is, for those persons who were accurate in their predictions of their mate's marital role attitudes, no assurance 93 Summary The sample of sixty-three couples was obtained through the adult education classes taught by the author for the Extension Division of the University of California. In each of the couples one or both of the partners had been previously married. As this is a replication study, the criteria for inclusion in the sample were the same as those of the exploratory study done by Miller. There were three social characteristics: Caucasian race, of non-Jevish religion, with the husband rated as middle or upper middle class on Hollingsheadfs "Two Factor Index of Social Position." The measurement of the variables was operationalized through the use of the Marital Bole Typology Scale, the Accuracy of Perception Scale, and the Interpersonal Check List. A two-way analysis of variance design, as well as Pearson’s product-moment coefficient of correlation techniques were used to test the six hypotheses at the .05 level of significance. CHAPTER V THE FINDINGS OF THE STUDY This chapter is divided into four parts, with the principal portion being devoted to a report of the findings on the six main hypotheses. First, however, the distribution of scores on the role typology scale and the accuracy of perception scale will be reported. Then, fallowing the discussion of the hypotheses, will come a consideration of some emergent findings and a concluding summary of the chapter. Distribution of Scores on Two Variables Distribution of Scores on the Marital Role Typology Scale Since low scores indicated equalitarian role attitudes and high scores traditional role attitudes, the median for the total sample was set as the cutting point. The potential range of scores on the 22 items finally determined as constituting the Marital Role Typology Scale was 22 to 110 points. The actual range was 27 to 95, the entire distribution by sex, of the Marital Role Typology scores can be seen in Table 9* Because the median was 5k there were 7k traditional subjects and 55 equalitarian subjects. A difference of means test (t - test) revealed that there 9k 95 TABLE 9 MABITAL BOLE TYPOLOGY, BY SEX Males Percent Accurate No. Females Percent No. Percent Total Percent 1 1.6 1 .8 90 - 94 85 - 89 1 1.6 1 .8 80 - 8*+ 75 - 79 3 if.8 1 1,6 4 3.2 70 - ?b 5 7.9 5 4.0 65 - 69 3 if.8 8 12.7 11 8.6 60 - 6b lif 22,2 7 11.1 21 16.7 55 - 59 i f 6.i f 16 25-4 20 15.9 50 - 3b 20 31.8 7 11.1 27 21.3 45 - b9 9 14,2 9 Ilf.2 18 14.3 4o - bb 5 7.8 3 4.8 8 6.4 35 - 39 2 3.2 2 3.2 4 3.2 30 - 3^ 1 1.6 3 4.8 4 3.2 25 - 29 1 1.6 1 1.6 2 1.6 63 100.0 63 100.0 126 100.0 Median 5b 57.5 54.0 Mean 3b,b 56.2 55.3 SD 10.9 11.9 11,6 t - test; t = .87 , Not Significant, p < .05 This table corresponds to Miller’s TABLE 8 in Appendix B. 96 was no difference between the males and females on the Marital Role Typology scale (t = .87)* The couples were then divided into three groups: (l) both traditional, (2) both equalitarian, and (3) mixed (one traditional one equalitarian). There were 25 traditional couples, Ik equalitarian and Zk mixed. Distribution of Scores on the Accuracy of Perception Scale The Accuracy of Perception score is the percentage of accurately predicted items divided by the total number of usable items. The potential range of scores is 0 percent to 100 percent. The actual range was ^.5 percent to 100 percent. The actual dis tribution of scores can be seen in Table 10. The median was again used as the cutting point. There were 6k "accurate per- ceivers" and 62 "inaccurate perceivers." A difference of means test (t - test) revealed that there was no difference between the males and females on the Accuracy of Perception Scale (t = 1.9). The couples were then divided into three groups 21 couples (both accurate), 22 couples (mixed), 20 couples (both inaccu rate) • Hypothesis I The difference between the interpersonal needs of marital partners will vary as a function of their marital role attitudes. 97 TABLE 10 ACCURACY OF PERCEPTION, BY SEX Percent Accurate Males No. Percent Females No. Percent Total Percent 100.0 3 4.8 3 4.8 6 8 95.0 - 99.9 1 1.6 2 3.2 3 2.4 90.0 - 94.9 3 4.8 1 1.6 4 3.2 85.0 - 89.9 7 11.1 3 4.8 IO 7.9 80.0 - 84.9 7 11.1 7 11.1 14 11.1 75.0 - 79.9 4 6.3 8 12.7 12 9.5 70.0 - 74.9 12 19*0 6 9.^ 18 14.3 65.0 - 69.9 11 17.4 7 11.1 18 14.3 60.0 - 64.9 10 15.9 9 14.3 19 15.0 55.0 - 59.9 2 3.2 3 4.8 5 4.0 50.0 - 54.9 3 4.8 5 7.8 8 6.3 45.0 - 49.9 2 3.2 2 1.6 40.0 - 44.9 3 4.8 3 2.4 35.0 - 39.9 2 3.2 2 1.6 30.0 - 34.9 25.0 - 29.9 20.0 - 24.9 1 1.6 1 .8 15.0 - 19.9 10.0 - 14.9 5.0 - 9.9 0.0 - 4.9 1 1.6 1 .8 100.0 100.0 100.0 Median 12,1% 68.2# 12.1% Mean 74.2% 67.6# 70.9% SD N = t = 1.91 Not 11.8 63 significant p .05 18.4 63 15.9 126 98 It is predicted that (l) the difference between the inter personal needs of traditional partners will be the greatest, (2) the difference between the interpersonal needs of mixed partners (male traditional-female equalitarian and male equalitarxan-female traditional) will be less great and (3) the difference between the interpersonal needs of equalitarian partners will be least. The instrument used to measure interpersonal need, the Interpersonal Check List, consists of two need dimensions, domi- nance-submission (DOM) and affiliation-hostility (LOV). A third need dimension, called the total difference of needs, is also examined, which is a summation of the difference of the dyad's DOM and LOV need scores. Since each of these three needs iB investigated in relation to marital role attitudes and accuracy of perception, three separate analysis of variance tests are required. The DOM, LOV, and total difference of needs tests are presented in Tables 11, 12, and 1J5 respectively. Hypothesis I is tested by the column effects. Thus according to the hypo thesis, X. > X._. The F ratio was compared with the * L d . 5 F .05 value as shown in the respective tables. Since none of the column values falls in the critical region for rejection of the null hypothesis (and thuB acceptance of our hypothesis), Hypothesis I is rejected. This same conclusion was reached by Miller. ‘ '"Miller, "Familial Hole Typology," p. 118. TABLE 11 99 DOM NEED DIFFERENCE Both Both Traditional Mixed Equalitarian Both Accurate Cell 11 X = 12.284 Z X = 73.707 z x2= 982.308 N s 6 Cell 12 x = 24.905 2 X = 149.431 2 X2= 4,440.956 N = 6 Cell 13 x = 36.906 2 X = 332.151 EX2=i8,36l.38o N = 9 Mixed Cell 21 X = 22.791 2 x = 227.909 z x2= 6,399.980 N = 10 Cell 22 x = 33.016 2 X = 264.127 2 x2=13,109.775 N = 8 Cell 23 X = 17.755 EX = 71.020 z x2= 2,338.130 N = 4 Cell 31 Cell 32 Cell 33 Both X = 27.419 X = 41.486 X = 13.951 Inaccu 2 X = 246.771 2 X = 414.856 EX = 13.951 rate Z X =11,674.664 I X =24,362.604 £ X = 194.620 N = 9 N = 10 N = 1 Summary of Analysis of Variance Source Sum of Squares DF Mean Square F F.05 Column 260.920 2 130.462 1.161 3.18 Row 18.154 2 9.077 0.080 3.18 Interaction 542.328 4 135.582 1.206 2.55 Error 54 112.368 100 TABLE 12 LOV NEED DIFFERENCE Both Traditional Mixed Both Equalitarian Cell 11 Cell 12 Cell 13 Both. X = 17.857 x = 31.032 X = 25*448 Accurate sx = 107.143 2 X = 186.194 2 X = 229.028 p 2 X = 4,065-776 2 X =10,734.600 2 X = 9,886.020 N = 6 N = 6 N = 9 Cell 21 Cell 22 Cell 23 x = 24.876 x = 30.063 X = 46.457 Mixed 2 X = 248.758 I X = 240.505 2 X = 185.826 2 X2=10,606.995 2 X2=12,302 » 352 2 x2=n,3^7.074 N = 10 N = 8 N = 4 Cell 31 Cell 32 Cell 33 Both x = 30.037 X = 38.485 X = 44.784 Inaccu 2 X = 270.336 2 X = 384.852 2 X = 44.784 p rate 2 X =251909.005 2 X =23,300.366 2 X = 2,005.565 N = 9 N = 10 N = 1 Summary of Analysis of Variance Source Sum of Squares DF Mean Square F F.05 Column 326.634 2 163.342 1.058 3.18 Row 265.844 2 132.922 0.861 3.18 Interaction 124.067 4 31.016 0.201 2.53 Error - 54 154.300 101 TABLE 13 TOTAL NEED DIFFERENCE Both Traditional Mixed Both Equalitarian Cell 11 Cell 12 Cell 13 Both X = 30.142 X = 55.938 X = 62.353 Accurate EX = 180.850 Z X = 335.625 z x = 561.179 2 X2= 71530.807 z X2=27,648.876 Z X2=46,625.729 N = 6 N = 6 N = 9 Cell 21 Cell 22 Cell 23 x = 47.667 x = 63,079 X = 64.212 Mixed £ x = 476.668 2 X = 504.632 z x = 256.846 Z X2=25,959.215 Z X2^44t547.902 z X2=22,433.491 N = 10 N = 8 N = 4 Cell 31 Cell 32 Cell 33 Both X = 57.456 X = 79.971 X = . 58.734 Inaccu z x = 517.108 £ X = 799.707 z x = 58.734 rate Z X2=42,353*562 z x =78,054.788 z x = 3,449.702 N = 9 N = 10 N = 1 Summary of Analysis of Variance Source Sum of Squares DF Mean Square F .05 Column 750.147 2 375.073 1.201 3.18 Row 381.246 2 190.623 0,610 3.18 Interaction 321.932 4 80.483 0. 257 2,55 Error — 54 312.254 102 Hypothesis II The difference between the interpersonal needs of marital partners will vary as a function of the interactional effects of their marital role attitudes and their accuracy of perception of the mate’s marital role attitude. Hypothesis II predicts that there will he interaction effects "which cannot he explained merely by adding the row and 2 column effects." This means that the accuracy of perception of the couple affects the relationship between the couple's marital role attitudes and their difference of needs. The underlying assumption is that the more accurately one partner perceives the other, the more able he is to choose a mate who gratifies his needs. Testing is against the null hypothesis of no interaction effect and as before is done on each of the need dimensions. On each of these needs three predictions are made with respect to the interaction effects of marital roles and accuracy of perception: 1. Among those dyads in which both members are able to perceive the other's marital role attitudes accurately, the difference of partners' needs is greatest between traditional, less between mixed (one member tradi- 2 Helen M. Walker and Joseph lev, Statistical Inference (New York: Rinehart and Winston, Inc., 1953)i P* 351* tional, the other member equalitarian), and least between equalitarian couples. 2. Among those traditional dyads, the difference of partners' needs is greatest between those couples who are mutually able to perceive the other's role attitudes accurately, less between those couples in which only one member is able to perceive his mate' role attitudes accurately, and least between those couples who are mutually unable to perceive the other's role attitudes accurately. 3* Among those equalitarian dyads, the difference of partners' needs is least between those couples who are mutually able to perceive the other's role attitudes accurately, greater between those couples in which only one member is able to perceive the other's role attitudes accurately, and greatest between those couples who are mutually unable to perceive the other's role attitudes accurately. Hypothesis II may be stated symbolically. These predic tions with reference to the DOM, LOV, and total difference of needs dimensions in Tables 11, 12, and 13 (pages respectively, are as followsJ 10*f To test the hypothesis, the F ratio is computed on the interaction of each need measure. The calculated F ratio for the three need dimensions for the interaction effects can be seen in Tables IX, 12, and 13 and are seen to be.non-significant. Thus the null hypothesis is accepted and Hypothesis II is 3 rejected. Miller reported the same findings. Hypotheses III, and IV Hypothesis III— there will be a negative correlation be tween the interpersonal needs of traditional marital partners. Hypothesis IV— there will be a positive correlation be tween the interpersonal needs of equalitarian marital partners. The correlations needed to test these two hypotheses against the null hypothesis of no correlation are presented in Table l*f. The correlations for the total sample are included because Miller included them. The total sample was composed of ^0 percent traditional couples, 22 percent equalitarian couples, and 38 percent mixed couples. T - tests on each of the six correlations revealed that none is statistically significant at the .05 level. Thus Hypotheses III and IV are rejected. Since none of the correla tions is statistically significant, nothing further can be said ^Miller, "Familial Role Typology," p. 12*f. **Ibid.. p. 126. 5ln Miller's study the composition was 35 percent tradi tional, 25 percent equalitarian, and 40 percent mixed couples. 105 TABLE lit CORRELATIONS BETWEEN THE MALES' AND FEMALES' SCORES AMONG SUBSAMPLES AND TOTAL SAMPLE Group N Correlation on DOM Need Correlation on LOV Need Mutually Traditional 25 -.058 .250 Mutually Equalitarian lit -ASk .171 Total Sample 63 -.109 .091 None significant at p < .05 This table corresponds to Miller's TABLE 15 in Appendix B. about them.^ It should be noted, however, that no hypothesis is possible on the total difference of needs dimension since each partner of a couple receives the same score on this dimension. Thus no correlation other than a perfect one is possible. 7 Miller reached the same conclusion. Hypotheses V and VI Hypothesis V— there will be a positive correlation be tween the difference of interpersonal needs and accuracy of perception of mate's marital role attitude for marital partners of the traditional type. Hypothesis VI— -there will be a negative correlation be tween the difference of interpersonal needs and accuracy of perception of mate's marital role attitude for marital partners of the equalitarian type. In Hypothesis V it is predicted that among mutually traditional couples, the greater a person's accuracy of percep tion score, then the greater the difference between (l) his and his mate's DOM needs, (2) his and his mate's LOV needs, and g It will be seen from Miller's Table 13 in Appendix B, p. li*5, that his sample achieved significant positive correla tions on DOM needs. This forced a rejection of his hypothesis Ilia (III of the present study), that the correlation for tra ditional couples would be negative. However, it also contradicts the theory of need complementarity. Such is not the case with the correlations in the present study. They are all negative, but not significant. ^Miller, "familial Hole Typology," p. 12^. 10? (3) his and his mate's total difference of needs. In Hypothesis VI it is predicted that among mutually equalitarian couples, the greater a person's accuracy of per ception score, then the smaller the difference between (1) his and his mate's DOM needs, (2) his and his mate's LOV needs, and (3) his and his mate's total difference of needs. The correlations needed to test these hypotheses are presented in Table 15. T - tests on each of the correlations indicated that none is statistically significant at the .05 level, with one exception. The correlation between partners' difference of DOM need scores and the females' accuracy of per ception score for all the females in the sample (N = 63) is r = .28*f, which is significant at p<.02 Ct = 2.^1). This high correlation appears anomalous, but will be considered in the next section of this chapter under the heading of emergent findings. Since the other correlations are not significant, no further inferences can be drawn and the hypotheses must be rejected. g Miller reported the same result. Emergent Findings In this section, those findings which emerged from the study as statistically significant, but about which no hypotheses Q Ibid., p. 132* 108 TABLE 15 CORRELATION BETWEEN THE PARTNERS' DIFFERENCES OF INTERPERSONAL NEED SCORES AND THE ACCURACY OF PERCEPTION SCORES BY SEX, OF SUBSAMPLES AND TOTAL SAMPLE DOM DIFF LOV DIFF Total DIFF vs vs vs Accuracy Accuracy Accuracy of of of Group N Perception Perception Perception Mutually Traditional Males 25 -.228 -.130 -.224 Females 25 .013 -.262 -.176 Mixed Males 24 -.062 .017 -.025 Females 24 -.509 .112 -.222 Mutually Equalitarian Males 14 - M b .169 .169 Females 14 0 00 K ' . • .018 .276 Total Sample Males 63 -.01? 1 * 0 H 00 -.002 Females 63 .284* .006 -.042 * Significant at p *02 This table corresponds to Miller's TABLE 14 in Appendix B. whatsoever were established before testing, will be discussed. There are three such findings to be considered here, two of which emerged from Miller’s work, and one of which emerged in the course of the present study. Since the second emergent finding Miller discussed was not adequately supported in the present study, it is considered first. Miller reported finding a statistically significant cor relation for the entire sample between mates on the marital role attitudes scale. He found that for his sample of two hundred 9 ten pre-nuptial couples, a positive correlation existed. Such a finding indicated the best predictor of a person's score on the Marital Role Typology scale to be his mate's score on that instrument. While the present study failed to achieve correlation of significance at the .05 level Ct = 1.8, p<.l), it was in the positive direction. At most, the present finding could be said to provide weak support, or at least not to contradict the Miller results on this issue. The other emergent finding identified by Miller was the relationship between accuracy of perception of couples, especially for traditional couples and the total sample, and by contrast, the equalitarian couples. He found a significant positive relation ship between partners on the Accuracy of Perception Scale for mutually traditional couples, as well as for the total sample (traditional, mixed, and equalitarian), but not for the mutually 110 equalitarian couples. These results were confirmed by the present study and are arrayed for comparison in Table 16. These results would seem to indicate that for traditional couples and for the total sample of couples in each, of these two studies, the more accurate one member of the pair is in predict ing his mate's marital role attitudes, the more accurate will be his partner’s perceptions. A corollary of this emergent finding for both studies, that the total sample of couples (traditional, mixed, and equalitarian) is significantly more perceptive than the most perceptive subsample, the mutually traditional couples, is a function of the statistical procedures which enhances the discriminative power of a test as the sample size is increased* So that by merely adding the three subsamples - one of which already has achieved statistical significance, and two of which are nearly so - the measurement base is improved by more than a merely additive factor. In this case, the whole is greater than the sum of the parts. The possibility that the higher correla tion of the total sample than of the traditionals could be explained by a high correlation of the mixed subsample - a truly confounding anomaly - was checked very carefully. No such anomalous result was found. The correlation was in fact the weakest of the three subsamples: n = 2*f, r = .3*f, t = 1*7, and as for equalitarian couples, not statistically significant. The third and last of the emergent findings to be dis cussed in this chapter is one not found by Miller in his sample, but one which developed in the present study. It was found as TABLE 16 CORRELATION BETWEEN THE MALE'S AND FEMALE'S ACCURACY OF PERCEPTION SCORES AMONG SUBSAMPLES AND THE TOTAL SAMPLE Group N Correlation on Present Study Accuracy of N Exception Millera Mutually Traditional 25 .^09** 70 .^5* * Mutually Equalitarian 1*+ .509 57 .19 Total Sample 63 ,if06*** 210 ** Significant at p < .05 *** Significant at p < .002 112 an artifact of the operations performed to test hypotheses V and VI concerning the relationship of accuracy of perception scores to both traditional and equalitarian couples' need scores.^ None of the statistical tests of association proved significant except that of the "difference of DOM needs" and "accuracy of perception" for females in the total sample* . This statistic indicates at least for the population from which the sample is drawn, that the accuracy of a female’s perception of her husband’s marital role attitudes varies directly with difference between his dominance (DOM) needs and hers* Therefore, if the difference between a couple’s DOM needs is great, the accuracy of her per ception of his role attitudes will be great. Whereas, if the difference between them on DOM needs is slight, then the accuracy of her perception of his role attitudes will be slight. This finding is congruent to the emergent finding discussed earlier that equalitarian couples are less accurate in their perception of spouse’s role attitudes. This finding specifies a portion of that relative inaccuracy, i.e. - in the present study females in equalitarian couples are significantly less accurate in perceiving their husband’s role attitudes than are females in traditional couples. No such finding was made for males, the statistic being non-significant in the present study. There are some implications from such findings which are worth noting. In the first place, it seems likely that accuracy ^Hypotheses V and VI are discussed on p. 107. of perception of mate's role attitudes is, as Miller suggested, less critical an issue for equalitarian couples than for tradi tional couples. At least, they exhibit lower accuracy scores. Whatever the male share of this lack of accuracy, it is clear from the present finding that the female contributes significantly to this phenomenon. And this greater contribution is congruent to the main thesis of this research, as well as Miller's, namely, that there is a difference in marital typologies* Traditional couples function on the basis of role differentiation heterog— amously in the way described by Parsons as instrumental and expressive, respectively for husband and wife. Since the wife's principal function is socioemotional, that is - expreBeive- integrative - she is by such terms more accommodating to the husband's more rigid instrumental role needs. Therefore, in order to fulfill this accommodative function, she must rely heavily on the accuracy of her perception of his role attitudes. This seems to be empirical evidence for the popular opinion that tradition ally is held of women as more "intuitive" than men* No such sex-linked role specialization is required of the partners in an equalitarian marriage according to the theoretical statement underlying this study* For in such a relationship, based as it is on the five functional norms of interpersonal process, no such rigid role definitions exist. And communication of needs directly by the husband replaces "intuiting" or "psyching-out" the needs by the wife (and vice-versa). Accuracy of perception as intra-personal achievement on the part of one 13A partner, institutionalized as the wife's role in the traditional marriage, is replaced in the equalitarian marriage by an inter personal process, communication* To summarize the finding, a high positive correlation was found for females in the total sample between dominance needs differences of partners and accuracy of perception of the partner's marital role attitudes. .Summary An array of frequency distributions on two scales, Marital Hole Typology and Accuracy of Perception, was presented, followed by the main body of the chapter in which the six research hypo theses and their related findings were discussed. Emergent findings, for which hypotheses had not been formulated a priori, were then presented. The findings were as follows: Hypothesis I— the difference between the interpersonal needs of marital partners did not vary as a function of their marital role attitudes. The criterion was measured by the column variance in a two-way analysis of variance design; the F tests were not statistically significant on DOM, LOV, nor total differ ence of needs. Therefore, this hypothesis was rejected. Hypothesis Hypothesis Hypothesis Hypothesis II— the difference between the interpersonal needs of marital partners did not vary as a function of the interactional effects of their marital role attitudes and their accuracy of perception. The criterion was measured by the interaction variance in the same two-way analysis of variance test used in Hypothesis I. The F tests were not statistically significant on DOM, LOV, nor total difference of needs. There fore, this hypothesis was rejected, III— A negative correlation between the inter personal needs of traditional partners was not found. On the contrary, a positive correlation was found. Therefore, this hypothesis was rejected. IV— A positive correlation between the inter personal needs of equalitarian type marital partners of sufficient statistical significant was found only for dominance needs, but not for love needs, nor for total needs. Therefore, this hypothesis was rejected. V— A positive correlation between the difference of interpersonal needs and accuracy of perception of mate's marital role attitude for marital partners of the traditional type, as hypothesized, was not found. Therefore this hypothesis was rejected. Hypothesis VI— A negative correlation between the differ ence of interpersonal needs and accuracy of perception of mate's marital role attitude for marital partners of the equalitarian type, as hypothesized, was not found. Therefore, this hypothesis was rejected. There were two related emergent findings, the first being that among traditionals, as well as the total sample of couples, accuracy of perception by one member of the partner's role attitude was correlated positively with accuracy of perception of that member's role attitude by the partner. The association for the total sample being an artifact of the statistical procedures, that of the traditional couples was attributed to the theorized function of accuracy of perception in that type of marriage. The other finding which emerged from the analysis of the data is related to the first, and that is that females throughout the sample are increasingly accurate in their perception of their partner's role attitude as the difference of dominance needs between spouses increases. Together, these two findings support the typology of marital roles predicted upon the traditional- equalitarian dichotomy. CHAPTER VI SUMMARY AND CONCLUSIONS A summary of this research study, together with the con clusions, as well as some implications for future research, are included in this final chapter. Summary The research sample was composed of sixty-three married couples, one or both of whom had been previously married, gathered through the adult education classes taught by the author for The Extension Division of the University of California. This was a middle and upper middle class sample, mostly college educated, non-Jewish Caucasians ranging in age from the early twenties through mid sixties. These and other social characteristics were obtained from a personal face sheet completed by each of the one hundred twenty-six respondents. Each participant also filled out three additional forms two of them identical in content, but referenced differently. These two forms were the Marital Role Typology scale for the self, and the Accuracy of Perception scale, a prediction of the mate's responses to the former scale. The third form was the Leary Interpersonal Check List used as a measure of interpersonal needs. The first two scales are 117 118 constructed of items from other instruments which were judged to discriminate responses on a typology of attitudes toward marital roles - traditional or equalitarian. Two statistical techniques were used for analysis of the data. These were the two-way analysis of variance, and the Pearsonian co-effecient of correlation. Each of the six research hypotheses was tested at the .05 level of significance. The theoretical framework within which the research was conducted was that of the sociological construct known as the "ideal type." The ideal typology of marriage proposed by Burgess and Locke was used. It was assumed that middle class marriages were of either the traditional institutional type, where role allocations were clearly differentiated along the lines of the husband-instrumental and wife-expressive as suggested by Parsons, or else of the equalitarian companionship type in which role allocations were convergent. A further category, "mixed," was added in which to place those couples who disclosed opposing role attitudes. Since different role expectations were attributed to the traditional and the equalitarian types of marriage, so differing interpersonal need sets were assumed to affect mate selection. Winch's theory of need complementarity in mate selection was considered the likely pattern for traditional couples expecting instrumental-expressive role differences. Whereas, need homogamy seemed the more likely pattern for equalitarian couples. Such a construction of the theoretical framework gave promise of resolving some of the contradictory 119 findings on the issue of need complementarity in mate selection. An extensive theoretical statement was offered, defining in detail the nature of the relationship within the framework of an ideal typology of marriage. Following this was a brief restatement of the theory in axiomatic form. From this was derived a set of logical propositions, which were then converted into the six research hypotheses. These hypotheses were as follows: I The difference between the interpersonal needs of marital partners will vary as a function of their marital role attitudes. IX The difference between the interpersonal needs of marital partners will vary as a function of the interactional effects of their marital role attitudes and their accuracy of perception of the mate's marital role attitude. Ill There will be a negative correlation between the interpersonal needs of traditional type marital partners. IV There will be a positive correlation between the interpersonal needs of equalitarian type marital partners* V There will be a positive correlation between the difference of interpersonal needs and accuracy of perception of mate's marital role attitude for marital partners of the traditional type. 120 VI There will be a negative correlation between the difference of interpersonal needs and accuracy of perception of mate's marital role attitude for marital partners of the equalitarian type. These hypotheses together with the theoretical framework from which they were derived, as well as the variables they were designed to measure through the administration of selected instruments and statistical procedures, were the same as those of the exploratory study done by Miller. Only the sample was changed in the hope of finding positive results. However, as was the case in Miller's research, none of the hypotheses was supported at a satisfactory level of statistical significance. However, certain findings did emerge about which no hypotheses had been generated. These were that a significant positive correlation existed between the traditional partners as well as for the sample of partners as a whole, for accuracy of perception. This did not hold true for equalitarian partners. Linked to this finding was another unhypothesized result indicat ing that for females in the total sample, accuracy of perception increased as difference in DOM needs between partners increased. These findings were interpreted to support the traditional- equalitarian typology of marital roles. This study had certain limitations, principally the field techniques of sample selection and questionnaire return. That is, the couples in the target population who failed to return their completed questionnaires were, in effect, refusing 121 to participate, They may have differed significantly from participant couples in other ways as well. Relying on the respondents to complete and mail their forms anonymously left no very effective follow-up procedure. In addition, there may have been confounding effects operating among the respondents, such as euphoric state of mind in relation to the marriage} deliberate intent to compare and cooperate on responses; and finally, the impossibility of detecting which "accurate" perceptions of the mate were true "perceptions" and not projections coinciding by chance with the mate's responses. Conclusions Even though none of the hypotheses was supported by the evidence presented, there are certain conclusions to be drawn from the study results. These will be principally related to the lack of significant findings. Since an adequate theory is a necessary base for sound research, it may be that the lack of significant findings in this study was due to the inadequacy of the theoretical framework. Though the conceptualization seemed promising, the framework as actually constructed was perhaps too simplistic. This is a criticism which has been levelled at the theory of need complementarity. And though the attempt was made to increase the complexity sufficiently to satisfy that criticism, while avoiding the equally serious risk of becoming too general, this would have fragmented the focus of the study, and thereby lost much of the power to generate relevant hypotheses. In short 122 it is concluded that the study may have lacked a theoretical framework with either the necessary complexity or the necessary relevance to produce confirmed hypotheses. Yet even if the theory were adequate, this study might have achieved the results it did due to deficiencies in the research, not the theory. In the first place, the research design may not have organized the variables in a sufficiently systematic way, nor generated a methodology to test those variables efficiently. The function of design is also to choose valid and reliable instruments to test the variables. Though the ICL remains an adequate choice for the interpersonal needs instrument, it would have enhanced the comparability of Miller's original design to have used either Schutz's FIRO-B test or the Edwards Personal Preference Schedule, or both. Especially since the study which used the former, reported support for need complementarity, while the published studies employing the EPPS have reported rejection of the hypothesis of complementary needs. To have used both of these tests in addition to the ICL would have greatly enhanced the importance of the present study. It also undoubtedly would have increased the rate of refusal among couples in the target population. On this last consideration, that of the sample, the conclusion seems justified that it was of an adequate size to allow the required tests of statistical significance to be per formed. The use of the computer rather than the hand scoring done by Hiller, facilitated the manipulations necessary for the analysis of the data. The particular statistical tests of significance used in this study, as well as by Miller, it is con cluded were justified by reference to the commonly accepted canons of statistical practice. There is a final conclusion, not related to the problem of the failure of the hypotheses or the methodology to generate significant support, but rather to the important additional findings which emerged from the study. This conclusion is that some underlying connection does exist between the type of marital role pattern and the phenomenon of accurate perception of mate's marital role attitude, especially for females. Just how the connection exists is not yet clear, but its existence seems established. In summary, it is necessary to conclude from the findings of the present study, as well as that of Miller and those cited in the review of the literature, that while identification of a couple's marital role attitudes may be a necessary component for the prediction of their interpersonal needs pattern, it is not sufficient. Implications for Future Research Based on the findings and conclusions arrived at in the present study, there are some important implications for further research. First, a more adequate theoretical base for research must be constructed for the analysis of factors related to mate selection and interpersonal needs. Such efforts as those of 12k Murstein and others to build a processual theory to replace the more static theories seems to hold promise for gratifying results. It must become more complex, achieve both increased generality and greater specificity. Second, in addition to more adequate theory, future research must continue to sharpen its tools and focus its attack more effectively on relevant variables. Third, there is an important opportunity for significant research in the study of marital role typology as related to accuracy of perception. This impinges upon many of the most critical current issues in addition to marital role typology. It is related to male-female relations from mate selection to marital adjustment, as well as sex-role learning and women's liberation. These areas of future research can be identified as relevant to this study of marital role typology, accuracy of perception, and interpersonal needs among a sample of married couples. In the course of analyzing the data on which this research is based, several phenomena were detected which, though not of sufficient statistical significance to enter in the findings, are worthy of note because of their possible Implications for future research. These occurrances will be discussed in relation to their appearance on the various tables of the previous chapters, as well as in reference to the main hypotheses. By comparing Table 10, p. 97 Accuracy of Perception by sex, with Hiller’s Table 9 (Appendix B), p. l^ it will be seen 125 that while males and females in the pre-nuptial sample (Miller) had virtually equal meanB (67*7 and 67*^, respectively), the females in the present study had a slightly lower mean (67*6) than the males (7^*2). Does this imply that perhaps the female Iossb some perceptive accuracy after marriage, and that males, on the average, gain some? Since the difference between the means for males and females in the present study is not statistically significant, it may be due entirely to change. However, this may be a fruitful avenue of inquiry as a part of research into the process of mate selection and family formation. Table 11, p. 99 on the DOM Need Difference displays the phenomenon of directionality exactly opposite to that hypo thesized, but not so as to be statistically significant. It is interesting to note that in Miller^ study, though not statisti cally significant, this directionality was as predicted (Table 10, Appendix B, p. Table 12, p. 100 indicates that for LOV Need Difference, the cell means do not go in the predicted direction, but appear mixed, except for the mutually traditional subsample in which the cell means go exactly opposite to the theoretically pre dicted direction. However, this event is not statistically significant. In Table 1J, p. 101 which displays the total needs differ ence dimension, the cell means in the array (except cell 33) go in the direction opposite to that hypothesized. This suggests that the theoretical relationship may be different than stated 126 here. Of interest in Table 14, p. 105 is the negative correla tions on the DOM needs and the positive ones on the LOV needs. Though none is statistically significant, this suggests that "need" is a more salient characteristic than "typology." For regardless of typology, all couples in the study show slight negative correlations on DOM needs and slight positive correla tions on LOV needs. Inspection of the array of correlations in Table 15, p. shows that five-sixths of them are in the direction opposite to that predicted. This is true both for the mutually traditional couples and for the mutually equalitarian couples. As no hypothesis was formulated for the events within the mixed couples, those correlations remain anomalous* Finally, the issue of DOM needs differences is of interest in relation to which has the higher score when there is a differ ence, the male or the female. This issue was resolved by checking each couple. The results may be summarized with remarkable uniformity. Regardless of typology in seventy per cent of the couples, the male reported a higher DOM score (traditional, sixty-eight percent; equalitarian, seventy-one percent; and mixed, seventy-one percent); whereas in thirty percent of the couples the female had a higher DOM score (traditional, thirty-two percent; equalitarian, twenty-nine percent; and mixed, twenty-nine percent). None of the correla tions between DOM need differences and Accuracy of Perception 127 with respect to this male-higher or female-higher occurrence proved to be statistically significant. Though none of the phenomena reported in this section occurs at the level required for statistical significance! there are important implications here for future research* principally those mentioned first in this discussion: the need for more adequate theory and improved research design* as well as data collection. APPENDIX A The Questionnaire 128 UNIVERSITY OF SOUTHERN CALIFORNIA C enter for the Study of the F a m ily L ife Cycle IBM Code 1 . 2. 4. 5. 6. 7. 8. 9. 10. 11 . 12. 13. 14. 15. T his study has been d esign ed for the pu rp ose of learn in g m ore about the a t titudes and v a lu e s of people who are m a rried . A ll in fo rm a tio n given for this r e sea r c h , sp o n so red by the U n iv e r sity of Southern C a lifo rn ia , is c o m p le te ly con fi d en tia l. No n a m es or identifying sta tem en ts appear on th ese form s. T h e re fo r e , p le a se be h o n est and frank in your a n sw e r s so that th is can be a v alu ab le study based on r e lia b le data. Sex: M D ate of birth: ( m o .) (day) .(y-rj Occupati0n<i f J ”^ P . I°y.ed; ________________________________________________________ pu joo; (e . g, tm ea tcu tter , tea ch er, clothing sa lesm a n , e t c .) C ir cle the h ig h est year of sch o o l com pleted: 12 or l e s s ; i i £ * j i 4 b i b , i i o A re you gain fu lly em ployed at this tim e? Yes______ No Annual in c o m e -- y o u r 's o n ly , not your m a te's: $ (A lim on y or child support paym en ts are not in com e fo r purposes of th is study. ) Ethnic origin: O riental______ ; A nglo______ ; M exican^_____ ; Negro_______; Other__ R elig io u s p referen ce: Cath._ ; Jew ish ; P r o t. When did you m a r r y your p r e se n t sp ouse? (m o. ) W ere you p r e v io u sly m a rried ? Y es______ No (day) ; Other ;None (yr.) G ive dates of p rev io u s m a r r ia g e s: 1st one from (yr. )_______ nated b y ______________________________ . (d iv o rce, death, annulm ent) G ive dates of p rev io u s m a r r ia g e s : 2nd one fr o m (yr. )______ nated b y _________________________________ . (d iv o rce, death, annulm ent) How long w e r e you engaged to your p r e se n t sp ou se? Years_ t0 (yr-) to (yr. ) term i- te r m i- M onths How long did you know your p r e se n t sp ou se b efore you becam e en gaged? Y ears__________M onths________ _ W as your p r e se n t spouse acquainted w ith your fo r m e r spouse before you r fo rm er m a r r ia g e ended? No______: Y es______ ; C lo se frien d s_______ ; Casual fr ie n d s______ ; A cquain tances only______ B efo re your p r e se n t m a r r ia g e did you and your p r e s e n t spouse have a ll______ , m o s t______ , so m e______ , or none_______ of your fr ie n d s in com m on at that tim e? H ave you and your sp o u se, as a m a r r ie d couple, changed a n y of your fr ie n d sh ip pat tern s fr o m those e a c h of you had b efo re your p r e se n t m a r r ia g e ? A nsw er below : 16. 17. 18. Husband retain ed all_ frien d s. m o st som e none , of h is fo r m e r Of the frien d s he has retain ed , a ll m o st som e none of them have b eco m e frien d s of h is p r e se n t w ife. W ife retained a ll______ , m o st______ , so m e______ , none_______ of her f o r m e r frien d s. h h f» fr-ipn/l c shi> has retained, a l l_____ . most_____ . Some P le a s e do not w rite in this colum n e, g. , m e jm iuinwM ) ■■■>; — 3. C ir c le the h ig h est y e a r of sc h o o l com p leted : 12 or le s s ; ^ 2 ^ 3 ^ 4 ’ ^C'1°°^ 4. A re you g a in fu lly em p lo y ed at this tim e? Y es_______ No______ 5. Annual in c o m e - - y o u r 's o n ly , not your m a te 's : $____________________________________ __ (A lim o n y o r child su p p ort p a y m en ts are not in co m e for p u r p o se s of th is study. ) 6 .. E thnic origin : O rien tal ; A nglo______ ; M exican______ ; N e g ro _______; Other_____ 7. R e lig io u s p r e fe r e n c e : Cath. ; J ew ish______ ; P r o t ._______; Other_______;None____ 8. When did you m a r r y your p r e se n t sp o u se? (m o. )________ (day)________ (yr. _)_________ 9. W ere you p r e v io u sly m a r r ie d ? Y es_______ No______ 10. G ive__d a tes of p r e v io u s_m a r r ia g e s : 1st one fr o m (yr. )_______ to (y r . )________ t e r m i nated b y ■ (d iv o rc e , d eath , annulm ent) 11. G ive__d a tes of p r e v io u s_m a r r ia g e s : 2nd one fr o m (yr. )_______ to (yr. )________ t e r m i nated b y _____ . (d iv o rc e , death, annulm ent) 12. How long w e r e you en gaged to your p r e se n t sp o u se? Y e a r s _________M onths________ 13. How long did you know your p r e se n t sp o u se b efo re you b e c a m e engaged? Y ea rs__________M onths_________ 14. Was your p r e se n t sp o u se acquainted w ith your fo r m e r sp o u se b efo re your fo r m e r m a r r ia g e ended? No______ ; Y es______ ; C lo se frien d s_______ ; C a su a l fr ie n d s_______; A cq u a in ta n ces only______ 15. B efo re you r p r e se n t m a r r ia g e did you and your p r e se n t sp o u se have a ll______ , m o s t______ , so m e_______, or none_______ of your frien d s in c o m m o n at that tim e? H ave you and your sp o u se, as a m a r r ie d co u p le, changed any of you r fr ie n d sh ip p a t te r n s fr o m th ose ea ch of you had b e fo r e your p r e se n t m a r r ia g e ? A n sw er below : 16. H usband retain ed a ll , m o s t______ , so m e , none , of h is fo r m e r fr ie n d s . 17. Of the fr ie n d s he has reta in ed , a ll_______ , m o s t____ , so m e_______ , none_______ of th em have b e c o m e fr ie n d s of h is p r e s e n t w ife . 18. Wife reta in ed a ll , m o s t______ , so m e _______.non e______ o f her fo r m e r fr ie n d s. 19. Of the frien d s she has retained) all________ , m o s t____ , so m e , none_______ of th em have b e c o m e fr ie n d s of her p r e s e n t husband. 20. A re th ere any c h ild re n livin g in your h om e at p r esen t? No_______ Y es_____ If the a n sw er to q u estio n 20 is y e s , p le a s e a n sw er the follow ing: 21. N um b er of w ife 's c h ild ren by fo r m e r m a r r ia g e :_____ . 22. N u m b er of hu sb and's c h ild ren by fo r m e r m a r r ia g e: 23. N um b er of n atu ral ch ild ren of this m a r r ia g e :_______ > . 24. N u m b er of adopted c h ild ren of this m a r r ia g e :_______ . 130 CHECK THE COLUMN WHICH MOST- CLOSELY E X PR E SSE S YOUR FEELINC strongly a g r ee a g r ee d isa g r e e str o n g ly - d is a g r e e The husband shou] _ The w ife should ta If the husband insi If a husband runs The husband shoul Husbands should t : A m a r r ie d woman What a husband dc The husband shou! W om an's place is The w ife should fi The husband's wii i M a rria g e is the b The husband shou M a rria g e should 1 • It's okay for the v A w ife should let A lm o st a ll m oney W om en who want m ea n s to be a \ W om en have as rr The w ife should li M arriage is a pai . . . . fa m ily with the A m a r r ie d womai i your feelings about each statement. 1 he husband should h elp w ith the h o u se w o rk , he w ife should take a job if she wants to. ' the husband in s is t s , the w ife should quit a needed job. ‘ a husband runs around, s o can his w ife . he husband should d ecid e how to spend a n y extra m o n ey . \1sband3 should be m o r e s tr ic t with th eir w iv e s. . m a r r ied w om an should not w ork o u tsid e of the hom e. rhat a husband does in his sp are tim e is h is own b u sin e ss, he husband should d ecid e w here to liv e . O m an's p lace is in the hom e. he wife should fit her life to her h u sb a n d 's. he husband's w ish e s should com e f ir s t in m o s t th ings. r » la rria g e is the b est c a r e e r for a w om an . he husband should w ea r the pants. la rria g e should be a fu ll tim e job for the w ife. :'s okay fo r the w ife to ea rn as m u ch a s h e r husband. . w ife should let her husband decide m o s t th in gs. d m ost a ll m on ey m a tte r s should be d e cid ed by the h u sb a n d .' fom en who want to r em o v e the w ord "obey" from the m a r r ia g e s e r v ic e don't u n d e rsta n d what it m eans to be a w ife. !om en have as m u ch righ t as m en to sow w ild oats. “ he w ife should help su p p ort the fa m ily w h en it is a b so lu tely n e c e s s a r y . la rria g e is a p a rtn ersh ip in which the w ife should s h a r e the r esp o n sib ility of su p p o rtin g the fa m ily with the husband w h enever p o s s ib le . . m a rried w om an should w ork if she i s a b le to and en jo y s w ork. 131 P R E T E N D THAT YOU ARE YOUR PRESENT MATE, AND CHECK THE COLUMN V FOR EACH STA TEM EN T. REM EM BER, ATTEM PT TO GIVE YOUR M ATE'S FEl s tr o n g ly a g r e e a g r e e d isa g r ee strongly d isa g ree The husband should he The wife should take a If the husband in s is t s , If a husband runs arou The husband should de Husbands should be m A m a rried woman sh o What a husband d oes i The husband should de W om an's place is in tl The wife should fit he The husband's w ish e s M arriage is the b e st < The husband should w M arriage should be a It's okay for the w ife A v/ife should let her A lm o st a ll m oney mai W omen who want to r < m eans to be a w ife. W om en have as m u ch The wife should help M arriage is a partnei • fa m ily with the hus A m a rried wom an shi CHECK THE COLUMN WHICH WOULD MOST CLEARLY EXPRESS YOUR MATE'S FEELINGS j IVE YOUR MATE'S FEELINGS, NOT YOUR OWN. The husband should help with the housew ork. The wife should take a job if she wants to. If the husband in s is ts , the wife should quit a needed job. If a husband runs around, so can his w ife. The husband should decide how to spend any extra m oney. Husbands should be moire str ict with their w ives. A m arried woman should not work outside of the home. 1 What a husband does in his spare tim e is his own business. The husband should decide where to liv e. ; . - * i * * i Woman's place is in the hom e. t The wife should fit her life to her husband's. The husband's w ish es should com e fir s t in m o st things. M arriage is the best c a r e e r for a w om an. \ The husband should w ear the pants. - I M arriage should be a fu ll tim e job for the wife. i It's okay for the wife to earn as m uch as her husband. . A v/ife should let her husband decide m o st things. ! A lm ost all money m a tters should be decided by the husband. Women who want to rem ove the word "obey" from the m arria g e se r v ic e don't understand what it m eans to be a wife. Women have as m uch right as m en to sow wild oats. The wife should help support the fam ily when it is absolutely n e c essa ry . M arriage is a partnership in which the w ife should share the resp on sib ility of supporting the fam ily with the husband w henever p o ssib le. A m arried woman should work if she is able to and enjoys work. 132 W rite on e a c h lin e: Please < heck chose items which describe YOURSELF if the 0 if the s ta te m e n t d e s c r ib e s y o u s ta te m e n t d o es not d e s c r ib e y ou . PI Dc well thought of makes a good impression able to give orders forceful se1f-respec ting independent' able to takE care of self can be indifferent to others can be strict if necessary firm but just can be frank and honest __critical of others can complain if necessary often gloomy able to doubt others frequently disappointed able to criticize self apologetic can be obedient usually gives in grateful admires and imitates others appreciative very anxious to be approved of cooperative eager to get along with others friendly affectionate and understanding considerate encourages others helpful big-hearted and unselfish i 1 2____ _often admired respected by others _good leader _likes responsibility ^self-confident ^self-reliant and assertive _businesslike _likes to compete with others _hard-boiled when necessary _stern but fair ^irritable■ ^straightforward and direct ^resents being bossed skeptical _hard to impress _touchy and easily hurt _easily embarrassed _lacks self-confidence _easily led modest _often helped by others _very respectful of authority _accepts advice readily _trustlng and eager to please _alway5 pleasant and agreeable _wants everyone to like him _sociable and neighborly _ y a r -i _kind and reassuring _tender and soft-hearted _enjoys taking care of others _gives freely of self always giving acts important bossy dominating boastful proud and sell _thfnks only of shrewd and cal impatient vltl self-seeking outspoken often unfrlenc bitter complaining Jealous slow to forgii self-punishinj shy passive and ui meek dependent wants co be l i lets others n v easily fooled too easily in, will confide fond of everyi likes everyboi forgives anyt! oversympathet generous to a overprotectlv 7---- £ 132 t heck those Items which describe YOURSELF if the s ta te m e n t d e s c r ib e s you if the s ta te m e n t does not d e s c r ib e y o u . P l e a s e w ork rap id ly, m ark in g a ll sp a c es. D o not d is c u s s your a n sw e r s w ith your sp o u se. often admired respected by others good leader __likes responsibility self-confident ■ self-reliant and assertive business like likes to compete with others hard-boiled when necessary stern but fair irritable • straightforward and direct resents being bossed skeptical hard to impress touchy and easily hurt easily embarrassed lacks self-confidence easily led modest often helped by others very respectful of authority accepts advice readily trusting and eager to please always pleasant and agreeable wants everyone to like him sociable and neighborly warn kind and reassuring tender and soft-hearted enjoys taking care of others gives freely of self always giving advice acts important bossy dominating boastful proud and self-satisfied thinks only of himself shrewd and calculating impatient with other's mistakes self-seeking outspoken often unfriendly bitter complaining jealous slow to forgive a wrong self-punishing shy passive and unaggressive meek dependent wants to be led lets others make decisions easily fooled too easily influenced by friends will confide in anyone fond of everyone likes everybody forgives anyching oversympathetic generous to a fault overprotective of others cries to be too successful expects everyone to admire him , manages others dictatorial somewhat snobbish egotistical and conceited selfish cold and unfeeling sarcast ic cruel and unkind ^frequently angry hard-hearted resentful rebels against everything stubborn distrusts everybody timid always ashamed of self obeys too willingly spineless hardly ever talks back clinging vine likes to be taken care of will believe anyone wants everyone’s love agrees with everyone friendly all the time loves everyone too lenient with others tries to comfort everyone too willing to give to others spoils people with kindness 4_ 8. DOM. LOV. U N IV E R S IT Y O F S O U T H E R N C A L IF O R N IA D E P A R T M E N T O F SO C IO L O G Y C E N T E R FO R T H E ST U D Y O F T H E F A M IL Y L IF E C Y C L E We a r e m o s t a p p r e c ia t iv e o f y o u r h e lp - - T H A N K Y O U . Y ou h a v e tak en p a r t in a r e s e a r c h , the r e s u l t s o f w h ic h w i l l a s s i s t m a r r i a g e c o u n s e l l o r s in th e ir w o r k w ith f a m i l i e s a s w e l l a s h u sb a n d s and w i v e s . Y ou m a y w is h to know w h a t t h e s e r e s u l t s a r e . T h e r e f o r e , a s p e c ia l m e e t in g in y o u r a r e a i s b e in g a r r a n g e d fo r th is p u r p o s e . If y o u w is h to jo in u s , and w c c e r t a in ly h o p e that you w ill, p l e a s e in d ic a te th is on the lin e b e lo w . We w ill n o tify y o u o f the d a te , tim e and p la c e . We a r e p la n n in g a f o llo w - u p stu d y in o r d e r to le a r n m o r e a b o u t how m a r r i a g e s p r o g r e s s o v e r a p e r io d of li m e . If you w o u ld lik e to p a r t ic ip a t e , p le a s e in d ic a t e th is on the lin e b e lo w . P l e a s e be a s s u r e d that a ll the in f o r m a t io n y o u h a v e g iv e n u s i s a b s o lu t e ly c o n f id e n t ia l. In fo r m m e of the g r o u p m e e t i n g . ___________ I w is h to p a r t ic ip a t e in Lhe f o ll o w - u p stu d y . N a m e_______________________________________________ ( p le a s e p rin t) A d d r e s s C ity APPENDIX B The Tables From The Miller Study 13^ / y r 135 TABLE X SOCIAL CLASS POSITIONS OF MALES FOR TOTAL INTERVIEWED SAMPLE AND THE SELECTED SAMPLE, BY PERCENTAGES Class Total Interviewed Sample N * 313 Selected Sample N = 210 1 3.8 - 2 20.1 30.0 3 h?.o 70.0 k 26.5 - 5 2.6 — 100.0 100.0 , v - / ¥ ' \ V * y 136 TABLE 2 OCCUPATIONAL POSITIONS OF MALES Occupational Positions N = 210 Per Cent 1 .5 2 20.0 3 39.0 b 31.0 5 9.5 6 — 7 100.0 TABLE 3 EDUCATIONAL POSITIONS OF MALES Educational Positions N = 210 Per Cent 1 5.2 2 31. ^ 3 51.5 k 11. ^ 5 ' .5 6 7 100.0 138 TABLE 4 AGE DISTRIBUTION OF SAMPLE, BY PERCENTAGES MALES FEMALES AGE GROUP N=210 N=210 48 and over .5 - 41-44 .3 - 37-40 2.9 1.9 33-36 2.9 1.4 30-32 7.4 5.3 30-28 24.3 10.0 21-24 52.9 45.2 17-20 8.6 36.2 100.0 100.0 TABLE 5 RELIGIOUS PREFERENCES OF THE BY PERCENTAGES SAMPLE, Religious Preference Male N = 210 Female N = 210 Protestant 57.1 58.1 Catholic 29.0 32.8 None 8.7 2.4 Other 5.2 6.7 TABLE 8 FREQUENCY DISTRIBUTION OF TRADITIONAL- EQUALITARIAN FAMILIAL ROLE SCORES OF THE SAMPLE Number Score Males Females Total 85-88 1 0 1 80-84 1 0 1 75-79 3 2 5 70-74 4 4 8 65-69 18 18 36 60-64 44 38 82 55-59 69 80 149 50-54 43 51 94 45-49 19 10 29 40-44 5 7 12 35-39 3 0 3 N 210 210 420 MEAN 57.3 57.3 57.3 MEDIAN 57 57 57 TABLE 9 FREQUENCY DISTRIBUTION OF THE ACCURACY OF PERCEPTION SCORES OF THE SAMPLE Number Score Males Females Total 100 0 1 1 95-99 0 0 0 90-91 + 5 3 6 85-89 Ik 5 19 80-8* f 15 20 33 75-79 23 17 40 7O-7A 53 53 106 65-69 28 32 60 60~6k 25 36 61 55-59 22 20 kz 50-5^ Ik 10 zh 45-^9 12 7 19 40-^ 2 5 7 55-59 1 1 2 N 210 210 ^20 MEAN 67.7 67.7 67.5 MEDIAN 70 67 70 ACCURACY OF PERCEPTION TABLE 10 ANALYSIS OF VARIANCE OF DIFFERENCE SCORES BETWEEN PRE-NUPTIAL PAIRS ON DOMINANCE (DOM) NEED BY FAMILIAL ROLE ATTITUDES AND ACCURACY OF PERCEPTION FAMILIAL ROLE ATTITUDES Traditional Male- Equalitarian Female or Traditional Male- Traditional Female Equalitarian Male- Traditional Female Equalitarian Male- Equalitarian Female Both Cell 11 Cell 12 Cell 13 Partners x = 8.36 x = 8.18 x 3 7.43 *x Accurate n - as n = 17 n » 14 One Cell 21 Cell 22 Cell 23 Partner x = 11.50 x = 9.17 x = 10.16 ix2 Accurate n » 26 n = 36 n » 25 Both Cell 31 Cell 32 Cell 33 Partners x = 9.19 x b 7.17 x = 10.89 **3 Inaccurate n b 16 n - 30 n ■ ■ 18 (i x - 29.05 $x 2 > 24.52 $X 3 = 28.48 *X.l Source oi Variation Sum of Sauares COMPLETED ANALYSIS OF VARIANCE Degrees of Freedom Mean Sauare F F.95 Rows 7.85 2 3.93 1.64 3.89 Columns 4.06 2 2.03 .85 3.89 Interaction 6.10 4 1.53 .64 2.41 Error 201 2.39 S+fl ACCURACY O F PERCEPTION TABLE 11 ANALYSIS OF VARIANCE OF DIFFERENCE SCORES BETWEEN PRE-NUPTIAL PAIRS ON AFFILIATION (LOV) NEED BY FAMILIAL ROLE ATTITUDES AND ACCURACY OF PERCEPTION Traditional Male- Traditional Female FAMILIAL ROLE ATTITUDES Traditional Male- Equalitarian Female or Equalitarian Male- Equalitarian Male- Traditional Female Equalitarian Female Both Cell 11 Cell 12 Cell 13 Partners x » .884 x » .812 x = .856 Accurate n - 28 n x 17 n - 14 J * * One Cell 21 Cell 22 Cell 23 £X2 Partner X a .800 x - .807 X . .798 Accurate n - 26 n - 36 n x 25 Both Cell 31 Cell 32 Cell 33 $*3. Partners x x .919 x - .941 x » .695 Inaccurate n - 16 n ■ 30 n - 18 ( x - 2.603 ixz - 2.560 *X.l *X.2 aScores have undergone a logarithmic transformation. ix 3 - 2.349 ^X.3 Source of Variation Sum of Sauares COMPLETED ANALYSIS OF VARIANCE Degrees of Freedom Mean Sauare F F.95 Rows .005 2 .003 .38 3.89 Columns .012 2 .006 .75 3.89 Interaction .028 4 .007 .88 2.41 trror 201 .008 - P - ACCURACY O F PERCEPTION TABLE 11 ANALYSIS OF VARIANCE OF DIFFERENCE SCORES BETWEEN PRE-NUPTIAL PAIRS ON TOTAL DIFFERENCE OF NEEDS SCORES BY FAMILIAL Traditional Male- Traditional Fenale ROLE ATTITUDES AND ACCURACY OF PERCEPTION FAMILIAL ROLE ATTITUDES Traditional Uale- Equalitarian Feeds or Equalitarian Male- Equalitarian Male- Traditional Female Eauelitarian Female Both Cell 11 Cell 12 Cell 13 Partner* x « 1.232 x - 1.169 x - 1.156 **1 Accurate n - 28 n » 17 n . 14 One Cell 21 Cell 22 Cell 23 Partner x - 1.210 x - 1.161 x - 1.202 <*2 Accurate n m 26 n ■ 36 n - 23 Both Cell 31 Cell 32 Cell 33 Partners x - 1.229 £ - 1.200 x - 1.210 Inaccurate n - 16 n > 30 n a 18 ' u * A - 3.671 i*.2 - 3-53° ix3 - 3.568 <*1 *X.2 *X.3 *Score* hare undergone a logarithmic transformation. COMPLETED ANALYSIS OF VARIANCE Sua of Decrees of Mean Source of Variation Sauare* Freedom Sauare F F.95 Bovs .001 2 .001 .25 3.89 Columns .003 2 .002 .50 3.89 Interaction .002 4 .001 .25 2.41 Error 201 .004 I*f5 TABLE 13 CORRELATIONS BETWEEN THE MALES' AND FEMALES’ SCORES AMONG SUBSAMPLES AND TOTAL SAMPLE Correlation Correlation on on Group N DOM Need LOV Need Mutually Traditional 70 .21* ,0* f Mutually Equalitarian 57 .26** -.02 Total Sample 210 .21* .08 *Significant at .05 level. **Significant at ,01 level. C ' : ' TABLE l*f CORRELATIONS BETWEEN THE PARTNERS' DIFFERENCES OF INTERPERSONAL NEED SCORES AND THE ACCURACY OF PERCEPTION SCORES, BY SEX, OF SUBSAMPLES AND TOTAL SAMPLE Group N DOM Differences and Accuracy of Perception LOV Differences and Accuracy of Perception Total Difference of Needs and Accuracy of Perception Mutually Traditional Male 6 70 -.13 -.08 i * H S J \ Females 70 C O 0 * 1 .07 -.01 Mutually Equalitarian Males 57 -.03 H H • .07 Females 57 -.17 .10 -.05 Total Sample Males 210 .02 -.05 -.02 Females 210 -.06 o o * i • o V Jl w Jr as 1^7 FIGURE 2 LAYOUT OF A TWO-WAY ANALYSIS OF VARIANCE DESIGN 11 12 13 21 22 23 31 32 33 I X 1* E X 3’ APPENDIX C The Circumplex Schema of the ICL 148 TABLE 1 1^9 ROLE ATTITUDES PERTAINING TO OCTANTS IN THE INTERPERSONAL CHECK LIST Role Type Octant Number Role Type (A) MANAGERIAL Octant 1 (P) AUTOCRATIC £ 1 Able to give orders 1 Well thought of 2 Forceful 2 Makes a good impression Good Leader Often admired Likes responsibility Respected by others 3 Bossy 3 Always giving advice Dominating Acts important Manages others Tries to be too 4a Dictatorial successful b Expects everyone to admire him (B) EXPLOITIVE Octant 2 (C) COMPETITIVE 1 Self-respecting 1 Able to take care of 2 Independent self Self-confident 2 Can be indifferent to Self-reliant and others assertive Businesslike 3 Boastful Likes to compete with Proud and self- others satisfied 3 Thinks only of himself Somewhat snobbish Shrewd and calculating b Egotistical and Selfish conceited b Cold and unfeeling A "I" role attitude is near the center of the circle and is considered an adaptive mechanism* The attitudes become less adaptive as they reach the periphery of the circle— the most maladaptive being a "V intensity* 150 AFFECTIONATE DOMINANCE HOSTILE DOMINANCE AP NO Lov Lov HI HOSTILE WEAKNESS ! Dom AFFECTIONATE . SUBMISSIVENESS FIGURE 1 A U T O C R A T I C P E R S O N A L I T Y N A R C I S S I S T I C P E R S O N A L I T Y M A N A G E R I A L P E R S O N A L I T Y '® H Y P E R N O R M A L \ P E R S O N A L I T Y \ a C O M P E T I T I V E P E R S O N A L I T Y R E S P O N S I B L E \ P E R S O N A L I T Y ' S A D I S T I C P E R S O N A L I T Y I ■ A G G R E S S I V E j P E R S O N A L I T Y 5 3--- \ R E B E L L I O U S \ P E R S O N A L I T Y \ \ C O O P E R A T I V E I P E R S O N A L I T Y j D O C I L E \ / P E R S O N A L I T Y ff / / V / S E L F ' ' E F F A C I N G D I S T R U S T F U L P E R S O N A L I T Y ^4/ P E R S O N A L I T Y 5 D E P E N D E N T P E R S O N A L I T Y M A S O C H I S T I C P E R S O N A L I T Y O V E R - C O N V E N T I O N A L / j P E R S O N A L I T Y j L M FIGURE 2 FIGURE 3 BIBLIOGRAPHY 153 BIBLIOGRAPHY Books Ackerman, Nathan W. The Psychodynamics of Family Life. New York: Basic Books, 1958. Baldwin, Alfred L* "The Parsonian Theory of Personality," The Social Theories of Talcott Parsons* Edited by Max Black. Englewood Cliffs, New Jersey: Prentice Hall, Inc., 1964. Bee, L. S. Marriage and Family Relations. New York: Harper Brothers, 1959. Bell, Norman W., and Vogel, Ezra F, (eds«). A Modern Introduc tion to the Family. Glencoe, 111.: The Free Press, I960. Blalock, Hubert M. Social Statistics. New York: McGraw-Hill Book Company, Inc., I960. Blood, Robert 0., Jr., and Wolfe, Donald M. Husbands and Wives. Glencoe, 111*: The Free Press, i960. Burchinal, Lee G., "The Premarital Dyad and Love Involvement," in Handbook of Marriage and the Family. Edited by Harold T. Christensen. Chicago: Rand McNally and Company, 1964. Burgess, E., and Cottrell, Leonard S. Predicting Success or Failure in Marriage. New York: Prentice-Hall, Inc., 1939. . and Locke, Harvey J. The Family: From Institution to Companionship. New York: American Book Company, 1953* . and Wallin, Paul. Engagement and Marriage. Phila delphia: J. B. Lippincott Company, 1953- Chance, Erika. Families in Treatment. New York: Basic Books, Inc., 1959. Cooley, Charles Horton. Social Organization. New York: Scribners, 1915* 154 155 Ellis, Albert. "The Value of Marriage Predictive Tests,1 ' Marriage and the Family. Edited by Robert F, Winch and Robert McGinnis. New York: Henry Holt and Company, 1952. Farber, Bernard, "Marital Integration Index," in Family Measure ment Techniques. Edited by Murray A. Straus. Minneapolis: University of Minnesota Press, 1969* Flugel, John C. Man. Morals and Society. New York: International Universities Press, 19^5* Foote, Nelson N., and Cottrell, Leonard S* Identity and Inter personal Competence. Chicago: The University of Chicago Press, 1955* Fromm, Eric. Escape From Freedom. New York: Rinehart, 19^1. Goode, W. and Hatt, P. Methods in Social Research. New York: McGraw-Hill, 1952. Grinker, Roy R. (ed.). Toward A Unified Theory of Human Behavior. New York: Basic Books, Inc., 1956. Guilford, J. P. Fundamental Statistics in Psychology and Educa tion. New York: McGraw-Hill Book Company, 1956. Hollingshead, August B., and Redlich, Frederich C. Social Class and Mental Illness. New York: John Wiley and Sons, 1958* Horney, Karen. Neurosis and Human Growth. New York: W. W. Norton and Co., 1950. Kenkel, William F. "Observational Studies of Husband-Wife Inter action in Family Decision-Making," Sourcebook in Marriage and the Family. Edited by Marvin B. Sussman. Boston: Houghton Mifflin Company, 1963. Kerlinger, Fred N. Foundations of Behavioral Research. New York: Holt, Rinehart and Company, 1966. Kirkpatrick, Clifford. The Family, as Process and Institution. New York: The Ronald Press, 1955* Ktsanes, Thomas and Ktsanes, Virginia. "The Theory of Comple mentary Needs in Mate-Selection," Selected Studies in Marriage and the Family. Edited by Robert F, Winch and Robert McGinnis. New York: Henry Holt and Company, 1953* Leary, Timothy. The Interpersonal Diagnosis of Personality. New York: The Ronald Press Company, 1957* 156 Locke, Harvey J* Predicting Adjustment in Marriage: A Compari son of a Divorced and a Happily Married Group. New York: Henry Holt and Company, 1951* Long, Joseph R. A Treatise on the Law of Domestic Relations. St. Paul: Keefe-Davidson Company, 1905* Luckey, Eleanore Braun. "Marital Interaction and Perceptional Congruence of Self and Family Concepts," Sourcebook in Marriage and the Family. Edited by Marvin B. Sussman. Boston: Houghton Mifflin Company, 19&3. Martindale, Don. The Nature And Types of Sociological Theory. Boston: Houghton Mifflin, i960. Meade, George Herbert. Mind. Self, and Society. Chicago: University of Chicago Press, 193^» Miller, Daniel R. and Swanson, Guy E. The Changing American Parent. New York: John Wiley and Sons, Inc., 1958. Murray, Henry A., et al. Explorations in Personality. New York: Oxford University Press, 1938. Newcomb, Theodore M, Social Psychology. New York: The Dryden Press, 1950* Nye, F. Ivan, and Berardo, Felix M. Emerging Conceptual Frame" works In Family Analysis. New York: The MacMillan Company, 1966. Parsons, Talcott, and Robert F. Bales. Family. Socialization and Interaction Process. Glencoe, Illinois: The Free Press, 1955* Parten, M. B, Surveys. Polls, and Samples. New York: Harper and Row, 1950« Peterson, James A. Education for Marriage. New York: Charles Scribner’s Sons, 1956* ________, Towards Successful Marriage. New York: Charles Scribner's Sons, 19o0. Rose, Arnold M. Sociology (2nd ed., revised). New York: Alfred A. Knopf, 1956. Sarbin, Theodore R. "Role Theory," Handbook of Social Psychology. 2 vols. Gardner Lindzey, (ed.) . Cambridge, Massachusetts: Addison-Wesley Publishing Company, 195^ • 157 Schutz, William C. FIRO: A Three-Dimensional Theory of Inter personal Behavior. New York: Rinehart and Company, 1958. Selltiz, Claire, et al. Research Methods in Social Relations. New York* Holt, Rinehart and Winston, 1963* Siegel, Sidney. Nonparametric Statistics for the Behavioral Sciences. New York: McGraw-Hill Book Company, Inc., 1956* Stryker, Sheldon. "Role-Taking Accuracy and Adjustment," Source book in Marriage and the Family. Edited by Marvin B. Sussman. Boston: Houghton Mifflin Company, 1963* Sullivan, Harry Stack. The Interpersonal Theory of Psychiatry. New York: W. W. Norton and Co., 1953* Sussman, Marvin G. Sourcebook in Marriage and the Family. Boston: Houghton Mifflin Company, 19^3* Toman, Walter. Family Constellation: Theory and Practice of a Psychological Game. New York: Springer Publishing Company, 1961. Udry, J. Richard. The Social Context of Marriage. New York: J* B. Lippincott Company, 1966. Walker, Helen M., and Joseph lev. Statistical Inference. New York: Holt, Rinehart and Winston, 1953- Waller, Willard, and Reuben Hill. The Family. A Dynamic Inter pretation. New York: Dryden Press, 1951* Winch, Robert F. Mate Selection: A Study of Complementary Needs. New York: Harper and Brothers, 1958* . The Modern Family. New York: Henry Holt and Company, 1952. . Robert McGinnis and Herbert Barringer, (eds.). Selected Studies in Marriage and the Family Crev. ed.) New York: Holt, Rinehart and Winston, 1965. Zelditch, Morris. "Role Differentiation in the Nuclear Family: A Comparative Study," in Family. Socialization and Inter action Process. Edited by Talcott Parsons and Robert Bales. New York: The Free Press, 1955• 158 Articles in Periodicals Alexander, Franz. "Psychoanalysis in Western Culture." American Journal of Psychiatry. CXII (March, 1956), 697-705. Armstrong, R. G. "The Leary Interpersonal Check List: A Reliability Study." Journal of Clinical Psychology, XIV (1958), 393-39^. Blazer, John A. "Complementary Needs and Marital Happiness," Marriage and Family Living. XXV (February, 1963)« 89-95* Bodarky, Clifford. "A Study of College Students Anticipated Role Behavior in Marriage," Dissertation Abstracts. XX (September, 1959)» 1093* Bolton, C. D. "Mate Selection as the Development of a Relation ship." Marriage and Family Living. XXIII (1961), 23^-240. Bowerman, Charles E. "Assortative Mating By Previous Marital Status: Seattle, 1939-19^6." American Sociological Review. XVIII (1953), 170-177. . and Day, Barbara R. "A Test of the Theory of Comple mentary Needs as Applied to Couples During Courtship," American Sociological Review, XXL (October, 1956), 602- Burgess, Ernest W., and Paul Wallin. "Homogamy in Personality Characteristics." Journal of Abnormal and Social Psycho logy. XXXIX (October, 19^), ^75-481. Centers, R. "Marital Selection and Occupational Strata." American Journal of Sociology. LIV (19^9), 530-535* Christensen, H. T. "Student Views on Mate Selection." Marriage and Family Living. IX (19^7), 83-88. Clark, E. L. "Educational Backgrounds in Af-sortative Mating." American Psychologist. IV (19^9), 287-288. Coombs, Robert H. "Reinforcement of Values in the Parental Home as a Factor in Mate Selection." Marriage and Family Living. XXIV (May, 1962), 155-157. Corsini, R. J. "Understanding and Similarity in Marriage." Journal of Abnormal and Social Psychology, LII (1956), 327-332. 159 Couch, Carl J. "The Use of the Concept 'Role' and Its Deriva tives in a Study of Marriage." Marriage and Family Living, XX (November, 1958), 353-357. Cronbach, Lee J. "Processes Affecting Scores on 'Understanding of Others' and 'Assumed Similarity.'” Psychological Bulletin. LII (May, 1955), 177-193. Dinitz, S., Banks, F., and Pasamanick, B. "Mate Selection and Social Class: Changes During The Last Quarter Century." Marriage and Family Living. XXII (i960), 3^8-351. Duvall, Evelyn M. "Growing Edges in Family Life Education." Marriage and Family Living. VI (19^), 18-25* Dyer, William G, "Analyzing Marital Adjustment Using Role Theory." Marriage and Family Living. XXIV (November, 1962), 371-375- , and Urban, Dick. "The Institutionalization of Equali tarian Family Norms." Marriage and Family Living, XX (February, 1958), 53-58. Dymond, Rosalind F. "The Relation of Accuracy of Perception of the Spouse and Marital Happiness." American Psychologist, VIII (August, 1953), 3^. Ehrenwald, Jan. "Family Diagnosis and Mechanisms of Psychosocial Defense." Family ProcesB, II (March, 1963), 121-131. Freeman, L. "Homogamy in Interethnic Mate Selection." Sociology and Social Research, XXXIX (1955), 369-377. Gage, N. L., and Cronbach, Lee J. "Conceptual and Methodological Problems in Interpersonal Perception." Psychological Review. LXII (November, 1955, ^11-^22. Gianopulos, Artie, and Mitchell, Howard E. "Marital Disagreement in Working Wife Marriages as a Function of Husband's Attitude Toward Wife’s Employment." Marriage and Family Living, XIX (November, 1957), 373-378. Glick, Paul C. "Intermarriage and Fertility Patterns Among Persons in Major Religious Groups." Eugenics Quarterly. vii (i960), 31-38. ________ , and Landau, E. "Age as a Factor in Marriage," American Sociological Review,XV (1950), 517-529. l6o Gray, Horace. "Psychological Types in Married People," Journal of Social Psychology, XIX (May, 19^9), 189-200. Hastorf, A. H., and Bender, I. E. "A Caution Respecting the Measurement of Empathic Ability." Journal of Abnormal and Social Psychology. XLVII (April, 1952), 57^-576. Heiss, Jerold S,, and Gordon, Michael. "Need Patterns and the Mutual Satisfaction of Dating and Engaged Couples." Journal of Marriage and the Family, XXVI (August, I96A), 337-339. Hill, Reuben, and Hansen, Donald A. "The Identification of Con ceptual Frameworks Utilized in Family Study." Marriage and Family Living, XXII (November, i960), 299-311. Hobart, Charles W., and Lindholm, Lauralee. "The Theory of Complementary Needs, a Re-Examination." The Pacific Sociological Review. VI (Fall, 1963), 73-79- Hollingshead, August B. "Cultural Factors in the Selection of Marriage Mates." American Sociological Review. XV (October 1950), 619-627. Jacobson, Alver Hilding. "Conflict in Attitudes Toward the Marital Roles of Husband and Wife." Research Studies of the State College of Washington. XIX (June, 1951), 103- 106. . "Conflict of Attitudes Toward the Roles of Husband and Wife in Marriage." American Sociological Review. XVII (April, 1952), 1^6-150. Katz, A. M., and Hill, R. "Residential Propinquity and Marital Selection: A Review of Theory, Method, and Fact." Marriage and Family Living. XX (1958), 27-3^. Kelley, E. L. "Psychological Factors in Assortive Mating." Psychological Bulletin. XXXVII (19^0), V?3. ________ • "Marital Compatibility as Related to Personality Traits of Husbands and Wives as Rated by Self and Spouse." "Journal of Social Psychology. XIII (February, 19^1), 193- 19^ Kent, D. "Subjective Factors in Mate Selection." Sociology and Social Research. XXV (l95l)» 391-398. 161 Kerckhoff, Alan C., and Davis, Keith E. "Value Consensus and Need Complementarity in Mate Selection." American Socio logical Review, XXVII (June, 1962), 295-30?* Kernodle, Wayne. "Some Implications of the Homogamy-Complementary Needs Theories of Mate Selection for Sociologial Re search." Social Forces, XXXVIII (December, 1959), 145-152. Kirkpatrick, C. "Statistical Investigation of the Psychoanalytic Theory of Mate Selection." Journal of Abnormal Psychology. (October, 1937), 427-430. Kogan, Kate L., and Jackson, Joan K. "Perceptions of Self and Spouse: Some Contaminating Factors." Journal of Marriage and the Family. XXVI (February, 196*0, 60-64. Komarovsky, Mira. "Cultural Contradictions and Sex Roles." American Journal of Sociology, LII (November, 1946), 184- 189. Ktsanes, T. "Complementary Needs in Mate Selection." Disserta.tiai Abstracts. XIV (1954), 882-883. ________ . "Mate Selection on the Basis of Personality Type: A Study Utilizing an Empirical Typology of Personality." American Sociological Review. XX (October, 1955), 547-551* LaForge, Rolfe, and Suczek, Robert F. "The Interpersonal Dimen sion of Personality: III. An Interpersonal Check List." Journal of Personality. XXIV (September, 1955), 94-112. Leary, Timothy, and Coffey, Hubert S. "Interpersonal Diagnosis: Some Problems of Methodology and Validation." Journal of Abnormal and Social Psychology. L (January, 1955), 110- 124. Levinson, Daniel J., and Huffman, Phyllis E„ "Traditional Family Ideology and Its Relation to Personality." Journal of Personality. XXIII (March, 1955), 251-273. Lu, Yi-Chuang. "Marital Roles and Marital Adjustment." Socio logy and Social Research. XXXVi (July, 1952), 364-368. . "Predicting Roles in Marriage." American Journal of Sociology. LVIII (July, 1952), 51-55* Luckey, Eleanore Braun. "Marital Satisfaction and Its Associa tion with Congruence of Perception." Marriage and Family Living. XXII (February, i960), 49-54. 162 Mangus, A. R. "Family Impacts on Mental Health." Marriage and Family Living. XIX (August, 1957), 256-262. . "Hole Theory and Marriage Counseling." Social Forces. XXXV (March, 1957)* 200-209. McKee, John P., and Sherriffs, Alex C. "Men's and Women's Beliefs, Ideals, and Self-Concepts." American Journal of Sociology. LXIV (January, 1959), 356-3^3* Middleton, Russell, and Putney, Shell. "Dominance in Decisions in the Family: Race and Class Differences." The American Journal of Sociology. LXV (May, i960), 605-609* Motz, Annabelle Bender. "The Role Conception Inventory: A Tool for Research in Social Psychology." American Sociological Review. XVII (August, 1952), ^65-^71* Murstein, B. "The Complementary Need Hypothesis in Newlyweds and Middle-Aged Married Couples." Journal of Abnormal Social Psychology. LXIII (1961), 19^-197. . "Stimulus— Value— Role: A Theory of Marital Choice." Journal of Marriage and the Family, XXXII (August, 1970), 4f)5“h6l. Powell, Reed M., and LaFave, Lawrence. "Some Determinants of Role-Taking Accuracy." Sociology and Social Research. (May-June, 1953)1 319-32(u Roos, D. "Complementary Needs in Mate-Selection." Dissertation Abstracts. XVII (1957), ^26-^27. Rosow, Irving. "Issues in the Concept of Need-Complementarity." Sociometry. XX (September, 1957), 216-233* Schellenberg, James A. "Homogamy in Personal Values and the 'Field of Eligibles,'" Social Forces, XXXIX (December, I960), 157-162. ________, and Bee, Lawrence S. "A Re-Examination of the Theory of Complementary Needs in Mate Selection." Marriage and Family Living, XXII (August, i960), 227-232. Schooley, M. "Personality Resemblances Among Married Couples." Journal of Abnormal Psychology, (October, 1936), 3^0-3^1. Spiegel, John P. "The Resolution of Role Conflict Within the Family." Psychiatry, XX (February, 1957), 1-16. 163 Steinmann, Anne G. "The Concept of the Femine Role in the American Family: A Study of the Concept of the Feminine Role of Fxfty-one Middle-Class American Families." Dissertation Abstracts. 19 (1958), 899* . "Lack of Communication Between Men and Women." Marriage and Family Living. XX (1958), 350-552. Strauss, A. "Personality Needs and Marital Choice." Social Forces. XXV (March, 19V?), 332-335* Stuart, I. R. "Complementary vs. Homogeneous Needs in Mate Selection." Journal of Social Psychology. LVI (1962), 291-300. Tharp, Roland G. "Psychological Patterning in Marriage." Psychological Bulletin. LX (March, 1963), 97-117* Winch, Robert F. "The Theory of Complementary Needs in Mate Selection: Final Results on the Test of the General Hypothesis." American Sociological Review. XX (October, 1955), 552-555* . Ktsanes, Thomas and Ktsanes, Virginia. "A Theory of Complementary Needs in Mate Selection: An Analytic and Descriptive Study." American Sociological Review. XIX (June, 195*0, 2*fl-2*f9. Unpublished Material Hollingshead, August B. "Two Factor Index of Social Position." New Haven, Connecticut: Copyrighted by August B. Hollings head, 1957* Mimeographed. Hurvitz, Nathan. "Marital Roles and Adjustment in Marriage in a Middle—Class Group." Unpublished Ph.D. dissertation, University of Southern California, 1958. King, Elsie V. "Personality Characteristics--Ideal and Perceived in Relation to Mate Selection." Unpublished Ph.D. disser tation, University of Southern California, 1961. Kotlar, Sally Lee. "Middle-Class Marital Roles— Ideal and Per ceived in Relation to Adjustment in Marriage." Unpublished Ph.D. dissertation, University of Southern California, 1961. 16k Lossner, Walter M. "Complementarity, Homogeneity, Heterogeneity and Marital Stability. Unpublished Ph.D. dissertation, University of Southern California, 1968. Miller, Louis I. "Familial Role Typology, Accuracy of Perception, and Mutual Needs Among Pre-Nuptial Partners." Unpublished Ph.D. dissertation, University of Southern California, 1966. Taylor, A. B. "Role Perception, Empathy, and Marital Adjustment." Unpublished Ph.D. dissertation, University of Southern California, 1965*

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

Power Relationships In Marital Discord

PDF

Communication Between Engaged Couples

PDF

Role Perception, Empathy, And Marital Adjustment

PDF

A Study Of The Relationships Between Religious Affiliation, Religious Practices And Marital Adjustment

PDF

Attitudes Toward Sex In Marriage And Patterns Of Erotic Behavior In Dating And Courtship Before Marriage

PDF

An Analysis Of The Changing Focus Of Marriage Counseling

PDF

Familial Role Typology, Accuracy Of Perception, And Mutual Needs Among Pre-Nuptial Partners

PDF

On Becoming A Parent: Attitude And Feeling Changes

PDF

A Study Of Relationships Between Occupational And Marital Roles And Marital Adjustment

PDF

Status consistency and its effects on marital adjustment and stability

PDF

An Exploratory Analysis Of Two Dimensions Of Family Separation

PDF

Complementarity, Homogeneity, Heterogeneity And Marital Stability

PDF

Power Struggle And Organizational Goals: A Case Study Of An Industrial Firm

PDF

An Analysis Of Factors In The Family'S Withdrawal Of A Patient From A Hospital For The Mentally Retarded

PDF

An Analysis Of The Positive And Negative Functions In The Contemporary Usage Of An Ancient Rite Of Passage

PDF

Some Characteristics Of A Selected Sample Of Free Methodist Pastoral Dropouts

PDF

Personality Characteristics: Ideal And Perceived In Relation To Mate Selection

PDF

The Parameters Of Singleness: An Inquiry Into Some Factors Influencing The Choice Of Singleness Over Marriage As A Way Of Life

PDF

A Study Of Group Development In Purposive Groups

PDF

Personality And Interpersonal Factors Associated With The Duration Of Marriage Counseling

Asset Metadata

Creator Fowler, Charles Raymond (author)

Core Title An Investigation Of Complementary Needs Between Marital Partners

Degree Doctor of Philosophy

Degree Program Sociology

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

Tag OAI-PMH Harvest,sociology, individual and family studies

Language English

Contributor Digitized by ProQuest (provenance)

Advisor Peterson, James A. (committee chair), Hill, William F. (committee member), Lasswell, Thomas E. (committee member)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c18-518638

Unique identifier UC11362325

Identifier 7127921.pdf (filename),usctheses-c18-518638 (legacy record id)

Legacy Identifier 7127921.pdf

Dmrecord 518638

Document Type Dissertation

Rights Fowler, Charles Raymond

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