University of Southern California Dissertations and Theses

A study of the effects of a race relations education program on attitudes of racial prejudice

(USC Thesis Other)

A study of the effects of a race relations education program on attitudes of racial prejudice

PDF

Download

Open document

Flip pages

Download a page range

Download transcript

Copy asset link

Request this asset

Transcript (if available)

Content A STUDY OF THE EFFECTS OF A RACE RELATIONS EDUCATION PROGRAM ON ATTITUDES OF RACIAL PREJUDICE by Frank Alan Rowe, Sr. 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 (Education) June 1976 UMI Number: DP24166 All rights reserved INFORMATION TO ALL USERS The quality of this reproduction is dependent upon the quality of the copy submitted. In the unlikely event that the author did not send a complete manuscript and there are missing pages, these will be noted. Also, if material had to be removed, a note will indicate the deletion. Dissertation PWMshing UMI DP24166 Published by ProQuest LLC (2014). Copyright in the Dissertation held by the Author. Microform Edition © ProQuest LLC. All rights reserved. This work is protected against unauthorized copying under Title 17, United States Code ProQuest LLC. 789 East Eisenhower Parkway P.O. Box 1346 Ann Arbor, Ml 48106- 1346 UNIVERSITY O F S O U T H E R N CA LIFORNIA TH E GRADUATE SCH O O L UNIVERSITY PARK LOS A NGELES. C A LIFO R N IA 9 0 0 0 7 This dissertation, written by FRANK ALAN ROWE, SR under the direction of A . . . ™ . Dissertation C om mittee, and approved by all its members, has been presented to and accepted by The Graduate School, in partial fulfillment of requirements of the degree of D O C T O R O F P H I L O S O P H Y Dean DISSERTATION COMMITTEE Chairman ACKNOWLEDGMENTS The author wishes to thank the following persons for their assistance and encouragement during this study: My dissertation committee Dr. Earl V. Pullias, Chairman Dr. Paul A. Bloland Dr. Anne M arie Bird My form er colleagues at March Air Force Base, Head quarters, USAF, and The Defense Race Relations Institute Dr. Leo Richards, an original member of the five-person dissertation committee The faculty and staff at the University of Southern California School of Education My wife, Alicia TABLE OF CONTENTS ACKNOWLEDGMENTS 11 LIST OF TABLES v Chapter I. THE PROBLEM 1 Background of the Problem Statement of the Problem Situation Purpose of the Study Importance of the Study Questions to be Answered Hypotheses Conceptual Assumptions Definitions Delimitations Organization of the Remainder of the Study Background of the Present Investigation Research Design and Control Procedures Subject Selection and Classification Pilot Study Data Collection, Processing, and Analysis Methodological Assumptions Limitations of the Study II. REVIEW OF THE LITERATURE 11 Conceptual Background Experimental Precedents Summary III. METHODOLOGY 32 IV. FINDINGS 50 in C hapter Effect of Participation in a Race Relations Education Program on Affective and Cognitive Scores Affective and Cognitive Scores Three Months after Treatment and Posttesting Effect of Treatm ent on Selected Demographic Groupings V. DISCUSSION AND INTERPRETATION............................. 64 VI. SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS.......................................................... 69 Summary Conclusions Recommendations APPENDIXES A. The Rowe Social Attitude Survey. . . ............................. 78 B . Answer Sheet . .................................................................... 83 REFERENCES......................................... 85 i iv LIST OF TABLES Table 1. Demographic Grouping of Subjects by Type of Test and Observation N um ber.................................................... 51 i 2. Pearson Correlation Coefficients of Affective and Cognitive Items, Intolerance of Ambiguity, and Social Problems Q uestionnaire......................................... 52 3. Analysis of Variance on Posttest Scores and Mean Scores of Affective Items for All S u b je c ts ................... 53 i 4. Analysis of Variance on Posttest Scores and Mean Scores of Cognitive Items for All S u b jects................... 54 5. Repeated M easures Analysis of Variance on Posttest and Delayed Posttest Scores and Mean Scores of Affective Items for All Subjects in Groups 3 and 4 . . . 55 6. Repeated M easures Analysis of Variance on Posttest and Delayed Posttest Scores and Mean Scores of Cognitive Items for All Subjects in Groups 3 and 4. . . 56 7. Repeated M easures Analysis of Variance of Pretest and Posttest Scores and Mean Scores of Cognitive Items for Subjects in the 17-22, 23-30, and 31-40 Year Age G ro u p s.................................................................. 59 8. Repeated M easures Analysis of Variance of Pretest and Posttest Scores of Affective Items for Black and White S ubjects................................................................ 60 9. Repeated M easures Analysis of Variance of Pretest and Posttest Scores of Cognitive Items for Black and White S ubjects............................................................... 61 v Table 10. Repeated M easures Analysis of Variance of Posttest and Delayed Posttest Scores of Affective Items for White and Black S u b je c ts ................ 62 11. Repeated Measures Analysis of Variance of Posttest and Delayed Posttest Scores of Cognitive Items for White and Black S ubjects.................................................... 63 ! vi CHAPTER I THE PROBLEM Background of the Problem When racial tension occurs in the armed forces, it is not only a divisive social problem, but also a detriment to mission effectiveness. Recognizing the need for social harmony, the Depart ment of Defense (DOD) in 1969 expressed its philosophy of human | relations through its Human Goals Proclamation. It declared that the I DOD, an arm of the government which exists to keep peace, must be I guided by a founding principle of the nation: that individual dignity i i and worth are of central importance. In more operational term s, j | this philosophy was to be expressed in DOD Directive 1322.11 which j | stated that one essential element in organizational and combat readi ness was harmony among m ilitary personnel. However, there was still evidence that by 1972 these goals were not being reached. Between April and November, 1972, DOD conducted a survey designed to determine the extent of racial discrim ination in the administration of m ilitary justice. Discrimination was suspected since there was a disparity in disciplinary rates between black and white servicemen. The task force discovered that blacks received nonjudicial punishment disporportionate to their numbers, longer duration of confinement for sim ilar offenses, and a lower pro portion of honorable discharges than whites of sim ilar aptitude and education (U. S. , 1972). In the belief that racial harmony could be created through education, in conjunction with other social actions, DOD initiated a program intended to become a national model of equal opportunity and treatm ent of its personnel. One part was the United States Air Force (USAF) Race Relations Education Program, to be conducted at all Air Force installations above a certain minimum population. The objec tive of the program was to change prejudicial behavior by, first, increasing the awareness of individuals about the histories, cultures, and values of various racial groups in this multiethnic society and, second, by increasing their knowledge of racial stereotypes, of racial ! myths, and of the unintentional and covert institutional racism that i ! exists in the society. This broad study of prejudice was expected to i reduce intolerant and nonharmonious behavior, especially if the educa tion portion was coordinated with a wide range of vigorous activities jaimed at overcoming institutional racism . Statement of the Problem Situation j Contemporary awareness of the social and economic costs 2 of allowing overt or covert discrimination to exist in any institution has caused an increased interest in program s such as the USAF Race Relations Education Program. However, such training is costly in i term s of man-hours expended. Everyone in a m ilitary uniform, regardless of rank, is directed to complete a 14-hour course. Recently, civilian employees of DOD were asked to participate. As of 31 December 1975, 938,315 m ilitary and 166,521 civilian personnel i l attended race relations training at A ir Force installations. To date (1976), there has been little em pirical data gathered to show whether race relations training could cause lasting changes in prejudicial attitude. At present, the Air Force is engaged in descriptive survey research designed to m easure possible behavior I |changes associated with the training. However, there has been no ) instrument employed to m easure cognitive acquisition or affective | changes by participants in the training. I Since cognition, affect, and behavior are the three compon- ( ents of an attitude (Zimbardo & Ebbesen, 1970), and the relationship of the three is closely interwoven (Harding, Kutner, Proshansky, & Chein, 1954), it seemed appropriate to evaluate the first two compon ents (cognition and affect) in this study of changes in prejudicial attitudes, since the Air Force had limited its study to behavioral changes. 3 Purpose of the Study This study was designed to generate data which would (1) evaluate the ability of a race relations education program to im part information on prejudice and racism and facilitate changes in preju dicial attitudes; (2) reflect the extent to which the effects of the pro gram tended to persist; (3) indicate if there were differential effects from the race relations education program on groups which varied in age, sex, education level, ethnic background, and geographic origin; I (4) aid program adm inistrators in curriculum decisions for future race relations education program s by assisting them in determining ! which demographic groups were most and least affected by the train - i | i n g . i j | Importance of the Study | In addition to evaluating the extensive use of resources and i the resulting curriculum improvements that could occur, a third benefit was seen. If the investigation indicated that the instrument i developed for this study was effective in measuring attitude change, then norms could be established and an easily adm inistered self testing program developed, so that each m ilitary installation could periodically evaluate the effectiveness of its own race relations education program. 4 Questions to be Answered This study sought to answer the following m ajor questions: 1. To what extent, if any, did participation in a race re la tions education program become associated with acquisi tion of knowledge and with changes in prejudicial attitudes? 2. To what extent, if any, did participation in a race re la tions education program have significantly differential effects on acquisition of knowledge of racism and preju dice and changes in prejudicial attitudes by personnel who differed in age, sex, education level, ethnic back ground, and geographic origin? 3. To what extent, if any, did the effect of participation in a race relations education program differ three months i after the treatm ent, compared to immediately after the treatm ent? 4. In light of this study, should the USAF Race Relations Education Program, Phase II, continue in the present format regarding its curriculum and target population? Hypotheses 1. In term s of a pretest and an immediate posttest designed to m easure changes in the cognitive and affective com ponents of attitudes regarding prejudice and racism , attendance at the USAF Race Relations Education Pro gram would cause a difference between the scores of subjects in the experimental group and the control group, as measured by the Rowe Social Attitude Survey (RSAS). 2. In term s of an immediate posttest and a delayed posttest designed to m easure changes in 1, above, attendance at the USAF Race Relations Education Program would cause a difference between the scores of subjects in the exper imental group and the control group, as measured by the RSAS. 3. In term s of the three tests mentioned above, attendance at the USAF Race Relations Education Program would cause the scores of subjects in the experimental group to differ from the scores of subjects in the control group. ■ as measured by the RSAS, regardless of their demo graphic groupings. Conceptual Assumptions The following conceptual assumptions were implicit in the investigation: 1. It was assumed that there was a relationship between 6 attitude and behavior; that an attitude consisted of cogni tive, affective, and behavioral components; and that cognitive and affective components were measurable on a unidimensional scale. 2. It was assumed that attitude change might precede but would more likely follow behavior change and would occur where there was dissonance among the components of attitude. 3. It was assumed that the personality which demonstrated an intolerance of ambiguity would be more prejudiced on racial m atters than a person displaying tolerance of ambiguity. 4. It was assumed that behavior change would result from a program which provided information about and increased the sensitivity toward minority groups, provided that the program was accompanied by a continuing environment containing clear evidence that institutional or personal racism would not be tolerated. Definitions Attitude. —An enduring organization of motivational, emo tional, perceptual, and cognitive processes with respect to some aspect of the individuars world (Krech & Crutchfield, 1948). 7 Cognition. --Any knowledge, opinion, or belief about the environment, about oneself, or about onefs behavior (Festinger, 1957). Ethnocentrism . --The tendency toward a pervasive and rigid distinction between ingroups and outgroups, with hostility toward outgroups and a submissive attitude toward and belief in the rightness of ingroups (Rogers, 1961). It contains two essential ingredients: erroneous generalization and hostility (Allport, 1954). Institutional racism . --The everyday policies and practices of organizations which deny equal opportunity to minority races. It is viewed as having dysfunctional consequences for the organization j as well as for minority groups (Pecorella, 1975). Intolerance of ambiguity. —The tendency to perceive ambigu ous situations as sources of threat. An ambiguous situation is one i which cannot be adequately categorized by an individual due to the lack of sufficient clues, such as a completely new situation in which there are no fam iliar clues, a complex situation with many clues, or an insoluble situation with contradictory elements (Budner, 1962). Prejudice. — A hostile or negative attitude toward a distin guishable group based on generalizations derived from faulty or incomplete information (Aronson, 1972). Prejudgments become prejudices only if they are not reversible when exposed to new knowl edge. A prejudice, unlike a simple misconception, is actively re s is t ant to all evidence that would change it. Its ingredients are hostility and categorical rejection (Allport, 1954). Racism. — Any attitude, action, or institutional structure which lim its the rights of a person or group because of color, race, or ethnic differences (U. S. , n .d .). Stereotype. —The attribution of identical characteristics to any person in a group, regardless of the actual variation among m em bers of the group. It is an overgeneralization and has negative connotation (Aronson, 1972). Tolerance. —Simply the absence of prejudice, either positive or negative (Martin, 1964). Delimitations 1. The study was limited to 200 randomly selected m ilitary personnel on active duty at March Air Force Base, California, from November 1975 to March 1976. 2. The Rowe Social Attitude Survey (RSAS) that was used in this study was constructed by the investigator and in cludes portions of the Social Problems Questionnaire (Robinson, Rusk, & Head, 1968) and the Intolerance of Ambiguity Questionnaire (Martin & Westie, 1959). 3. The study was limited to the present course entitled the USAF Race Relations Education Program, Phase II. 4. The instrum ent used in this study was designed to m easure the cognitive and affective components of attitudes, but not the behavioral component. Organization of the Remainder of the Study Chapter II presents a theoretical framework of attitude and behavior change, plus selected literature and previous research in m easurem ent of prejudicial attitudes. Chapter III sets forth the methodology of the study, includ ing a discussion of the development of the test instrum ent, the pilot i j study, research design and control procedures, subject selection and j classification, data collection, processing, and analysis, methodo- | i j logical assumptions, and limitations of the investigation. | Chapter IV presents the findings of the study. i ! Chapter V provides a discussion and interpretation of the i findings. Chapter VI consists of a summary of the study and its find ings, conclusions, and recommendations. CHAPTER II REVIEW OF THE LITERATURE | This chapter presents the theoretical and experimental i i framework upon which this investigation of the effectiveness of a course in race relations rested. The conceptual background, which is discussed first, served as a point of reference for the study. Various theoretical positions that led to the formation of the hypothe ses are the basis of the conceptual framework. A second section of the chapter examines existing studies relating to the m easurement of i i prejudice and educational program s designed to change prejudicial attitudes and behavior. Conceptual Background i The major determinants of prejudice are thought to be (1) displaced aggression, (2) personality needs, (3) economic and politi cal forces, (4) conformity to existing social norms (Aronson, 1972). The framework of these determinants serves as a focal point for expanding on the causes of prejudicial attitudes and on methods avail able in overcoming institutional racism . The first two causes, which 11 may be thought of as having a psychological basis, are included in the discussion under the headings of the nature of attitudes and attitude i |change. This background leads to the important concept of the preju- jdiced personality. The third and fourth causes of prejudice have a I sociological foundation. The third determinant of prejudice, namely, the effect of economic and political forces, is discussed only briefly t f | since it is beyond the scope of this research and is generally not i relevant in a m ilitary environment. The fourth cause of prejudice is included in the section on behavior change and the section that d is cusses the attempt by the Air Force to overcome institutional racism ; through such change. ! The nature of attitudes j The investigator chose Krech and Crutchfield’s (1948) defini tion of attitude for this discussion. They stated that an attitude is i i !”an enduring organization of motivational, emotional, perceptual and jcognitive processes with respect to some aspect of the individual’s world” (p. 152). This definition is complementary to the ’’ component” theory which states that an attitude consists of cognitive, affective, [and behavioral components. Zimbardo and Ebbesen (1970) stated: i ! The affective component consists of a person’s evaluation of, liking of, or emotional response to some object or person. The cognitive component has been conceptualized as a person’s belief about or factual knowledge of the object or person. The behavioral component involves the person’s overt behavior directed toward the object or person, (p. 7) _ 12 Many other authors (Gordon, 1972; Harding et al. , 1954; McGuire, 1969; Rokeach, 1968) support the component view which is central to this investigation. Two other aspects of attitudes, the unidimensional concept and the relationship between the components, require exploration, i An attitude is a learned predisposition to respond to an object or I class of objects in a consistently favorable or unfavorable way (Allport, 1935). Thus, the favorable or unfavorable aspect of an attitude can be thought of as a unidimensional concept and one that lends itself to measurem ent. Fishbein (1967) concurred with All- p ort’s concept of measuring an individual’s attitude along a contin uum. F urther, he found that while one given belief or a behavioral intention may be uncorrelated with an individual’s attitude, a totality (or even a larger set) of an individual’s behavioral intentions gives a good estim ate of his attitude. f I Regarding the interaction between the components, there is i no simple straightforward relationship between the cognitive and j affective parts of attitudes and behavior. Allport cautioned that the j unidimensional concept could be misleading, in attempting to predict behavior from attitude. For example, two people may feel equally favorable toward an object, yet have differing feelings about specific parts that com prise the object. The unidimensional character of attitude m easure does not cover the qualitative aspect of attitudes. Dobb (1947), in discussing this relationship, stated that although an attitude is a learned predisposition to respond, once the attitude is learned, the subject must also learn what response to make toward the object of the attitude. Severy (1974), in his analysis of attitudes and their relationship to behavior, stated that how someone feels about an object may or may not be reflected in his behavior. ’’ A tti tudes are simply one aspect of the behavioral situation. In a vacuum, one’s attitudes would lead directly to behavioral characteristics of these attitudes” (p. 2). However, behavior is continuously being constrained, limited, or prescribed by the social-psychological situa tio n (environment). McGuire’s (1969) findings indicated that there was a rather low correlation between a person’s verbal report of his attitude by a paper and pencil inventory and his actual behavior toward the object of the attitude. However, it is important to note that his ! results showed that the relationship between the affective and cogni tive components was more highly correlated than either of these with I behavior. t Not all theorists agree on the variance of intercorrelations ; among the three components of attitude. For example, Harding et al. (1954) reported that the three components correlated high in most studies. They concluded that since the relationship was so close, it would make little difference which components were used to rank individuals’ attitudes toward specific ethnic groups. Gordon (1972) __________________ . ____________ _______________________________________ 14 found that when affective and cognitive statements dealt with a given issue, they were highly related statistically to each other. McGuire (1969) affirmed this when he stated: Given the less than perfect state of our measuring proce dures, the three components have proven to be so highly intercorrelated that theorists who insist on distinguishing them should bear the burden of proving the distinction as worthwhile, (p. 157) When discussing the three components of attitude, it is important to note that they develop under varying circum stances. Attitudes are learned either from direct experience which relates to the cognitive and affective components or from other people, such as family or friends, who influence the behavioral component (Triandis, 1971). However, direct experience can also influence behavior since the three components interact and tend to become consistent with each other. This concept formed the basis for the race relations education training developed by the DOD. ! I ’ Attitude change Attitudes can be changed in a variety of ways (Triandis, 1971). The cognitive component may be altered involuntarily by the receiving of new information from m ass media, or through people, or voluntarily through the seeking of new data. The affective component may change through direct experience with the attitude object involv ing a pleasant or unpleasant episode with that object. A third way to change attitude is to force a person to behave in a way that is _______________________ 15 inconsistent with his existing attitudes, such as through legislation that^ would prohibit certain types of discrimination. Triandis cited a fourth way to change attitudes--through a "fait accompli. " Once an event has occurred, attitudes change to become consistent with the implication of the event. For example, a president -elect' s popularity is greater after the election than before. M alec's (1971) cognitive consistency theory of attitude change states that a single attitude exists within a field of many other a tti tudes. However, when attitudes are relevant to each other, they strive toward balance, symmetry, or consonance: "Attitudes, in their affective, behavioral and cognitive components, must ’fit' with each other so that the field is relatively stable. Any attitude which does not fit must change” (pp. 40-41). F estinger’s (1957) theory of cogni tive dissonance, one of the cognitive theories of attitude change, states that dissonance exists when relevant cognitive elements, or "cognitions, ” do not fit. The magnitude of the dissonance depends upon the importance and the proportion of the cognitions that are d is sonant. The key element in this theory is that dissonance produces pressure to change one of the cognitive elements. Once pressure occurs, dissonance may be reduced by (1) changing a cognitive ele ment, (2) adding new cognitive elem ents, or (3) changing a behavioral element, such as a smoker who stops his habit upon learning that it is detrim ental to his health. i 16 Festinger (1964) also cautioned about the stability of attitude change. When opinion or attitude is changed through a momentary persuasive communication, this change is "unstable and will dis~ i appear or remain isolated unless an environmental or behavioral change can be brought about to support and maintain it" (p. 414). If support is not provided, the same influences that caused the initial attitude will continue to act and will diminish the effect of the change, j F urther, people tend to avoid exposing themselves to new information i which they expect would increase dissonance. If forcibly exposed to it, they will evade its impact by disbelief, misperception, or some ; other device. i i | While a change of attitude does not necessarily predict a | change in behavior (Cohen, 1964; Tittle, 1965), the reverse is often | true (Aronson, 1972). Attitudinal change creates a context in which i much behavioral change becomes possible or, as Triandis (1971) ; stated, it can be a contributing cause. "What has only recently begun i to be understood is that changes in behavior can affect changes in attitudes" (Aronson, 1972:193). Greely and Sheatsley (1971) cited, for example, that the increasing support for school integration over i the past 30 years makes it somewhat easier for official policies of school integration to be pursued. Bern (1970) stated that although it is often difficult to separate cause and effect, "a close analysis . . . | indicates that favorable attitudes toward a particular desegregation I _______ : ____________________________________________________________ 17 m ore typically follow rather than precede the move" (p. 68). And last, regarding attitude change, Sherif, Sherif, and Nebergall (1965) established the principle important to this investigation that moderate attitudes are more susceptible to change than extreme attitudes. i The prejudiced personality i | Pertinent to a study of prejudice and attitude change is an I i i understanding of the prejudiced personality. When the origins of the f j prejudiced personality are explored, two of the causes of prejudice i | em erge, namely, displaced aggression and personality needs. One of the earliest w riters to develop the theme of the i j 1 "authoritarian personality" was Erich Fromm (1941), who saw this i ! I | trait as a basic foundation of Fascism . He postulated that man, i | freed from group ties due to the rise of capitalism , found him self i I alone, helpless, and isolated. Unless these feelings were overcome i ! by love of one’s fellowman and productive work, he would escape I i I from the freedom that accompanies the process of individuation, i i according to From m . Such escape mechanisms include authoritari anism , destructiveness, and conformity. Briefly, the authoritarian i wants to hurt and dominate others, while at the same time he wants to be hurt and dominated. He adm ires authority and tends to submit to it, while desiring to be an authority him self and have others sub- | mit to him. This authoritarian personality structure is rooted in a basic feeling of powerlessness and characterized by a worship of power and power relations, together with a lack of love and human tenderness. j Stimulated by From m 's theory, Adorno, Frenkel-Brunswik, Levinson, and Sanford (1950) developed scales of ethnocentrism and authoritarianism . They concluded that the authoritarian character | structure stem s from childhood experiences, such as subjection to harsh, inflexible discipline, high dependence upon parents, conditional parental love, a hierarchical family structure, and a cold clim ate in the home. This combination sets the stage for the emergence of an adult with a high degree of anger which, because of fear and insecur ity, takes the form of displaced aggression or scapegoating against powerless groups, while the individual maintains an outward respect jfor authority. I One of the traits of the authoritarian personality that Adorno | et al. isolated and classified under cognitive personality organization was rigidity or intolerance of ambiguity. The relationship between ethnocentrism and intolerance of ambiguity had been verified by other investigators (Frenkel-Brunswik, 1949; O’Connor, 1952; Taft, 1956). The individual who was intolerant of ambiguity tended to be rigid in his beliefs, possess "conventional” values, be intolerant of weakness in him self and others, and be highly punitive and suspicious. The intolerant person seemed to have a need for absolute dichotomies and I I 19 unambiguous solutions. In their study, Martin and Westie (1959) found that the tolerant person was able to perceive gradation, v aria tion, and relativity and was able to recognize that each individual, i regardless of group assignment, was unique. In fact, they deter- i mined that the characteristics of the two personality types differed enough in many social and personal respects to justify reference to prejudice and tolerance syndromes. Besides relating ethnocentric attitudes to intolerance of ambiguity, O’ Connor (1952) discovered that the intolerant subjects had poor abstract ability where intelligence was not a significant variable. Frenkel-Bruns wik (1948), in her test of children 11 to 16 years of age, found that subjects who were intolerant of ambiguity became anxious or resisted facts or situations that were not clearly structured. They also tended to dichotomize their thinking about sex roles and to regard position in their family in i I | term s of hierarchical roles. Frenkel-Brunswik concluded that both i i j tolerant and prejudiced thinking about ethnic groups was a reflection i ' I j of a total style of cognitive operation. An earlier study by Allport and K ram er (1946) showed sim i- | lar results using adult subjects. The highly prejudiced would endorse j such statem ents as: i +The world is a hazardous place in which men are basic ally evil and dangerous. +We do not have enough discipline in our American way of life. Although these propositions appear to have nothing to do with preju dice, the authors concluded that prejudice was often broadly woven into a total way of thinking—a conclusion sim ilar to that of Frenkel- Brunswik (1948). The prejudiced style of thinking is a reflection of the preju diced person’s way of thinking about anything. Allport (1954) saw a parallel between the dynamics of prejudice and cognition. The need for definiteness was likely to lead to a constriction of cognition, since the person failed to see all relevant sides to his problems. Allport found that tolerant people tended to take longer to establish a norm for themselves.^ Intolerant people clung to past solutions, feared saying "I don’t know, ” and developed an urge for quick and definite answers. Since their self-im ages were confused, and they I had suffered more psychological deprivation as children, prejudiced people were m ore susceptible to frustration. There was a strong need for clear-cut social order where lodges, schools, churches, or the nation served as leaning posts in the form of memberships or feelings of intense patriotism . Conversely, the high degree of self insight among tolerant subjects was related to a sense of humor. Apparently, one who could laugh at him self was unlikely to have the need to feel superior to others. And last, in looking at the value system s of tolerant and prejudiced personalities, Rim (1970) applied i Rokeach’s (1968) term inal and instrum ental values to Budner’s (1962) 21 Intolerance of Ambiguity Scale. He discovered that intolerant sub jects ranked the values of "polite, " "ambitious, ” "national security, " "a comfortable life, " and "social recognition" as significantly im por tant. Tolerant subjects stressed the values of "cheerful, " "broad minded, " "happiness, " and "inner harmony. " Demographers have contributed to an understanding of prejudice. Many theorists have stated that there is a relationship between the amount of education and the tolerance of a person, though they disagreed on the extent. Allport (1954) determined that general education appreciably raised the level of tolerance and that the gain would be passed on to the next generation. Berger, Leibly, and Meunier (1973) found that college seniors were significantly more i flexible than freshmen in their attitudes toward seven foreign nation- | alities. They also found that females were more flexible than m ales. j Cautioning that the variables were likely to be complex and in ter- ! active, Bettelheim and Janowitz (1964) found that the less prejudiced i | person was likely to be younger, better educated, and of a higher i i socioeconomic status. However, they warned that the more educated may be more skillful in obscuring their prejudices. Authoritarian ism was "particularly evident" among members of disadvantaged m inorities, lower socioeconomic strata, or the less educated, according to Stewart and Hoult’s (1959:277) study of the F Scale. ! Regarding geographic origin and sex, Selznick and Steinberg (1969), in their study on antiblack attitudes, found no difference in attitude among male and female subjects in the Northern sample, but the Southern sample showed some variability based on sex. Seventy-six percent of the males versus 69 percent of the females had negative attitudes about blacks. When geographic origin was considered by itself, 72 percent in the South and 40 percent in the North demon strated antiblack attitudes. ! Behavior change I As mentioned ea rlie r, the fourth m ajor cause of prejudice is conformity to existing social norms (Aronson, 1972). Though racial prejudice historically was greater against blacks in the South, particularly in towns where economic competition was great, the number of authoritarian personalities in the South was relatively the same as in the North (Pettigrew, 1959). This apparent discrepancy was simply caused by people conforming to existing norm s in the region where they lived. Pettigrew observed that the need to con form to social norm s outweighed economic factors, frustration, and personality as a cause of prejudice. Aronson concluded that the lack of accurate evidence or a preponderance of misleading information could lead people to adopt negative attitudes on the basis of hearsay. Fendrich (1967) found that prejudice often developed before any contact with minority groups occurred. Similarly, evidence showed that children tended to im itate the prejudices of their parents and ____________________________________________________________ 23 other adults (DuBois & Li, 1971). Any m easure to counter racial prejudice must take into account the whole structure of the prejudiced outlook (Adorno et a l . , j 1950). Programs limited to rational arguments, appeals to sympathy, or simply closer association with m inorities would all fail. Allport (1954) stated that pure informational program s were useless since [they were quickly forgotten or distorted. Several authors stressed that participating minority and m ajority m em bers must be of equal status (Bettelheim & Janowitz, 1964). Aronson (1972) showed that if blacks and whites could be brought into direct contact, prejudiced individuals would come into contact with the reality of their own i iexperience, not just a stereotype. Equal status was essential to [overcome the stereotype of whites toward blacks performing menial tasks, and to overcome the resentm ent of blacks in a position of p e r forming such tasks. Whether prejudice was largely a function of economic conflict, conformity to social norm s, or deeply rooted personality needs, it was not easily changed by an information pro gram since the thought pattern had persisted over the years. Aronson concluded his analysis by stating that "the best way to produce even tual interracial harmony would be to launch into behavioral change” (p. 196). Festinger (1957) agreed that behavior change was prim ary and that once it had occurred, a change in beliefs would likely follow, jlnsko (1967) also encouraged researchers to concentrate on attitude I 24 change following behavior change rather than the reverse, since a causal sequence existed, upon which the dissonance theory has focused. Social support through agreem ent of others was a m ajor way i that dissonance reduction could occur, regardless of how and where the dissonance had arisen. By obtaining social support for some i opinion, the person would add cognitive elements which were conso- j nant with that opinion and thus would reduce the total magnitude of dissonance (Festinger, 1957). Another method of dissonance reduc tion was forced compliance to public pressure prior to an accom panying change in private opinion. It might result from either the threat of punishment for noncompliance or the offer of reward. Once i | it had occurred, there was dissonance between opinion and behavior. The magnitude of the dissonance was related to the amount of reward or punishment present to induce a behavior change and the im por tance of the opinion or behavior involved. Assuming that compliance occurred, the g reater the reward or punishment, the less the l | dissonance. In summary, dissonance from forced compliance could be reduced by lessening the importance of the opinion or behavior, changing the opinion so that it agreed with the public behavior, or magnifying the reward or punishment used to induce the behavior. One of A llportrs (1954) conclusions stated that ’’ contacts that bring knowledge and acquaintance are likely to engender sounder beliefs concerning minority groups, and for this reason contribute to the reduction of prejudice" (p. 268). However, to reduce prejudice, the acquaintance must be meaningful, since there was evidence that a casual contact was more likely to increase rather than to dispel i prejudice. Finally, it was relevant to this study to observe the relation ship between attitudes and stereotypes. While investigating cognitive consistency theories, Pavlos (1972) used a bogus pipeline paradigm to predict change in racial attitudes and stereotypes. He found that only when subjects changed their racial attitudes did they also change their related racial stereotypes. i A program of behavior change With this theoretical base as a background, it would be i 1 appropriate to examine the program used by the DOD to eliminate ■ i j racial discrim ination. Relying partly upon the writings of two behav- ! lior theorists, Dean and Rosen (1955), the DOD accepted the following assumptions: F irst, the widely held belief that prejudice was the only cause of segregation and discrimination was refuted. Dean and Rosen's proposition that "conformity with the practices of segregation and discrim ination is often quite unrelated to the intensity of prejudice in the individuals who conform" (p. 58) was accepted. The authors stated that "within wide lim its, prejudiced persons will accept and participate in a thoroughly mixed and integrated setting if integrated patterns are established and accepted as appropriate by other p a r ticipants in that situation” (pp. 58-59). The authors* second proposition was that "desegregation i that proceeds by firm and decisive steps backed by the responsible authorities is more readily accepted and taken for granted than a i halting desegregation that appears unsure of itself” (p. 70). They found that when clear lines of authority existed, people tended to accept the rules even though their initial response to them was un favorable. This was consistent with Pettigrew*s (1961) findings. He stated that violence had generally occurred in localities where civic leaders gave hints of resisting desegregation. In other words, if j ; people were not given the opportunity to reduce dissonance, there I | was violence. Aronson (1972) stated that ”the sooner the individuals are made to realize that integration is inevitable, the sooner their ! prejudiced attitudes will begin to change” (p. 196). ! i The DOD attacked the problem of discrim ination by two methods. F irst, discrim inatory behavior would be changed through race relations training. Seminars consisting of a heterogeneous mix of no m ore than 25 people were assem bled for a two-day pro gram . Individuals were treated as having equal social status throughout the program --a necessary requisite for change (Allport, 1954). A verbal sem inar contract was established with each p artici i 27 pant to insure active individual participation. Methods of effective communication and b a rrie rs to communication were explored. Film s, tapes, and lectures provided cognitive information to increase aw are ness of the nature of prejudice, institutional racism , and majority and minority cultural values. Discussion of feelings toward outgroups through group dialogue stressed the affective component. Role play- ! ing through experiential activities designed to place majority members in minority positions in a simulated society provided the 1 Vicarious experience” that Allport (1954:488) cited as a tool to overcome preju dice. Reinforcement occurred through the requirem ent that all p e r sonnel attend the Race Relations Education Program periodically. Course content was altered to avoid repetition. Stress was placed on | the need for positive action through individual involvement in over coming institutional racism . Also emphasized was the DOD's contin- j uing commitment to eradicate institutional racism within its own structure through affirm ative action. Briefly, the second portion of DOD’s plan was to reduce institutional discrim ination. It included a wide range of actions such as establishing a Social Actions Office on every installation, continu ous pressure on commanders and supervisors to insure equal tre a t ment of all personnel, swift action against those in authority who resisted, ongoing review of promotion policies, judicial actions and types of.discharges, periodic analysis of the racial clim ate on installations, and frequent statem ents on the continuing policy of equality of treatm ent for all and on the need to eliminate discrim ina tion. Experimental Precedents Despite concern over the effects of prejudice and racism in an institution as large as the DOD, there seem s to be little system - i iatic, rigorous experimental work that is available which m easures the cognitive and affective components of attitudes following exposure to an education program designed to reduce prejudice. Gaines and Hoine (1973) evaluated the effectiveness of a race relations sem inar designed to change prejudicial attitudes. One of the few purely experimental studies in this area of m easurem ent, it yielded some interesting results. The sem inars consisted of open-ended dialogue among 342 black and white servicem en conducted by two minimally trained i j facilitators. Using the Woodmansee Multifactor Racial Attitude I ! Inventory, the study indicated that the treatm ent groups responded in a more equalitarian fashion than nontreatment groups (p <.01). There was a decrease in equalitarian scores from the posttreatm ent test to the delayed posttreatm ent test given three months later, though scores were still statistically significant (p . 05). The authors recommended that the sem inar form at be adopted for race relations education program s. The authors attributed the decline after the three-m onth period to the subjectsr returning to an environment that had fostered their attitudes for a number of years. A possibility exists that the organization had not effectively implemented the required DOD actions aimed at eliminating institutional racism . Similar results of attitude decay were found in two separate experiments conducted by Caldwell (1973) and Kleg (1971). In Caldwell’s study, college students were submitted to a 12-week course on the nature of prejudice. They were pretested, posttested, and delayed posttested, using the Remmers Scale to M easure A tti tudes Toward Defined Groups. One conclusion proposed that signifi cant student attitudinal modification was not sustained beyond three months after completion of a specialized curriculum . Again, there were no accompanying overt policy or procedural changes within the college environment to eliminate institutional racism . The study by Kleg (1971) attempted to determine whether attitude change would occur through a program designed only to pro vide information about prejudice, race, social class, and caste in a group of inner city white high school students. The Remmers Scale was also employed. The findings indicated statistically significant cognitive-affective attitudinal change immediately after treatm ent, but delayed posttesting indicated no change. Kleg concluded that knowledge and attitudes are in constant interaction, that cognition can tem porarily influence ethnic attitudes, but as newly acquired 30 knowledge is forgotten, attitudes begin to revert to their original level. Summary i Prejudice can be viewed as consisting of a psychological and a sociological component. The psychological aspect manifests itself in displaced aggression and the needs of a prejudiced personality, j which may be characterized by rigidity, intolerance to ambiguity, and fear of outgroups, among other traits. The sociological com ponent consists of institutional practices and policies which overtly or covertly support racism and of peer group or societal pressure to conform to prejudiced practices. Attitude is seen as composed of three parts: the cognitive, i the affective, and the behavioral. The DOD has attempted to change patterns of prejudicial behavior through race relations training and through action designed to rid itself of institutional racism . 31 CHAPTER III METHODOLOGY This chapter deals with methodological background and pro cedures involved in the study. The first section presents a discussion of the development of the test instrum ent used for the investigation of I I the cognitive and affective aspects of attitude change. Other sections consider (1) the basic research design and control procedures em ployed, (2) subject selection and classification, (3) pilot study, (4) techniques of data collection, processing, and analysis, (5) methodo logical assumptions, and (6) limitations of the study. i Background of the Present Investigation ! Early in the investigation it was decided that existing test instrum ents designed to m easure attitude change would be unsuitable | I for evaluating the USAF Race Relations Education Program. Three reasons prompted the development of a new m easurement tool. F irst, to m easure the cognitive acquisition of information that was part of the learning outcome objective of the course syllabus, it was neces sary to structure an instrum ent that referred to this information. 32 Second, an instrum ent was needed that would determine alterations in the affective component of attitudes, through statem ents that required the subject to evaluate or express an emotional response about a racial m atter. Third, as there is an interaction between cognitive and affective statem ents, it was necessary to word items not only in a consistent voice but also in a contemporary style. The Rowe Social Attitude Survey (RSAS) evolved out of a pool of 150 statem ents drawn up by a team of race relations special ists who had graduated as qualified instructors from the Defense Race Relations Institute and had practiced in the field. The pool was then reduced to 35 statem ents and 5 questions by the team of special ists. Ten were designed to m easure the affective component, in that they required an evaluative or emotional response. Twenty-five I | statem ents required a cognitive response, in that they sought to i j | m easure knowledge of factual information or degree of belief in a i | stereotype. Five demographic questions perm itted subjects to be i ! classified for later analysis. The demographic classifications were j sex, age, ethnic background, educational level, and geographic origin. Wording of statem ents often followed examples in other attitude m easures found in the literature. C are was used to avoid an "acquiescence response set" (Triandis, 1971:121) by reverse word ing of some of the statem ents. Item sequence was also considered ! 33 important. The instrum ent was designed so that demographic ques tions were asked first, followed by a series of cognitive statem ents, then by a mix of affective and cognitive statem ents. Two statements that were thought to trigger the strongest emotional responses, since they referred to sex and ethnicity, were placed in the last half of the i : test. This procedure followed a recommendation by Horowitz, |Sedlacek, and Brooks (1972), who found that form sequence was important in m easures o f racial attitudes. Use of least specific wording in the beginning of an instrum ent would avoid cueing the subject that his racial attitude was being m easured. Finally, in following Adorno et al. 's (1950) suggestion, the negatively worded ! item s were phrased so that they expressed subtle hostility without seeming to offend the democratic values which most prejudiced people feel they must maintain. The RSAS was sent to Headquarters USAF for approval and j i I | evaluation by another group of race relations specialists. Several t i J suggested changes were incorporated into the instrum ent, such as the elimination of three items based on negative ethnic stereotypes. The final list of 35 statem ents was reviewed by a third group of race relations instructors who helped determine which item s were cogni tive and which were affective within the definitions expressed by the investigator. It was verified that each cognitive question related to a course learning outcome. Thus, content validation of the cognitive „ ■ _________ 34 questions was assumed since "expert opinions frequently serve as criteria of validity" (Remmers, Gage, & Rummel, 1965:313). However, of continuing concern to the investigator was the validity of the affective statem ents which are a direct m easure of the i construct of prejudice. It was determined that construct validity could be established for the affective statements if two previously j validated m easures of prejudice were tested concurrently and the i separate scores were sufficiently correlated. If not, the affective statem ents would be eliminated and, instead, the two instrum ents described below would serve as the m easure of prejudice. The first instrument chosen was the Intolerance of Ambiguity Scale (Martin, 1964). This scale was designed to m easure mental rigidity and the inability to perceive gradation, variation, and rela- I tivity. Martin and Westie (1959), in their study of prejudice toward j blacks, randomly selected 429 adult subjects from Indianapolis and gave them a short prognostic scale (Martin, 1964). Of the 429, 41 i ! qualified as tolerant and 59 showed prejudicial scores. These 100 subjects then took a battery of measurement tests which included Adorno's F Scale and the Intolerance of Ambiguity Scale. Scores of the prejudiced subjects correlated at .72 on these two scales, while j the tolerant subjects correlated at . 32. Both scores were significant at the . 05 level. Robinson and Shaver (1969) reported that the Intol erance of Ambiguity Scale bore close resem blance to several items I 35 on the original F Scale of Adorno, M thus suggesting relevant face validity” (p. 322). Regarding the reliability of the Intolerance of Ambiguity Scale, M artin’ 1 ' stated, based on his experiences in admin istering the instrum ent in its final form, ”1 am satisfied that there is a very high reliability quotient, in the neighborhood of .95. ” The second instrum ent selected was Harding and Schuman’s i I Social Problems Questionnaire (Robinson et a l . , 1968). It describes i j discrim inatory incidents that emphasize the presence of strong pressures to support discrim inatory treatm ent toward a minority. i i j The respondent is asked to choose between a course that either j injures the minority or offends a m ajority person. The subject who j does not discrim inate has the capacity to identify m ore strongly with the minority member than with fellow m ajority m em bers (Schuman & 2 Harding, 1963). Harding estim ated the construct validity of the Social Problems Questionnaire, correlated with other m easures of j prejudice used in his study, to be . 80. A split-half reliability of . 93 ! was found in the sample of the adult population of Boston, M assachu setts, used in Harding’s study. This investigator reduced the Social I Problems Questionnaire from 12 stories to 6, for reasons of brevity I and relevance. It was desired to lim it the time required to complete J. M artin, personal communication, September 5, 1975. 2 J. Harding, personal communication, September 10, 1975. 36 the RSAS to 30 minutes, the time recommended by Severy (1974) to maintain subject attention. Situations that were no longer relevant to contemporary society were eliminated. The word ’’ Negro, n where used, was changed to ’’ black, ” and a story relating to school integra tion was altered slightly to indicate that there was no busing avail - i able. i ! j Now that RSAS has been validated and determined reliable, i i j it can be shortened to the first 40 item s, thus eliminating the Intoler- I ance of Ambiguity Scale (items 41-48) and the Social Problems Ques tionnaire (items 49-66). A copy of the RSAS and the answer sheet is contained in appendixes A and B respectively. The Likert Scale was chosen as the response mode for the RSAS for a number of reasons. F irst, it is a rating device designed to reveal both the direction of the individual’s stand on an issue and i the intensity with which he holds it. Each item has five response I i possibilities, and each possibility was given a value by the investi gator. A high value reflected an unprejudiced response, while a low i value indicated the opposite. Since the Likert method provides infor- i mation on a unidimensional scale, opinion statem ents were worded to correspond with this concept. Other reasons for choosing the Likert included the fact that it does not assum e equal intervals between the scale values. This i allows the scale to provide information on the ordering of people’s i ! . - 37 attitudes, but does not indicate how close or far apart different attitudes may be. As it is an ordinal scale, it makes possible the ranking of individuals or groups in term s of the favorableness of their attitudes toward an object. The investigator agreed with Adorno et al. fs (1950) reasons for choosing the Likert for the F Scale, in that it was easier to apply, required fewer item s, and avoided the use of judges that was essential in the Thurston Scale. Also, the Likert was chosen for RSAS since it corresponded to the scale construction used by Harding and Martin in their instrum ents and fit the form at of the widely distributed and inexpensive Air Force i answer sheet (see Appendix B). Research Design and Control Procedures The Solomon Four-Group Design (Campbell & Stanley, 1963: i J 24) was used to test the following null hypotheses: 1. In term s of a pretest and an immediate posttest designed to m easure changes in the cognitive and affective com ponents of attitudes regarding prejudice and racism , attendance at the USAF Race Relations Education Pro gram would not cause a difference between the scores of subjects in the experimental group and the control group, as m easured by the Rowe Social Attitude Survey (RSAS). 2. In term s of an immediate posttest and a delayed posttest 38 designed to m easure changes in 1, above, attendance at the USAF Race Relations Education Program would not cause a difference between the scores of subjects in the experimental group and the control group, as m easured by the RSAS. 3. In term s of the three tests mentioned above, attendance at the USAF Race Relations Education Program would not cause the scores of subjects in the experimental group to differ from the scores of subjects in the control group, as m easured by the RSAS, regardless of their demographic groupings. The four-group research design was modified by the investi- ! gator to accommodate a delayed posttest of groups 3 and 4. The design was as follows: This design was selected on the basis that it "is strong and aesthetic ally satisfying. It has potent controls" (Kerlinger, 1973:339) and is "very useful in studies of attitude change” (Triandis, 1971:146). It is a strong design as the requirem ent for comparison is accomplished between groups 1 and 2, and groups 3 and 4. Random selection of subjects assures equivalence of the groups. History, maturation, I testing, m ortality, instrumentation, regression, and interaction of (Experimental) (Control) (Experimental) (Control) Group 1 R Group 2 R Group 3 R Group 4 R 0» X Oz 0 3 O 4. X Os- 0 7 0 6 O s L 39 selection and maturation (Campbell & Stanley, 1963) are controlled with this design. It overcomes the external invalidity caused by the possible interaction effect that a pretest might have on the treatm ent itself (Solomon, 1949). Also, the design was chosen and then modified so that each respondent took the test only twice to minimize the effect of memory. The independent variable examined in the study was the USAF | Race Relations Education Program, Phase II. Scores on the RSAS (containing M artin’s Intolerance of Ambiguity Scale and Harding and Schuman’s Social Problems Questionnaire) served as the dependent variable. Control procedures for organism ic contamination in the ! ! investigation included random selection of m ilitary subjects, the use | of a large sample, and the clustering of subjects into demographic : groups. The control procedures for environmental contamination | were as follows: Subjects used rooms in the same building at March i ! A ir Force Base; the experimental groups were taught by the same instructor; the investigator was the sole adm inistrator of the instru ment throughout the experiment; and the same form s of the RSAS and answer sheet were used. F urther, the instructions on the front were designed to be self-explanatory so that no other instructions would be required during the test. The anonymous aspect of the RSAS was stressed in the instructions. Also, at every test adm inistration, the 40 investigator verbally emphasized that anonymity was assured, and encouraged respondents to answer how they felt about the statem ents in the RSAS that day. This procedure was followed to build rapport i between the respondent and investigator and to create the im pression that the questioner would not disapprove of any view that was ex pressed. This step was recommended by Cook and Selltiz (1964) as j a way to overcome the tendency to provide the socially acceptable i ! answer. i j | Subject Selection and Classification t i i | Two hundred m ilitary personnel were selected on the basis of their random arriv al, between November and December, 1975, as newly-as signed permanent m em bers of March Air Force Base (AFB). These subjects were divided into four equal groups of 50 each. i j Classification variables used in hypothesis testing were j | ! j sex, age, ethnic background, education level, and geographic origin. j 1 | The demographic classifications were: i t 1. Sex: male vs. female 2. Age: 17-22, 23-26, 27-30, 31-40, 41 and over 3. Ethnic background: white, Mexican American/Chicano, black, Oriental American, American Indian, other 4. Education level: some high school, finished high school, some college, finished a four-year college, finished a-graduate ____________________________________ 41 degree 5. Geographic origin: Northwest, Northeast, Southwest, Southeast, other Classification of subjects into these various demographic f ! cells was necessary due to the requirem ent of insuring anonymity of subjects* test responses. They were instructed not to place their i names or other means of identification on the answer sheets. i F urther, the RSAS would be considered a "direct” technique of attitude assessm ent (Sherif & Sherif, 1969:367), as the respondent was asked to indicate his attitude on a subject. Shaw and Wright (1967) cautioned that the reliability and validity of direct attitude tests could not be assured when used as m easures of individual attitude, * i but only in research designed to m easure group com parisons. C or roborating this use of group m easures, M artin and Westie (1959) suggested that the predictive utility of m easures such as the tolerance of ambiguity is much greater when applied to categories or groups i than when applied to individuals. And last, to help program admin istrato rs determ ine which respondents were most and least affected by race relations training, it was necessary to use demographic groupings. Also, it was desired to compare the results of this study to those of demographic studies mentioned ea rlier in the section of Chapter II on the prejudiced personality. 42 Pilot Study In early November 1975, 30 randomly chosen m ilitary personnel at March AFB were given the RSAS under the identical ' environmental and test adm inistration conditions that would be em - ! ployed during later testing. The demographics of the subjects 1 chosen for this pilot study were comparable to the Base m ilitary i i j population. The p'ilot study was used to determine the average length i i of time required to complete the RSAS, whether misunderstandings in instructions or test items existed, and whether the printing was legible, since photographic reduction was used to print the entire test on one sheet. No discrepancies were noted, and the time for completion ranged from 18 to 27 minutes. | The second purpose of the pilot study was to eliminate items | that did not discrim inate sufficiently. To insure that the test was i internally consistent, it was necessary that each item exhibit a sub- i ! stantial correlation with the total score, or that items elicit different i responses from those whose total score was high and those who scored low (Selltiz, Jahoda, Deutsch, & Cook, 1966). Five items were replaced because answers clustered around one opinion, i whether it was strongly agree/agree, or strongly disagree/disagree. The subjects in the pilot study were not used in any other portion of ; the investigation. Data Collection, Processing, and Analysis The RSAS was adm inistered to all subjects selected for the experiment. Groups 1 and 2 were tested during their inprocessing orientation program . Group 1 was scheduled to attend race relations training not earlier than four weeks hence. This time gap was based on Shaw and Wright’s (1967) recommendation to allow a two- to six- week interval to minimize the effects of memory or attitude in te r action between the tests. Group 2 was withheld from the training, but was scheduled for a second testing so that it would coincide with the completion date of Group l ’s training. Group 1 was given the posttest immediately after completing the USAF Race Relations Education Program. Group 3 took the RSAS immediately after r e ceiving the training, and then three months later. Group 4 received no training but took the RSAS on the same dates as Group 3. The | j delayed posttest was considered an important element in the investi- ! j gation since many investigators who m easure attitude change encour- j age the use of a posttest-delayed posttest design to determine attitude decay (Belasco & T rice, 1969; Kleg, 1971; Triandis, 1971). i After completion of the delayed posttests, demographic data and answers were entered on IBM punch cards. Also entered were code numbers for the type of test (i.e. pretest, posttest, or delayed posttest) and the group designation of the respondent. The investigator assigned w eights of 1 to 5 to each item _____________ 44 response. A value of 5 reflected a tolerant response and a value of 1 indicated a prejudiced answer. For the 61 item s on the RSAS, the score range was 61-305. A split-half reliability corrected with the Spearman-Brown prediction formula (Helmstadter, 1964) was used to determine the internal consistency of the Intolerance of Ambiguity Scale, the Social Problems Questionnaire Scale, the affective item s, and the cognitive item s. A Pearson correlation coefficient was performed on the four sets of item s. Further refinement was established for the 61 item s i by use of factor analytic methods. The first hypothesis was tested by an analysis of variance using unweighted means analysis to determine the effect of treatm ent. A repeated m easures analysis of variance was performed on the posttest and delayed posttest to test the second hypothesis and on the pretest, posttest, and delayed posttest scores of selected demographic sub populations to test the third hypothesis. | Data processing was done at the University Computer Center i at the University of Southern California on an IBM 370, using the Statistical Package for the Social Sciences, Version 6, an IBM 1800 at the University of California, Los Angeles, and an IBM 360 at the i University of California, Riverside, using the Educational Statistics Package. 45 Methodological Assumptions 1. The research design, control methods, subject selection procedures, and data processing techniques used in this study were appropriate to the intent of the investigation. 2. Subjective attitudes could be measured by a quantitative technique resulting in a num erical score that represents the subjects’ opinion. 3. A particular test item and the available responses to the item had the same general meaning for all respondents. 4. The method of determining the reliability and validity of the RSAS was appropriate to the purposes of the investigation. 5. Subjects used in the study were representative of all Air Force personnel who had not completed the USAF Race Relations Education Program, Phase II. 6. The responses of subjects chosen for the study were not affected by earlier exposure to other race relations education program s. 7. Subjects in the experimental and control groups respond ed honestly and did not develop a response or acquies cence set. 8. The time space between adm inistrations of the identical test instrum ent to the same respondents was sufficient _______________________________________ 46 to negate the effects of memory or test interaction. 9. Data were accurately gathered, recorded, stored, and analyzed. 10. The cognitive and affective item s in the RSAS could be interpreted and analyzed separately with negligible interaction in subjects' responses. 11. It was possible to designate which item s in the RSAS were cognitive and which were affective. 12. The data yielded by the study were appropriate for analysis by inferential param etric statistical procedures 13. Results of data analysis were interpreted fairly and / accurately. 14. The conclusions drawn from data analysis followed logically and accurately from the objective analysis and did not reflect experim enter bias. Limitations of the Study Limitations which may have affected the generalizability of the results of the study included the following: 1. Although a total number of 170 subjects from the origi nal 200 was adequate for the analysis procedures em ployed, a larg er sample would have increased the strength of the conclusions. 47 2. The subjects used in the study were representative only of the m ilitary population assigned to March AFB during November and December 1975. 3. The subjects who were placed in the various demo graphic groupings were representative only of the population from which they were drawn. Generalizations to personnel with other demographic variables should be avoided. Demographic cells containing fewer than five subjects were not included in the analysis. i 4. Subjects may have responded to sensitive or personal items in a socially acceptable manner instead of ex pressing their true beliefs or feelings. Subjects may have doubted the anonymous aspect of the tests. ! 5. The historical or contemporary racial clim ate of March AFB may have affected the outcome of the inves tigation. 6. The personalities or the professional competence of the instructors in the USAF Race Relations Education Pro gram , Phase II, at March AFB during December 1975 may not have been representative of the population of USAF race relations instructors. 7. True randomness in selecting the 200 subjects was c u r tailed by the constraint that only those subjects who ! i I 48 were eligible for the training due to no previous attend ance at the USAF Race Relations Education Program , Phase II, were included in the experiment. 8. Conclusions drawn from data analysis were limited to the extent to which assumptions underlying those pro cedures had been satisfied. 49 CHAPTER IV FINDINGS The demographic grouping of the 200 subjects who were I * j adm inistered the first test and the 170 subjects who were available for the adm inistration of the second test is presented in Table 1. Internal consistency was determined for each of the four i | portions of the RSAS using split-half reliability corrected with the i Spearman-Brown prediction formula (Helmstadter, 1964). The i | j cognitive item s showed an £ of . 82; the affective item s, . 69; the Intolerance of Ambiguity, . 71; and the Social Problems Question naire, .77. A Pearson correlation coefficient of the same four sec tions gave the results that are shown in Table 2. Since the correlation between the Social Problems Question naire Scale and the Intolerance of Ambiguity Scale with the affective questions was low, a factor analysis between all 61 test items was performed to investigate whether the two scales were measuring another dimension of prejudice. The factor analysis confirmed this possibility as 8 of the 10 affective item s clustered around two closely related but different factors of prejudice. The factors were labeled # i l I 50 TABLE 1 DEMOGRAPHIC GROUPING OF SUBJECTS BY TYPE OF TEST AND OBSERVATION NUMBER Demographic Groupings Pretest o, o 3 Posttest and Pretest O a. O* Posttest, No Pretest Os Oe Delayed Posttest 0 7 Os Sex Male 90 79 86 73 Female 10 7 14 11 Age 17-22 44 38 40 32 23-26 24 21 21 19 27-30 11 9 12 9 31-40 20 17 21 20 41+ 1 1 6 4 Ethnic background White 69 61 78 65 Mexican American 7 4 3 3 Black 18 17 14 12 Asian American 2 1 3 3 American Indian 1 1 0 0 Other 3 2 2 1 Education level ! Some high school 3 1 3 2 : Completed high school 43 38 39 33 ! Some college 37 31 43 36 ! Completed college 12 11 9 8 j Completed grad, school 5 5 6 5 1 Geographic origin Northwest 6 5 8 7 Northeast 39 34 39 34 Southwest 37 30 31 27 Southeast 17 17 21 16 Other 1 0 1 0 Total within each group 100 86 100 84 The Four-Group Research Design (Experimental) Group 1 R O, X 0* (Control) Group 2 R 0 3 O* (Experimental) Group 3 R X Os 0 7 | (Control) i' i ----------- ------------------------------------ Group 4 R 0 6 0 © ------ 51 TABLE 2 PEARSON CORRELATION COEFFICIENTS OF AFFECTIVE AND COGNITIVE ITEMS, INTOLERANCE OF AMBIGUITY, AND SOCIAL PROBLEMS QUESTIONNAIRE Affective Cognitive Intolerance of Ambiguity Social Problems Affective 1.0000 0. 6310 0. 3808 0. 5050 Cognitive Intolerance of Ambiguity Social Problems 1.0000 0. 4535 1. 0000 0. 5325 0. 4431 1.0000 personal involvement or commitment (items 10, 16, 20, 25) and social awareness (items 18, 24, 32, 39). The items from the two i scales did not cluster around either of these factors. Therefore, despite the fact that the reliability of the Social Problems Question naire Scale was somewhat higher than the affective item s on the i 1 RSAS, it was decided to use the RSAS affective item s and consider i ' i them a m easure of prejudice, while recognizing the need for replica tion due to the moderate r of .69. The sm all number of affective item s (10) may have contributed to the moderate level of reliability. Effect of Participation in a Race Relations Education Program on Affective and Cognitive Scores In testing the first hypothesis, a two-tailed analysis of 52 variance was perform ed on the posttest scores of the 10 affective item s (numbered 10, 16, 17, 18, 20, 24, 25, 30, 32, 39) and the 25 cognitive items (the remaining item s numbered 6 to 40) for the four groups of subjects. Due to the disproportionality of cell frequencies, unweighted means analysis was utilized to calculate the appropriate | F value. The two-way interaction of treatm ent and pretest and the I | main effect of pretest failed to show significant differences for both sets of item s. However, the effect of treatm ent was significant in both cases. Information on posttest scores is presented in tables 3 and 4. TABLE 3 ANALYSIS OF VARIANCE ON POSTTEST SCORES AND MEAN SCORES OF AFFECTIVE ITEMS FOR ALL SUBJECTS Source of Variation SS df MS F P Treatm ent (T) Pretest (P) T X P Residual 586.124 1 586.124 25.6 < .0 0 1 50.866 1 50.866 2.2 NS 0.131 1 0.131 0.006 NS 4166.695 182 22.894 Total 4803.695 185 Mean S.D. Treatm ent with pretest Treatm ent, no pretest No treatm ent with pretest No treatm ent, no pretest 38.6 37.6 35. 1 34.0 3. 8 5.4 4.7 4.9 53 TABLE 4 ANALYSIS OF VARIANCE ON POSTTEST SCORES AND MEAN SCORES OF COGNITIVE ITEMS FOR ALL SUBJECTS Source of gg Variation df MS F P Treatm ent (T) 1560. 742 1 1560. 742 13.57 c . 001 Pretest (P) 27. 297 1 27.297 0. 24 NS T X P 274.591 1 274. 591 2.39 NS Residual 20811.242 182 114. 979 Total 22673.872 185 Mean S.D. Treatm ent with pretest 91.4 11.0 Treatm ent, no pretest 93.0 10.3 No treatm ent with pretest 88.2 9.3 No treatm ent, no pretest 85.0 12.0 Since there was no significance attributable to testing effect, it was unnecessary to use an analysis of covariance. The 10 affec I | tive item s, ranging in scores from 26 to 49, and the 25 cognitive I item s, ranging from 62 to 117, served as the dependent variables. Affective and Cognitive Scores Three Months after Treatm ent and Posttesting The second hypothesis was tested by use of a repeated m easures analysis of variance program on the posttest and delayed i j posttest scores of the subjects in groups 3 and 4 to investigate the effect of time on subjects' attitudes. Adjusting of the population sample of groups 3 and 4 was accomplished by eliminating the post- test scores of the subjects who were not available for the delayed posttest. The results for the affective and cognitive items are p re sented in tables 5 and 6. TABLE 5 REPEATED MEASURES ANALYSIS OF VARIANCE ON POSTTEST AND DELAYED POSTTEST SCORES AND MEAN SCORES OF AFFECTIVE ITEMS FOR ALL SUBJECTS IN GROUPS 3 AND 4 Source of Variation SS df MS F P Between subjects Treatm ent (T) 154.190 1 154.190 3.99 NS Residual 3089. 157 80 38. 614 Within subjects Post/delayed (PD) 170. 126 1 170. 126 19.88 c . 001 T X PD 111.086 1 111. 086 12.98 c . 001 Residual 684.344 80 8. 554 Total 4208. 885 163 Mean S.D. Treatm ent with posttest Treatm ent with delayed posttest No treatm ent with posttest No treatm ent with delayed posttest 37.9 34.2 34.3 34.0 4.8 4.7 4.9 4.8 55 TABLE 6 REPEATED MEASURES ANALYSIS OF VARIANCE ON POSTTEST AND DELAYED POSTTEST SCORES AND MEAN SCORES OF COGNITIVE ITEMS FOR ALL SUBJECTS IN GROUPS 3 AND 4 Source of Variation SS df MS F p Between subjects Treatm ent (T) 1290. 203 1 1290. 203 6.43 NS Residual 16248. 002 81 200. 593 Within subjects Post/delayed (PD) 330.817 1 330. 817 10.02 c . 005 T X PD 324.334 1 324.334 9.82 < . 005 Residual 2675. 500 81 33.006 Total 20868. 856 165 Mean S.D. . Treatm ent with post test 93.7 9.4 Treatm ent with delayed posttest 88.1 9.7 No treatm ent with posttest 85.4 11.5 No treatm ent with delayed posttest 85.3 11.9 Regarding the attitude decay for subjects receiving treatm ent and those receiving none, a repeated m easures analysis of variance was perform ed. Due to first order interactions, it was necessary to split the repeated m easures analysis of variance on the treatm ent dimension. This enabled the examination of the posttest/delayed posttest scores to determ ine the magnitude of attitude decay. The 56 second repeated m easures analysis of variance on posttest/delayed posttest scores showed an .F (1, 83) = 11. 40, p c . 001, on affective item s for subjects receiving treatm ent, and no significant difference for nontreatment subjects, I? (1, 83) = < 1 , NS. For the cognitive item s, posttest/delayed posttest scores of subjects receiving tre a t ment indicated an .F (1, 83) = 6.98, p < . 01, while nontreatment scores were not significant, F _ (1, 83) = < 1 , NS. Effect of Treatm ent on Selected i Demographic Groupings The testing of the third hypothesis required the most exten sive use of statistical tools. The scores of subjects who were not ! present for the second test adm inistration were eliminated from the | sample. A repeated m easures analysis of variance was perform ed |on pretest/posttest and posttest/delayed posttest scores of cognitive I j and affective item s in search of interaction effects for all demographic i j subpopulations containing a cell size of five or more. Due to this i lim itation, several demographic groups required adjustment. F or the variable of ethnicity (subsequently referred to as Race), the Mexican Am erican, Asian American, American Indian, and other categories were too small to analyze, so they were elim i nated. This left black and white subjects for the category Race. For the variable Age, the cells of subjects 23-26 and 27-30 years of age 57 were collapsed to provide a more robust sample size. The 41 and over age group contained four subjects, so it was eliminated. And last, the number of subjects who had not completed high school was two, so this group was removed from the education variable, j The only demographic variable showing interaction was Age j for the pretest/posttest scores of cognitive item s. Though the v ari- i able Race showed no interaction, there was a main effect of race present in all the test scores. Looking at the variable Age first, the results of the pretest/posttest scores of cognitive item s are shown in Table 7. i Due to the second order interaction of Treatm ent by Age by | Pre/Post Scores (I? [2,77] = 6. 61, p c .O l) , a 2 X 3 analysis of i variance was performed on variables Pre/Posttest versus Age for the | Treatm ent and the No Treatm ent groups. There was no significant difference for the No Treatm ent group. However, for Treatm ent, since there was no first order interaction present, the main effects I | of age may be interpreted directly. Significance was found among the i i j cognitive item s of .F (1,165) = 3. 95, p < . 05. i The second demographic variable requiring further analysis was Race. All scores for Race are depicted in tables 8, 9, 10, 11. Since there was no interaction of Race with Pre/Post or Post/ Delayed Post Scores for either the affective item s (tables 8, 10) or cognitive item s (tables 9, 11) the main effect of race may be interpre- j I 58 TABLE 7 REPEATED MEASURES ANALYSIS OF VARIANCE OF PRETEST AND POSTTEST SCORES AND MEAN SCORES OF COGNITIVE ITEMS FOR SUBJECTS IN THE 17-22, 23-30, AND 31-40 YEAR AGE GROUPS Source of Variation SS df MS F p Between subjects Treatm ent (T) 44. 0854 1 44.0854 0.23 NS Age (A) 1558.5051 2 779.2525 4.12 NS T X A 352.1158 2 176.0579 0.93 NS Residual 14538.2519 77 188.8085 Within subjects Pre/Post (PP) 335.4656 1 335.4656 39.09 ^ .0 0 1 T X PP 488. 7239 1 488.7239 56.95 < .001 A X PP 83.4406 2 41.7203 4.86 NS T X A X PP 113.5247 2 56.7623 6.61 *<.01 Residual 660. 7501 77 8. 5811 Total 18174.7501 165 17* Pre -22 Post 23- Pre -30 Post Treatm ent 88.0 90.6 87.6 94.7 No Treatm ent 87.9 87.3 90.6 90.8 31-40 Pre Post 75. 1 85. 6 86.5 85.1 ted directly. Significance was found among the cognitive item s of F (1,72) = 6.97, p c . 01, and F _ (1,72) = 18. 37, p c .0 0 1 , indicating that blacks scored significantly higher than whites for the cognitive item s for all three tests. Significance due to race also was found among the affective item s of _ F (1,71) = 9. 80, p <.. 005, and I? (1,73) = 5.19, p < . 05, which shows that blacks scored higher than whites for i j the affective statem ents for the three test adm inistrations. 1 TABLE 8 REPEATED MEASURES ANALYSIS OF VARIANCE OF PRETEST AND POSTTEST SCORES OF AFFECTIVE ITEMS FOR BLACK AND WHITE SUBJECTS : Source of Variation SS df MS F P Between subjects Treatm ent (T) 40.7977 1 40. 7977 1. 28 NS Race (R) ■ 164.4849 1 164.4849 5.19 c . 05 T X R 12. 9400 1 12. 9400 0. 40 NS Residual 2312.9379 73 31. 6840 ' Within subjects Pre/Post (PP) 53. 5912 1 53.5912 16. 67 c .001 T X PP 73.3431 1 73.3431 22. 82 c . 001 R X PP 14.5758 1 14. 5758 4.53 NS T X R X PP 2. 2950 1 2. 2950 0.71 NS Residual 234.5937 73 3.2136 Total 2909.5593 153 60 TABLE 9 REPEATED MEASURES ANALYSIS OF VARIANCE OF PRETEST AND POSTTEST SCORES OF COGNITIVE ITEMS FOR BLACK AND WHITE SUBJECTS Source of Variation SS df MS F P Between subjects Treatm ent (T) 3. 4080 1 3. 4080 0. 02 NS Race (R) 1181.7128 1 181.7128 6.97 < .0 1 T X R 19.4745 1 19. 4745 0.11 -N S Residual 12205. 5019 72 169. 5208 Within subjects Pre/Post (PP) 85.6877 1 85.6877 7.93 A o H - * T X PP 157. 5487 1 157.5487 14.58 <.001 R X PP 27.4590 1 27.4590 2. 54 NS T X R X PP 4. 3817 1 4.3817 0. 40 NS Residual 777.7501 72 10. 8020 Total 14462.9244 151 61 TABLE 10 REPEATED MEASURES ANALYSIS OF VARIANCE OF POSTTEST AND DELAYED POSTTEST SCORES OF A FFECTIVE ITEMS FOR WHITE AND BLACK SUBJECTS Source of Variation SS df MS F P Between subjects Treatm ent (T) 81. 9968 1 81.9968 2.24 NS Race (R) 358.8203 1 358.8203 9.80 < .0 0 5 T X R 1. 0192 1 1. 0192 0. 02 NS Residual 2598.8694 71 36. 6051 Within subjects Post/Delayed (PD) 66. 6713 1 66. 6713 7.26 < .01 T X PD 36.1841 1 36.1841 3.94 NS R X PD 1. 2194 1 1. 2194 0. 13 NS T X R X PD 2. 4389 1 2.4389 0. 26 NS Residual 651. 5938 71 9.1773 Total 3798.8132 149 62 TABLE 11 REPEATED MEASURES ANALYSIS OF VARIANCE OF POSTTEST AND DELAYED POSTTEST SCORES OF COGNITIVE ITEMS FOR WHITE AND BLACK SUBJECTS Source of Variation SS df MS F P Between subjects Treatm ent (T) 242.2822 1 242.2822 1. 42 NS Race (R) 3131.4246 1 3131. 4246 18. 37 <c. 001 T X R 28. 3149 1 28. 3149 0.16 NS Residual 12270. 0019 72 170. 4167 Within subjects Post/Delayed (PD) 69. 4737 1 69.4737 1. 99 NS T X PD 41.1587 1 41. 1587 1. 18 NS R X PD 22.1848 1 22.1848 0. 63 NS T X R X PD 50. 9376 1 50. 9376 1. 46 NS Residual 2511.0004 72 34.8750 Total 18366.7788 151 63 CHAPTER V DISCUSSION AND INTERPRETATION Within the lim itations of the sample and method used in this t study, the findings seem to w arrant the following interpretations: 1. In reference to the first null hypothesis, attendance at the USAF Race Relations Education Program, Phase II, causes a significant difference in pretest and posttest scores of all subjects, as m easured by the RSAS. An F ratio of 25. 6 (p . 001) for the affective item s and 13. 57 (p < . 001) for the cognitive item s resulted. Therefore, it appears that the treatm ent has a signifi- j cant influence on subjects, when m easured immediately | after the training, in both the affective and cognitive components of attitude. This led to the rejection of the first null hypothesis. 2. In reference to the second null hypothesis, the decline in affective and cognitive item scores m easured three months after the treatm ent and immediate posttest was t j significant. The results showed an _ F ratio of 8. 40 I i ! 64 (p < . 005) for the affective item s and an F _ ratio of 6.98 (p < . 01) for the cognitive item s. Therefore, the presence of attitude decay during the time interval indicates that subjects1 prejudicial attitudes return to a level sim ilar to that of those subjects who have no treatm ent (see Table 5). However, for the cognitive item s, mean scores of subjects receiving treatm ent rem ain above those of the nontreatment subjects (see Table 6). This fact indicates that there is some reten tion of new information about racism and prejudicial m atters. Since score change is significant, the second null hypothesis was rejected. 3. The testing of the third hypothesis indicated that in te r action among all of the demographic variables was present only for the p retest/posttest scores of the cognitive items for the variable Age. This resulted in an ]? = 3.95 (p < . 05). It appears that older subjects gain m ore cognitive information about prejudice and racism than younger subjects, as shown by the mean score differences in Table 7. This may be due to the possibility that m ature subjects have had more experi ences and therefore are better able to absorb an exten sive amount of cognitive information within a relatively short period of tim e. Regarding the demographic variable Race, scores of black subjects were signifi cantly higher than scores of white subjects on the three tests for both affective and cognitive item s, as indicated in tables 8, 9, 10, 11. However, blacks are likely to be more aware of the type of information included in the cognitive item s, such as the frequent disadvantages of 0 urban renewal, the prices of goods in ghetto neighbor hoods, or the income level of blacks. The observation that blacks score higher than whites on the affective item s should be viewed in light of the nature of the items The wording of the m ajority of the statem ents appears to be racially neutral, in that the response would not be influenced by a subject’s ethnic background. An exam ple is Item 16, which states, ”1 feel better about other ethnic groups since I have had some training in race relations. " However, Item 24 asks the subject to r e spond to the statem ent, "Most of the problems m inori ties have they bring on them selves. ” It is likely that a m inority person would respond m ore readily in a negative manner to the item because of personal experi ences as a member of a minority group. Therefore, the generalizability of the finding that blacks respond in 66 a m ore tolerant manner than whites must be limited by the observations cited above. Due to the presence of interaction of treatm ent with a demographic subpopula tion, the third null hypothesis was rejected. 4. The reliability of the RSAS appears to be adequately demonstrated. Reliability coefficients fall within the acceptable range of r = .82 for the 25 cognitive item s and r = . 69 for the 10 affective item s. Caution must be used, however, in attempting to generalize too broadly about the discrete nature of the two types of item s. As mentioned in the section on methodological assumptions in Chapter III, item responses may derive partly from the subjects' cognitive domain and partly from the affective domain. 5. One of the affective item s should have been eliminated after the pilot test. Responses to Item 30, "When I see a white man and a black woman together, I feel . . . , ” failed to cluster around either of the two factors of the factor analysis. Also, analysis of the statistics of the 61 item s for all 370 respondents showed the standard deviation of Item 30 to be sm aller than for any other item. This may have been caused by respondent indifference or the tendency to answer in a socially acceptable manner. 6. Although it was desired to analyze the effect of tre a t ment on subjects with dem onstrated tolerant and prejudiced personalities, the sample size of the highest and lowest quartiles, as m easured by the affective item scores, was too sm all (N=ll) to make meaningful com parisons. 68 CHAPTER VI SUMMARY, CONCLUSIONS, AND RECOMMENDATIONS Summary A review of the literature revealed that relatively little em pirical research had been done to determ ine the effect of an j education program on the cognitive and affective components of i racial attitudes. However, a large study recently completed by H eadquarters, United States A ir Force attempted to determ ine b e havior changes toward prejudice and racism by the use of descriptive | survey research methods. j Purpose i This study was designed to generate data which would (1) i evaluate the ability of a race relations education program to im part information on prejudice and racism and facilitate changes in preju dicial attitudes; (2) reflect the extent to which the effects of the pro gram tended to persist; (3) indicate if there were differential effects from the race relations education program on groups which varied in age, sex, education level, ethnic background, and geographic origin; (4) aid program adm inistrators in future curriculum decisions by demonstrating which demographic groups were most and least affected by the training. Research methods A pretest/posttest/delayed posttest experim ental-control j group design, using 200 randomly selected Air Force personnel recently assigned to March A ir Force Base, was employed. The subjects were randomly divided into four groups of equal size. The Solomon Four-Group Design was modified to allow for the adm inis tration of a delayed posttest. The test instrum ent was the Rowe i | Social Attitude Survey. It consisted of 5 demographic questions, 10 i t j affective item s, 25 cognitive item s, M artin's Intolerance of Ambigu- j ity Scale, and a shortened version of Harding's Social Problems i I | Questionnaire Scale. The demographic variables were sex, age, ! ethnic background, education level, and geographic origin. A two- way unweighted means analysis of variance was completed on the posttest scores of all groups for the affective item s, which were considered a direct m easure of the construct of prejudice, and for the cognitive questions, which were used as a m easure of the acqui sition of information about prejudice and racism . The analyses were perform ed to m easure treatm ent effect and the presence of testing effect. A repeated m easures analysis of variance was perform ed on the posttest and delayed posttest scores of groups 3 and 4 to investigate attitude decay. Also, a repeated m easures analysis of variance was completed on pretest, posttest, and delayed posttest scores of all groups to analyze the effect of treatm ent and attitude decay 011 selected demographic subpopulations. 1 I Findings | j Within the lim itations of the sample and methods used in the 1 study, the following findings are summarized: 1. A significant effect (F [1,182] = 25. 6, p < . 001) was obtained for treatm ent due to attendance at a race 1 relations education program , as evidenced by the differ ence in scores of the affective item s for the experimental and control groups. A significant effect (F [1,182] = 13. 8, p < c . 001) was obtained for score differences of the cognitive item s. There was no testing effect present i in the investigation. 2. Delayed posttest scores for the affective item s decreased to a level that was not significantly different from the scores of subjects who had no treatm ent. For the cognitive item s, although attitude decay was present, delayed post treatm ent scores remained above the scores of subjects who did not undergo treatm ent. 3. Only the variable Age showed interaction with treatm ent versus type of test among all of the demographic group ings that were tested. Older subjects scored signifi cantly higher on the acquisition of cognitive item s regarding m atters of racism and prejudice. Regarding the variable Race, black subjects were more tolerant in their responses than white subjects. More intimate awareness of and sensitivity to minority economic and social conditions by blacks could account for some of the noted differences. Conclusions On the basis of the analysis of the data the following con clusions were drawn: 1. The curriculum of the race relations training is effective in that it raises subjects' aw areness level on m atters of racial discrim ination and prejudice. 2. The curriculum is equally effective for demographic groups based on sex, ethnicity, education level, and geographic origin, since the statistically m easured effect of treatm ent, with the exception of the variable Age, was the sam e for all subpopulations. 3. The decline in affective scores three months after the training indicates that treatm ent subjects respond at a level sim ilar to that of subjects who have had no tre a t ment. This should not be considered a weakness or failure of the program since its stated objective is behavioral change rath er than change in the affective domain. The less extensive decay of cognitive item scores indicates that the form at presently employed by the A ir Force is useful in causing m ore lasting changes in cognitive scores. 4. Since it is statistically shown that there is no main effect of testing, the tim e lapse of a minimum of four weeks between the pretest and posttest is suitable for minimizing the effect of memory. 5. The use of a repeated questionnaire is appropriate as a means of m easuring attitude change in a quantifiable manner, as long as rigorous environmental and adm in istrative controls are maintained. The Likert Scale lends itself well to group com parisons since it provides ordinal data. The questionnaire approach is useful in analyzing treatm ent effect by demographic subpopula tions. The use of a delayed posttreatm ent test is recommended for investigations that attempt to m easure attitude change since attitude decay in varying degrees 73 is found in the literature on attitude studies as well as in this experiment. Recommendations Based on these results and conclusions, the following recommendations seem to be warranted by this study: 1. It is recommended that attitude m easurem ent in stru ments employing a pretest/treatm ent/posttest design at m ilitary personnel accession points be used. This would allow for the testing of people who had not been exposed to ea rlie r race relations training. 2. In order to investigate attitude decay in the affective and cognitive domains, yearly race relations training ' should be employed, accompanied by pre/post/delayed posttesting of subjects. Appropriate curriculum changes could be made to keep the m aterial from becoming repetitive. 3. In order to evaluate the race relations training in the context of the component theory of attitudes, the results of this investigation should be combined with the results of the USAF study of behavioral changes in racism and prejudice. 4. Since age was one demographic factor which indicated ______________________________________________________________ 74 differences in cognitive acquisition, the curriculum should be evaluated with a view toward increasing its impact on younger subjects. 5. If repeated attitude m easures are employed and memory is not considered an important source of bias, it is recommended that the same subjects be used for p re / post/delayed posttesting. The resulting sim pler design i would provide comparable control of internal and external sources of invalidity provided by the Solomon Four-Group Design, without the extensive use of time i and subjects it demands. i 6. Regarding the development of new attitude measuring i instrum ents, cognitive item s should be based on learn ing outcomes of the course. Affective changes are m ore difficult to assess. It is recommended that affec tive test item s be worded so that they avoid a minority i j bias. If two or m ore scales are combined in order to help establish validity of a new attitude measuring instrum ent, it is recommended that the scale item s be interm ixed, if possible, to avoid a response set by subjects. As tim e and resources perm it, extensive use of pilot testing combined with factor analysis would ! I assist investigators in further development of valid t \ \ I 75 m easurem ent item s. Especially recommended for further study are statem ents within the affective domain. Use of a delayed posttest is encouraged. 7. A larger number of subjects in the pre/posttest group would allow for the analysis of treatm ent effect on subjects with demonstrated tolerant and prejudiced personalities, determined by highest and lowest quartile rankings of pretest scores for affective item s. t i 76 APPENDIXES APPENDIX A THE ROWE SOCIAL ATTITUDE SURVEY THE ROWE SOCIAL ATTITUDE SURVEY THANK YOU FOR PARTICIPATING. IT IS IMPORTANT TO FIND OUT HOW YOU ACTUALLY FEEL ABOUT THE SUBJECTS COVERED IN THIS SURVEY. ALTHOUGH THE SURVEY IS VOL UNTARY, PLEASE MAKE YOUR FEELINGS KNOWN. MOST OF THE ITEMS IN THE SURVEY ASK YOU HOW MUCH YOU AGREE OR DISAGREE WITH A STATEMENT. JUST PUT DOWN THE ANSWER THAT FIRST COMES TO YOUR MIND. DO NOT SPEND TOO MUCH TIME ON ANY ITEM. SELECT ONLY ONE ANSWER TO EACH QUESTION; MARK YOUR ANSWERS ON THE ANSWER SHEET WITH A NO. 2 PENCIL ONLY. PLEASE DO NOT MAKE ANY MARKS ON THE SURVEY ITSELF. YOU WERE REQUESTED TO PARTICIPATE IN THIS SURVEY BASED ONLY ON A RANDOM SELECTION. EVERYONE PARTICIPATING WILL REMAIN ANONYMOUS, DO ML PUT YOUR NAME OR SOCIAL SECURITY NUMBER ON THE ANSWER SHEET. THERE IS NO WAY FOR YOU TO BE IDENTIFIED WITH YOUR ANSWERS, SO PUT DOWN HOW YOU REALLY FEEL ABOUT THINGS TODAY. NOW, PLEASE BEGIN. APPROVED BY HQ USAF/ACMR USAF SCN 75-88 1 . SEX A MALE ' B FEMALE 2. AGE A B C D E 3. W HICH OF THE FOLLOW ING DO YOU CONSIDER YOURSELF? A W H ITE B M EXIC AN A M E R IC A N /C H I CANO C BLACK D O R IE N T A L AM ERICAN E AM ER IC AN IN D IA N F OTHER (W R IT E IN ON ANSWER SH E E T) E D U C ATIO N LE V E L A SOME H IG H SCHOOL B F IN IS H E D H IG H SCHOOL C SOME COLLEGE D F IN IS H E D A FOUR YEAR COLLEGE E F IN IS H E D A GRADUATE DEGREE 5. WHERE D ID YOU F IR S T ENTER A C T IV E DUTY? A NORTHWEST (A L A S , ID A H O , MONT, N D AK, N E B , O R E, S D A K , WASH, WYO) B NORTHEAST (C O N N , D E L , I L L , IN D , IO W A, M A IN E , M A R Y L, M ASS, M ICH M IN N , NHAMP, N J , N Y , O H IO , PENN, R I , V E R , W VA, W IS C ) C SOUTHWEST ( A R IZ , C A L IF , CO LO , HAW, K A N , N E V , NMEX, O K , T X , Ut ) D SOUTHEAST (A L A , A R K , F L A , G A, K E N , L A , M IS S , MO, NC AR , SCAR, TE N N , V A ) E OTHER (W R IT E IN ON ANSWER SH E E T) 6 . URBAN RENEWAL IS A SO LU TIO N TO THE M IN O R IT Y HOUSING SHORTAGE. A STRONGLY AGREE B" AG R EE' . C UNDECIDED D D ISA G R E E E STRONGLY D ISAG REE 7. C H ICANO S FACE MANY OF THE SAME PROBLEMS BLACKS DO. A STRONGLY AGREE B AGREE i C UNDECIDED D D ISAG R EE E STRONGLY D ISAG REE 5UCH GREAT S T R ID E S HAVE BEEN MADE IN EQUAL O PPO R TU N ITY THAT IT WOULD NOT BE HARD TO L IV E IN AM E R IC A TODAY AS A M IN O R IT Y PERSON. A STRONGLY AGREE B AGREE , C UNDECIDED D D ISAG R EE "E STRONGLY D ISAG R EE MOST JEWS ARE GOOD BUSINESSM EN BECAUSE THEY W IL L DO A N Y TH IN G FOR A D O LLA R . A STRONGLY AGREE B AGREE C UNDECIDED D D ISA G R E E E STRONGLY D ISAG R EE 10. I B E L IE V E THAT I WOULD GET PERSONALLY IN VO LVED TO HELP SOLVE A R A C IA L PROBLEM. A STRONGLY AGREE B AGREE C U N D ECIDED D D ISAG R EE E STRONGLY D ISAG REE 1 1 . GENERALLY S P E A K IN G , D IS C R IM IN A T IO N A G A IN S T M IN O R IT IE S IS ON THE D E C L IN E AND IS R E A LLY NOT MUCH OF A PROBLEM ANYMORE. A STRONGLY AGREE B AGREE C UNDECIDED D D ISAG R EE E STRONGLY DISAG R EE 12. BLACKS COULD MOVE OUT OF THE GHETTO IF THEY R EA LLY WANTED TO . A STRONGLY AGREE B AGREE C UNDECIDED D D ISAG R EE E STRONGLY D ISAG R EE 13. A L L SCHOOLS GET MONEY FROM THE S T A T E , SO SCHOOLS IN M IN O R IT Y NEIGHBORHOODS ARE COMPARABLE TO THOSE IN W H ITE NEIGHBORHOODS. A STRONGLY AGREE B AGREE C U N D ECIDED D D ISA G R E E E STRONGLY D ISAG R EE 14. M IN O R IT IE S SHOULD BE S A T IS F IE D W IT H THE PROGRESS THEY HAVE MADE IN THE PAST TEN YEARS. A STRONGLY AGREE B AGREE C UNDECIDED D D ISAG R EE E STRONGLY D ISAG R EE 15. THE CARTOON THE "F R IT O B A N D IT O " S H O U LD N 'T BE D IS T U R B IN G TO C H ICANO S S IN C E IT IS JU S T AN A D V E R T IS E M E N T . A STRONGLY AGREE B AGREE C U N D ECIDED D D ISAG R EE E STRONGLY D ISAG R EE 16. I FE E L BETTER ABOUT OTHER E T H N IC GROUPS S IN C E I HAVE HAD SOME T R A IN IN G IN RACE R E L A T IO N S . A STRONGLY AGREE B AGREE C UNDEC ID ED D D ISAG R EE E STRONGLY D ISAG REE 17. A BLACK DOCTOR HAS THE SAME C A P A B IL IT Y IN P R A C T IC IN G M E D IC IN E AS A W HITE DOCTOR. A STRONGLY AGREE B AGREE C UNDECIDED D D ISAG R EE E STRONGLY D ISAG REE 18. C A L L IN G A T T E N T IO N TO R A C IA L PROBLEMS ONLY MAKES T H IN G S WORSE. A STRONGLY AGREE B AGREE C UNDECIDED D D ISAG REE E STRONGLY D ISAG R EE 19. MOST M IN O R IT Y PEOPLE WOULD RATHER BE ON WELFARE THAN WORK. A STRONGLY AGREE B AGREE C UNDECIDED D DISAG R EE E STRONGLY D ISAG R EE 20. THE T IM E AND MONEY SPENT ON RACE R E LA T IO N S T R A IN IN G IS W ORTHW HILE. A STRONGLY AGREE B AGREE C UNDECIDED D D ISAG R EE E /S T R O N G LY D IS AG R EE 21. TO D A Y , THE AVERAGE BLAC K EARNS ENOUGH MONEY TO L IV E AS W ELL AS H IS W H ITE COUNTERPART. A STRONGLY AGREE B AGREE C UNDECIDED D D IS AG R EE E STRONGLY D ISAG R EE 22. CHICANO S SHOULD SPEAK E N G LIS H A T HOME SO THEY W O N 'T BE SO SLOW IN SCHOOL- ' A STRONGLY AGREE 1 B AGREE C UNDECIDED D D ISAG R EE . E STRONGLY D ISAG R EE 23. IN T E G R A T IO N OF C H ILD R E N OF D IF F E R E N T RACES IN P U B L IC SCHOOLS IS A ID E A , x A STRONGLY AGREE - B AGREE C UNDECIDED D D ISAG R EE E STRONGLY D IS A G R E E , 24. MOST OF THE PROBLEMS M IN O R IT IE S HAVE THEY BR IN G ON TH EM S E LV E S . A STRONGLY AGREE B AGREE C UNDECIDED D D ISAG R EE E STRONGLY D ISAG R EE 25. I AM IN FAVOR OF A C T IV E RECRUITM ENT OF M IN O R IT IE S TO BALANCE EMPLOY MENT. A STRONGLY AGREE B AGREE C UNDECIDED D D ISAG R EE E STRONGLY D ISAG R EE 26. MOST O R IE N T A LS ARE S L Y . A STRONGLY AGREE B AGREE C UNDECIDED D D ISAG R EE E STRONGLY D ISAG R EE 27. BECAUSE OF T H E IR BACKGROUND M E X IC A N AM ERICANS USU ALLY PREFER M IGRANT F IE L D WORK L IK E H A R VESTIN G ORANGF5 - A STRONGLY AGREE B AGREE C UNDECIDED D D IS AG R EE E STRONGLY D ISAG R EE 28. P O L IC E TEND TO OVERLOOK C R IM ES COM M ITTED A G A IN S T BLACKS S IN C E THEY DON T A FFEC T THE W H ITE M A JO R IT Y . A STRONGLY AGREE B AGREE C UNDECIDED D D ISAG R EE E STRONGLY D ISAG R EE 29. R A C IA L D IS C R IM IN A T IO N AND P R E JU D IC E IS S T IL L A PROBLEM IN THE A IR FORCE A STRONGLY AGREE TODAY, B AGREE C UNDECIDED D DISAG REE E STRONGLY DISAG REE 30. WHEN I SEE A W H ITE MAN AND A BLAC K WOMAN TO G ETHER, I F E E L . . . A VERY ANGRY B ANGRY C UNCONCERNED D GLAD E VERY GLAD 31. UNDER THE DEM OCRATIC SYSTEM , EVERYBODY IS F A IR L Y REPRESENTED S IN C E WE A L L HAVE THE R IG H T TO VO TE. A STRONGLY AGREE B AGREE C UNDECIDED D D ISAG R EE E STRONGLY D ISA G R E E 32. THERE IS NOTHING WRONG W ITH U S IN G R A C IA L SLURS AS LONG AS THEY ARE IN JO K E S . A STRONGLY AGREE B AGREE C UNDECIDED D DISAG REE E STRONGLY DISAGREE GOOD I P R E JU D IC E IS OFTEN TAUGHT BY PAR EN TS. A STRONGLY AGREE B AGREE C UNDECIDED D D ISAG REE E STRONGLY D ISAG R EE BLACK POWER MEANS A V IO L E N T R E V O LU T IO N . A STRONGLY AGREE B AGREE C UNDECIDED D DISAG REE E STRONGLY DISAG REE 10 TESTS ARE A GOOD IN D IC A T IO N OF TRUE IN T E L L IG E N C E . A STRONGLY AGREE B AGREE C UNDECIDED D DISAG REE E STRCNGLY D ISAG REE THE CLENCHED F IS T IS A SIG N OF BLACK M IL IT A N C Y . A STRONGLY AGREE B AGREE C UNDECIDED D D ISAG R EE E STRONGLY D ISAG R EE P R IC E S IN GHETTO NEIGHBORHOODS ARE U SUALLY HIG HER THAN THOSE IN W HITE NEIGHBORHOODS. A STRONGLY AGREE B AGREE C UNDECIDED D D ISAG R EE E STRONGLY D ISAG R EE MOST OF THE PROBLEMS OF L IF E ON AN IN D IA N R ESER VATIO N ARE DUE TO NEGLECT BY THE FEDERAL GOVERNMENT. A STRONGLY AGREE B AGREE C UNDECIDED D DISAG REE E STRONGLY D ISAG R EE BLACKS ARE MORE CONCERNED W ITH SEX THAN W H IT E S . A STRONGLY AGREE B AGREE C UNDECIDED: D DISAG R EE E STRONGLY D ISAG R EE S IN C E BLACKS L IK E TO L IV E TO GETHER, HOUSING PROJECTS IN GHETTO AREAS ARE SUCCESSFUL. A STRONGLY AGREE B AGREE C UNDECIDED D DISAG R EE E STRONGLY D ISA G R E E THERE ARE TWO K IN D S OF PEOPLE IN THE WORLD; THE WEAK AND THE STRONG. A STRONGLY AGREE B AGREE C UNDECIDED D D ISA G R E E E STRONGLY D ISA G R E E A, PERSON IS E IT H E R ONE HUNDRED PERCENT AM ERICAN OR HE I S N 'T . A STRONGLY AGREE B AGREE C UNDECIDED D DISAG R EE E STRONGLY D ISAG REE A PERSON E IT H E R KNOWS THE ANSWER TO A Q UESTIO N OR HE DOESN T . A STRONGLY AGREE B AGREE C UNDECIDED D D ISAG R EE E STRONGLY D ISA G R E E THERE ARE TWO K IN D S OF WOMEN; THE PURE AND THE BAD . A STRONGLY AGREE B AGREE C UNDECIDED D D ISAG R EE E STRONGLY D ISA G R E E YOU CAN C L A S S IF Y ALMOST A L L PEOPLE AS E IT H E R HONEST OR CROOKED. A STRONGLY AGREE B AGREE C UNDECIDED D D IS AG R EE E STRONGLY D ISAG R EE F IR S T IM P R E S S IO N S ARE VERY IM PO R TAN T. A STRONGLY AGREE B AGREE C UNDECIDED D D IS AG R EE E STRONGLY D IS AG R EE IT D O E S N 'T TAKE VERY LONG TO F IN D OUT IF YOU CAN TRUST A PERSON. A STRONGLY AGREE B AGREE C UNDECIDED D D ISAG R EE E STRONGLY D ISAG R EE THERE IS ONLY ONE R IG H T WAY TO DO A N Y T H IN G . A STRONGLY AGREE B AGREE C UNDECIDED D D ISAG R EE E STRONGLY D ISAG R EE IN T H IS PART OF THE Q U E S T IO N N A IR E YOU W IL L F IN D A NUMBER OF S T O R IE S , ‘ FOLLOWED BY Q U E S T IO N S . ANSWER A L L THE Q U ESTIO N S IN THE WAY YOU T H IN K IS ' B E S T , IF YOU ARE SURF THAT YOUR ANSWER TO A Q U ESTIO N IS Y E S , THEN MARK A ' ON YOUR ANSWER SH E E T. ! Y E S ‘ y e s § NO 8 0 ; i i HOWEVER, IF YOU T H IN K THAT YOUR ANSWER TO A Q U ESTIO N IS YES BUT YOU HAVE SOME B O U f i l S ABOUT I T , THEN MARK ' B ON YOUR ANSWER S H E E T. I IF YOU S IM P L Y CANNOT MAKE UP YOUR M IN D WHETHER YOUR ANSWER IS " Y E S " O R ~ " n o " THEN MARK ' c ' ON YOUR ANSWER S H E E T. TRY TO USE THE Q U ESTIO N MARK AS L IT T L E AS P O S S IB L E . IF YOU T H IN K THAT YOUR ANSWER TO A Q UESTIO N IS " n o " , BUT YOU HAVE SOME DOUBTS ABOUT I T , THEN MARK d " ON YOUR ANSWER S H EET. AND F IN A L L Y , IF YOU ARE SURE THAT YOUR ANSWER IS NO YOUR ANSWER SH E E T. THERE ARE THREE Q U ESTIO N S AFTER EACH STORY. Q U E S T IO N S . THEN MARK E ON BE SURE TO ANSWER A L L THE AN ELDERLY BLACK ENTERED A CHIC AG O BARBERSHOP. BEFORE HE COULD S IT DOWN, THE HEAD BARBER S A ID : ' 1 ' M SORRY, WE CAN T G IV E YOU A H A IR C U T . IT TAKES S P E C IA L C LIP P E R S AND E X P E R IE N C E W HICH NONE OF OUR BAPBERS H AVE. T H E R E 'S A PLACE ACROSS THE STREET THAT S P E C IA L IZ E S IN H A IR C U TS FOR B LA C K S , AND I* M SURE THEY WOULD BE GLAD TO SERVE YOU. 4 9 . SHOULD THE HEAD BARBER HAVE BEEN W IL L IN G TO G IV E THE BL.ACK A H A IR C U T ? y£s B YES D NO 50, IS IT A L L R IG H T FOR SOME BARBERSHOPS TO REFUSE TO SERVE B LA C KS? 51. WOULD YOU M IN D GOING TO A BARBERSHOP THAT REFUSED TO SERVE B LA C KS? Y E S YES ' ? NO Y E S YES ? NO I K o ‘N O ' ) IN A TOWN IN SOUTHERN C A L IF O R N IA WHERE THERE IS NO B U S IN G FOR IN T E G R A T IO N , M E X IC A N AM ERICAN AND W HITE C H ILD R E N ARE SENT TO D IF F E R E N T P U B L IC SCHOOLS. THE SCHOOL A U T H O R IT IE S SAY TH AT MOST OF THE M E X IC A N AM ERICAN C H ILD R E N READ AND UNDERSTAND E N G L IS H SO POORLY TH A T’ NEARLY A L L OF THEM WOULD HAVE TO BE IN S P E C IA L CLASSES ANYWAY. ALSO THEY REPORT THAT IN TOWNS WHERE M EXIC AN AM ERICAN AND W HITE C H ILD R E N ARE PLACED IN THE SAME SCHOOLS, F IG H T S BETWEEN THE TWO GROUPS ARE VERY FREQUENT. 52. 53. 54. IS IT A GOOD ID E A TO HAVE SEPARATE SCHOOLS FOR M EXIC AN AM ERICAN AND W HITE C H ILD R E N ? Y E l S YES D NO SHOULD THE SCHOOL A U T H O R IT IE S TRY P LA C IN G YES YES M EXIC AN AM ERICAN AND W H ITE C H ILD R E N IN THE SAME SCHOOLS, IN S P IT E OF THE DANGER OF F IG H T S ? IF THE PARENTS OF A W H ITE C H IL D DO NOT WANT H IM A S S IG N E D TO A SCHOOL W ITH A LOT OF M EXIC AN AM ERICAN C H IL D R E N , SHOULD THEY BE A BLE TO TRANSFER H IM TO AN A L L -W H IT E P U B L IC SCHOOL? N O y(:S B YES D NO n 6 IN A LARGE WESTERN C IT Y THERE HAS BEEN FOR MANY YEARS AN INFO RM AL GROUP C A LLED THE BUSINESSM EN S LUNCHEON C LU B . NONE OF THE MEMBERS IS J E W IS H . RECENTLY ONE OF THE MEMBERS PROPOSED THEY ADD T O 'T H E GROUP A JEW WHO WAS HEAD BUYER FOR ONE OF THE LARGER C LO TH IN G STO RES. A M A JO R IT Y OF THE MEM BERS VOTED A G A IN S T T H IS , THE P R E S ID E N T OF THE CLUB S A ID : 1 T H IN K MR. ROTHMAN IS A F IN E PERSON, BUT I DON T T H IN K WE SHOULD ADD H IM TO THE LU N CHEON C LU B . T H IS HAS ALWAYS BEEN A S M A LL, CLO SELY K N IT P R IV A T E GROUP; AND IF WE TOOK IN MR, ROTHMAN WE WOULD SOON NEED TO A D M IT EVERY JE W IS H B U S I NESSMAN IN TOWN WHO WANTED TO J O IN . 55. IF YOU HAD BEEN A MEMBER OF THE C LU B , WOULD YOU HAVE O BJECTED TO THE P R E S ID E N T 'S STATEM ENT? 56. IF THE CLUB DOES A D M IT JE W IS H MEMBERS, SHOULD IT BE A L IT T L E MORE CAREFUL IN P IC K IN G THEM THAN IN P IC K IN G NO N -JEW S? 57. SHOULD B U S IN E S S AND P R O FE S S IO N A L CLUBS A D M IT MEMBERS W ITHO UT P A Y IN G ANY A T T E N T IO N TO WHETHER THE NEW MEMBERS ARE J'E W IS H ? y £ s B YES D NO .n 8 Y E S YES Y E S YES ? n o N O ARTHUR BRADY WAS A BLACK LAWYER IN A TOWN NEAR BOSTON. HE WANTED TO J O IN THE LOCAL T E N N IS CLUB SO HE COULD P LAY ON T H E IR COURTS. HE KNEW TH AT A NEW MEMBER HAD TO BE RECOMMENDED BY AN OLD MEMBER, SO .HE ASKED A F R IE N D OF H IS WHO BELONGED TO THE CLUB IF HE WOULD RECOMMEND H IM . THE F R IE N D S A ID : I M T E R R IB L Y SORRY, BUT 1 CAN T . THE T E N N IS CLUB HAS ALWAYS HAD A P O L IC Y OF NOT A D M IT T IN G B LA C K S . A PROPOSAL TO CHANGE IT WAS BROUGHT UP AT THE LA S T M EM BERSHIP M E E T IN G , BUT IT WAS VOTED DOWN O V E R W H E LM IN G LY ." 53. IS IT A L L R IG H T FOR A P R IV A T E T E N N IS CLUB TO REFUSE TO A D M IT BLAC KS? 59. SHOULD MR. B R A D Y 'S F R IE N D TRY TO GET THE CLUB TO CHANGE IT S RULES SO THAT HE COULD BE A D M IT T E D ? . 60. IS IT A GOOD ID E A TO HAVE SEPARATE T E N N IS C LU B S , BOW LING LE A G U ES, AND SO O N , FOR W H ITES AND BLAC KS? y(:S D NO n 5 Y E S YES ? Y E S y e s ? n o N O A YOUNG JE W IS H MAN GRADUATED FROM A FAMOUS NORTHERN E N G IN E E R IN G C O LLE G E , WHERE HE S P E C IA L IZ E D IN IN D U S T R IA L D E S IG N . AFTER G RADUATION HE A P P L IE D FOR A JOB W ITH A NORTHERN MANUFACTURER, AND THE PERSONNEL MANAGER TOLD H IM : I D H IR E YOU EXCEPT FOR.ONE T H IN G . THE THREE MEN YOU WOULD HAVE TO WORK W ITH MOST C L O S E L Y --T H E PLAN T MANAGER, THE ACCOUNTANT, AND THE HEAD OF THE DRAFT INC. RO Q M --AL.L HAVE STRONG F E E LIN G S ABOUT JEW S. IF YOU WERE GOING TO WORK DOWN IN THE P L A N T / I ' D NEVER G IV E YOUR R E L IG IO N A THO UG HT; BUT AT THE LE V E L OF .TOP MANAGEMENT ANY SER IO U S F R IC T IO N CAN R U IN AN O R G A N IZ A T IO N . 61. WAS IT A L L R IG H T FOR THE PERSONNEL MANAGER Y ^S YES ^ NO n 5 TO REFUSE TO H IR E THE JE W IS H ENG IN EER TO A V O ID F R IC T IO N W ITH THE OTHER EMPLOYEES? 62. SHOULD THE PERSONNEL MANAGER HAVE ASKED THE OTHER MEN HOW THEY WOULD FE E L ABOUT Y E S YES ? NO N O WORKING W ITH A JE W IS H E N G IN E E R / AND THEN . MADE H IS D E C IS IO N BASED ON T H E IR W IS H E S ? 63. SHOULD EMPLOYERS H IR E MEN FOR TOP MANAGE MENT JO BS W ITHOUT P A Y IN G ANY A T T E N T IO N TO Y E S YES ? NO N O WHETHER THEY ARE C H R IS T IA N OR JE W IS H ? IN A LARGE FACTORY JU S T O U T S ID E C IN C IN N A T I, O H IO , THERE WAS NEED FOR A NEW FOREMAN IN ONE DEPARTM ENT. BOTH A BLACK AND A W H ITE MAN WERE E L IG IB L E FOR THE JOB IN TERMS OF T R A IN IN G , BUT THE BLACK HAD BEEN WORKING IN THAT D E PARTMENT FOR A LONGER T IM E . AFTER SERIO US THOUGHT, THE PERSONNEL D E PAR T MENT P IC K E D THE W H ITE MAN FOR THE J O B , BUT ALSO SAW TO IT THAT THE BLACK R E C E IV E D AN EQUAL R A IS E IN S A L A R Y , THOUGH HE D ID NOT GET A PROMOTION IN D U T IE S . THE PERSONNEL MANAGER E X P L A IN E D TO THE BLACK THAT S IN C E MANY OF THE WORKERS WERE FROM THE SOUTH AND NOT USED TO H A V IN G A BLACK OVER THEM , IT M IG H T S E R IO U S LY DAMAGE THE HARMONY AND P R O D U C T IV IT Y OF THE PLANT IF A BLAC K WERE MADE FOREMEN. A B C 64. WAS IT WRONG FOR THE PERSONNEL MANAGER TO Y E S YES ? P IC K THE W HITE MAN FOR THE JOB OF FOREMAN, ASSUM ING THE REASONS FOR H IS D E C IS IO N WERE R E A LLY THE ONES HE GAVE? 65. SHOULD THE BLACK ACCEPT THE R A IS E IN SALARY AND NOT ATTEM PT TO BECOME FOREMEN? Y E S YES ? 6 6 . IF YOU WORKED IN A FA C TO R Y, WOULD YOU OBJECT TO H A V IN G A BLACK PUT OVER YOU AS A FOREMAN? Y E S YES ? NO n [) NO N O NO N O APPENDIX B ANSWER SHEET 00 A B C O E 1 I I i i I I i i i i A B C D E 2 i i I : I I i i i i A B C D E 3ii i i i i i i i i ABODE ABODE 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 X I 2 3 u. o £ 2 4 < Q 2 5 2 6 2 7 2 8 2 9 ABODE ABODE ABODE ABODE A B O D E ABODE ABODE ABODE ABODE ABODE ABODE ABODE ABODE A B O D E ABODE ABODE ABODE ABODE A B O D E ABODE ABODE ABODE A B O D E A B O D E A B O D E 31 3 2 3 3 3 4 3 5 3 6 3 7 3 8 3 9 4 0 41 4 2 4 3 4 4 ;; ii ii h ii A B O D E 45 ii ii ii ii ii A B O D E A B O D E A B O D E A B O D E A B O D E A B O D E A B O D E A B O D E A B O D E A B O D E A B O D E A B O D E A B C D E 61 ii ii ii ii ii A B O D E 62 ii ii ii ii ii A B O D E 63 ii ii ii ii ii A B O D E 64! A B O D E 65 ii ii ii ii ii A B O D E 66 ii ii ii ii ii A B O D E 67 i A B O D E 68 A B O D E 69 ii ii ii ii ii A B O D E 70 ii ii ii ii ii A B O D E A B O D E A B O D E 71 7 2 7 3 ii II II II II a b o d e 7 4 II II II II II A B O D E 7 5 II II II II II 91 9 2 9 3 9 4 9 5 9 6 9 7 9 8 9 9 100 101 102 103 1 0 4 105! A B C D E :: II II II ;; .. . I. .. • • .. A B c D E jj :: II II j; :: ii II :: II A B c D E •• «• .. •• • • , v f a ,, •• »• • • • • A B c D E .. .. .. • « .. A B c D E A B c D E A 8 c D E II II A B c D E ii ii II II A B c D E j; II ;; II II ;; A B c D E :: ;; II II A B c D E A B O D E A B C D E A B O D E A B O D E BE SURE YOUR MARKS ARE HEAVY A N D BLACK. ERASE COMPLETELY A N Y A N SW ER YOU W ISH T O CHANGE. A B O D E A B O D E A B O D E A B C D E 30 ii ii AF PRT 15 DEC. 51 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 A B O D E A B O D E A B O D E A B O D E A B O D E A B O D E A B O D E A B O D E A B O D E A B O D E A B O D E A B O D E A B O D E A B O D E 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 A B O D E A B O D E A B C D E A B c D E II II II II II :: :i n :: :: A B c D E ,, H ii A B c D E A B c D E ii ;; ii II II A B c D E JJ ii ii ii ii •* •• m m " A B c D E A B c D E a. • « aa aa aa II {; II II II A B c D E II II II II II A B c D E A B c D E ;i II II II II :: :: :: :: :: A B c D E " :: " 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 A B O D E A B O D E A B O D E A B O D E A B O D E A B O D E A B O D E A B O D E A B O D E A B O D E A B O D E A B O D E A B O D E A B O D E 121 122 12 3 12 4 125 126 127 12 8 129 130 131 132 133 1 34;; A 13511 A B C D ' E. ii ii ii ii ii A B c D E A B .c r D E J J ii * J ii ii •« • • i A B c- D E • a aa a. a. •• *• •• *• " - " w •• A B c D E A B c D E ii ii II A B c D E ,, ,a ,, ,, II II :: II A B c D E :: II :: II A B c D E II II II :: A B c D E ii ii ii ii A B c D E A r-B c D -E II i " II .. II a. A ■ B c D E ii ii A ; b c D E i ” i “ ;; II II A 1 B . c D E u . A B O D E 136 A B O D E 137! A B O D E 138 A B O D E 139! A B O D E 140! A B O D E 141 ABODE 142 ABODE 143 ABODE 144! ABODE 145 A B O D E 146 ii ii ii ii ii A B O D E 147I I I I I I I I I I A B O D E 14811 II II II II A B O D E 149 A B O D E 150 81 C 43— 16— 6 9 4 2 3 -3 GPO : 1967 0 - 2 4 9 - 5 2 5 REFERENCES REFERENCES Adorno, T. W. , Frenkel-Brunswik, E. , Levinson, D. J. , & Sanford, R. N. The authoritarian personality. New York: H arper, 1950. Allport, C. W. Attitudes. In C. Murchinson (E d .), Handbook of social psychology. W orcester, Mass. : Clark University Press, 1935. Allport, G. W. The nature of prejudice. Reading, M ass. : Addison- Wesley, 1954. Allport, G. W. , & K ram er, B. M. Some roots of prejudice. Journal of Psychology, 1946, 22, 9-39. I Aronson, E. The social animal. San Francisco: W. H. Freem an, j 1972. i Belasco, J. A. , & T rice, H. M. The assessm ent of change in training and therapy. New York: McGraw-Hill, 1969. Bern, D. J. Beliefs, attitudes, and human affairs. Belmont, Ca. : Brooks/Cole, 1970. Berger, L. , Leibly, J. , & Meunier, B. An attitude survey of college freshmen and seniors: Preference toward seven nationali- ties. Psychological R eports, 1973, 33, 837-838. Bettelheim, B. , & Janowitz, M. Social change and prejudice, j including dynamics of prejudice. New York: F ree Press of I Glencoe, 1964. i Budner, S. Intolerance of ambiguity as a personality variable. Journal of Personality, 1962, 30, 29-50. Caldwell, F. L. Change among selected college student groups in j aw areness regarding the nature of prejudice. Unpublished doctoral dissertation, University of Southern California, 1973. Campbell, D. T. , & Stanley, J. C. Experimental and quasi - experimental designs for research. Chicago: Rand McNally, 1963. Cohen, A. R. Attitude change and social influence. New York: Basic Books, 1964. Cook, S. W. , & Selltiz, C. A m ultiple-indicator approach to attitude m easurem ent. Psychological Bulletin, 1964, 62(1), 36-55. Dean, J. P. , & Rosen, A. A manual of intergroup relations. Chicago: University of Chicago Press, 1955. Dobb, L. W. The behavior of attitudes. Psychological Review, 1947, 54, 135-156. | DuBois, R. D. , & Li, M. Reducing social tension and conflict through the group conversation method. New York: Association Press, 1971. Fendrich, J. M. Perceived reference group support: Racial attitudes and overt behavior. American Sociological Review, 1967, 32, 960-970. Festinger, L. A theory of cognitive dissonance. Evanston, 111.: Row, Peterson, 1957. Festinger, L. Behavioral support for opinion change. Public Opinion Q uarterly, 1964, _28, 404-417. Fishbein, M. Attitude and the prediction of behavior. In M. Fish- bein (E d .), Readings in attitude theory and m easurem ent. New York: John W. Wiley & Sons, 1967. Frenkel-Brunswik, E. A study of prejudice in children. Human Relations, 1948, 1, 295-306. Frenkel-Brunswik, E. Intolerance of ambiguity as an emotional and perceptual personality variable. Journal of Personality, 1949, 18, 108-143. 87 From m , E. Escape from freedom . New York: Holt, Rinehart & Winston, 1941. Gaines, R. N ., & Hoine, H. An evaluation of a race relations sem inar. Millington, T en n .: Naval Technical Training Command, April 1973. Gordon, L. V. Value correlates of student attitudes on social issues: A multination study. Journal of Applied Psychology, 1972, 56, 305-311. Greely, A. , & Sheatsley, P. B. Attitudes toward desegregation. Chicago, 1 1 1 . : National Opinion Research C enter, 1971. Harding, J. , Kutner, B. , Proshansky, H. , & Chein, I. Prejudice and ethnic relations. In G. Lindzey (Ed. ), Handbook of social psychology (Vol. 2). Reading, M a ss.: Addison-Wesley, 1954. H elm stadter, G. C. Principles of psychological m easurem ent. New York: Appleton-Century-Crofts, 1964. Horowitz, J. L. , Sedlacek, W. E. , & Brooks, G. C. , Jr. Related m easures in racial attitude m easurem ent. Research report No. 8-72. College Park, M d.: University of Maryland, Cultural Study C enter, 1972. Insko, C. A. Theories of attitude change. New York: Appleton- j Century-Crofts, 1967. i 'K erlinger, F. N. Foundations of behavioral research (2nd e d .). New York: Holt, Rinehart & Winston, 1973. Kleg, M. Attitudinal change in white students after instruction in an ethnic relations unit. Paper presented to the American Educational Research Association Annual Meeting, New York, February 1971. Krech, D. , & Crutchfield, R. S. Theory and problems of social psychology. New York: McGraw-Hill, 1948. Malec, M. A. Attitude change. Chicago: Markham Publishing Co. , 1971. M artin, J. G. The tolerant personality. Detroit: Wayne State University Press, 1964. 88 M artin, J. G. , & W estie, F. R. The tolerant personality. American Sociological Review, 1959, 24, 521-528. McGuire, W. J. The nature of attitudes and attitude change. In G. Lindzey & E. Aronson (E d s.), Handbook of social psychology (Vol. 3, 2nd e d .). Reading, M a ss.: Addison- Wesley, 1969. M iller, D. C. Handbook of research design and social m easurem ent. New York: David McKay, 1964. O’Connor, P. Ethnocentrism, intolerance of ambiguity, and abstract reasoning ability. Journal of Abnormal and Social Psy chology, 1952, 47, 526-530. Pavlos, A. J. Racial attitude and stereotype change with bogus pipeline paradigm. Proceedings of the Annual Convention of the American Psychological A ssociation, 1972, 7 _ , 291 -292. Pecorella, P. A. Predictors of race discrim ination in the Navy. Ann Arbor: University of Michigan, Institute for Social Research 1975. Pettigrew, T. Regional differences in anti-Negro prejudice. Journal of Abnormal and Social Psychology, 1959, 59, 28-36. Pettigrew, T. Social psychology and desegregation research. American Psychologist, 1961, 16, 105-112. Rem m ers, H. H. , Gage, N. L. , & Rummel, J. F. A practical introduction to m easurem ent and evaluation (2nd e d .). | New York: Harper & Row, 1965. | Rim, Y. Values and attitudes. Personality, 1970, 1(3), 243-250. j | Robinson, J. P. , Rusk, J. G. , & Head, K. B. M easures of political I attitudes. Ann Arbor: University of Michigan, 1968. Robinson, J. P. , & Shaver, P. R. M easures of social psychological attitudes. Ann Arbor: University of Michigan, 1969, Rogers, C. R. On becoming a person. Boston: Houghton Mifflin, 1961. Rokeach, M. The role of values in public opinion research. Public Opinion Q uarterly, 1968, 32(4), 547-559. 89 Schuman, H. , & Harding, J. Sympathetic identification with the underdog. Public Opinion Q uarterly, 1963, 27, 230-241. Selltiz, C. , Jahoda, M ., Deutsch, M ., & Cook, S. W. Attitude methodology. In M. Jahoda & N. W arren (E d s.), Attitudes. Baltimore, M d.: Penguin Books, 1966. Selznick, G. J ., & Steinberg, S. The tenacity of prejudice: Anti sem itism in contem porary A m erica. New York: H arper & Row, 1969. ISevery, L. J. Procedures and issues in the m easurem ent of attitudes. I TM report 30. Princeton, N.J.: Educational Testing ; Service, 1974. i Shaw, M. E. , & Wright, J. M. Scales for the m easurem ent of attitudes. New York: McGraw-Hill, 1967. Sherif, E. W. , Sherif, M. , & Nebergall, R. E. Attitude and attitude change: The social judgement-involvement approach. | Philadelphia: Saunders7 1965. iSherif, M. , & Sherif, C. W. Social psychology. New York: Harper & Row, 1969. Solomon, R. L. An extension of control group design. Psychological Bulletin, 1949, 46, 137-150. | Stewart, D. , & Hoult, T. A social-psychological theory of the i authoritarian personality. American Journal of Sociology, i 1959, 65, 274-279. i j Taft, R. Intolerance of ambiguity and ethnocentrism . Journal of I Consulting Psychology, 1956, 20(2), 153-154. j Tittle, C. R. Attitude m easurem ent and prediction of behavior: An i evaluation of five m easuring techniques. Unpublished | doctoral dissertation, University of Texas, Austin, 1965. iTriandis, H. C. Attitude and attitude change. New York: John Wiley & Sons, 1971. U. S. A ir Force. United States A ir Force race relations education program , phase II, course syllabus. [Patrick Air Force Base, Florida], n. d. 90 U.S. Department of Defense. Report of the task force on the adm inistration of m ilitary justice in the arm ed forces. Washington, D.C. : Author, 1972. Zimbardo, P. G. , & Ebbesen, E. B . Influencing attitudes and changing behavior. Reading, M ass.: Addison-Wesley, 19707 91

Linked assets

University of Southern California Dissertations and Theses

Conceptually similar

PDF

A study of the effects of percentage and locus of burn on body image, self-concept, and social perception for an adult burn population

PDF

A study of the effects of tutoring upon alienation as measured by an original scale

PDF

The effect of a primary grade level interethnic curriculum on racial prejudice

PDF

Christian counselors: A descriptive study of their counseling practices

PDF

Psychoeducational status in relation to duration and quality of dietary treatment in early and later diagnosed and treated phenylketonuric children, adolescents, and young adults

PDF

Jean-Paul Sartre's concept of freedom and its implications for American education

PDF

The effects of sign language on picture naming in four retarded children possessing normal hearing

PDF

Factors which appear to influence the subjective evaluation of a motion picture

PDF

Videotape focused feedback techniques in marathon group therapy: Effects on self-actualization and psychopathology

PDF

A sociological case study of bilingual education and its effects on the schools and the community

PDF

An experimental study of the effect of a listening skills training program in kindergarten on the development of pre-reading skills

PDF

The community health program at Ramathibodi Medical School, Mahidol University, Thailand: Its effectiveness and its influences on attitudes

PDF

An exploratory investigation of women's studies in selected institutions of higher education with emphasis upon the historical background of the status of women and the special needs of women in ...

PDF

A study of the influence of attitude on the outcome of therapy in a captive population

PDF

A survey of the attitudes of school teachers toward problems in education

PDF

Attitudes toward supervision in civil service

PDF

An evaluation of the work study program of Pasadena High School

PDF

A study of the causation and effects of pupil placement and programing in education ability groups at La Colina Junior High School

PDF

A study of counselor training program value structures and their effect on the value system of the counselor

PDF

A survey of educator attitudes and opinions concerning ability grouping in the elementary schools

Asset Metadata

Creator Rowe, Frank Alan (author)

Core Title A study of the effects of a race relations education program on attitudes of racial prejudice

Degree Doctor of Philosophy

Degree Program Education

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

Tag Education, Social Sciences,OAI-PMH Harvest

Language English

Contributor Digitized by ProQuest (provenance)

Permanent Link (DOI) https://doi.org/10.25549/usctheses-c26-449385

Unique identifier UC11246049

Identifier usctheses-c26-449385 (legacy record id)

Legacy Identifier DP24166.pdf

Dmrecord 449385

Document Type Dissertation

Rights Rowe, Frank Alan

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