.1“- , H..-.,«.q.u-..uq..-—~._..a t—..._‘_.-‘-~.~........_..... ..._.-, ~ “—4—- - _-.-..- ... - < ,, . RACE, BELIEF, AND PERCEIVED SOCIAL PRESSURE AS DETERMINANTS OF INTERPERSONAL BEHAVIORAL INTENTIONS I Thesis for the Degree of Pb; D.‘ MICHIGAN STATE UNIVERSITY. LOUIS W03 MEZEI 197.0 ‘--Vo-‘,r :::: .7.,_. IIII IIIIIIIIII IIIIIIIIIIIIII IIIIIIIII 1293 01109 5639 IHESIb LIBRARY " Michigan State University This is to certify that ‘the thesis entitled i RACE. BELIEF, AND PERCENED SOCIAL PRESSURE I AS DETERMINANTS OF INTERPERSONAL BEHAVICRAL INTENTIONS presented by I Louis Lajos Masai has been accepted towards fulfillment of the requirements for Ma—_ degree in _P_3_ xcholoy . -, . 4 I " Aft AZ”Z\ p, L3A’r’fid- A Major professor '_/7 4 , q "4/ //1 W799 Date 0-169 fflmgyg MAR o 9 2000 ABSTRACT RACE. BELIEF. AND PERCEIVED SOCIAL PRESSURE AS DETERMINANTS OF INTERPERSONAL BEHAVIORAL INTENTIONS By Louis Lajos Mezei The study was undertaken in order to reconcile contra- dictory explanations of empirical findings about the relative influ- ence of race and belief similarities on interpersonal behavioral in- tentions. Rokeach, Smith, and Evans (1960) have demonstrated that friendship acceptance depends much more on belief similarity than on racial or ethnic similaritys Rokeach et a1. (1960) have proposed that race prejudice is a special case of belief prejudice, which operates through the mechanism of attributing different values and beliefs to people against whom race rejection is practiced, and then rejecting them.on the basis of belief incongruence. Studies in the literature on the existence and formation of stereotypes illustrate how the belief congruence principle can operate in situations where race prejudice exists due to inferred beliefs. Triandis and Davis (1965) demonstrated that friendship acceptance operates according to the belief congruence principle, but behavioral intentions for small social distance interactions, such as marriage, or dating, are governed more by racial considerations. They have proposed that the degree of intimacy of the interaction will determine whether race or belief is the more important determinant of behavioral intention. Louis Lajos Mezei This study was designed to demonstrate that an extension of the belief congruence principle to include beliefs of the subjects about approval and disapproval of a given interaction by significant others, such as parents and friends, could serve as an explanation of the data reported in the literature. Eighty college student subjects were administered a questionnaire in which they indicated their acceptance or rejection of four stimulus persons; a white communist, a Negro communist, a white anti-communist, a Negro anti-communist for ten social interactions. Perceived social pressure for each be- havioral intention was measured by asking each subject to indicate the degree of approval or disapproval expected from their parents, and friends, if they engaged in each of the interactions with each stimulus person. The experiment was repeated using 79 subjects and variations in beliefs about civil rights legislation, instead of beliefs about communism. Similar results were obtained using both belief variables, although belief about communism was found to be a stronger variable than belief about civil rights legislation. Only for three behavioral intentions: marry, date, and accept as kin by marriage were relative rejections of the stimulus persons stronger on the basis of race than on the basis of belief. For the interactions; accept as intimate friend, fish with, eat with, be commanded by, admire the ideas of, and treat as equal belief prejudice was found to be higher than race prejudice. In the case of race prejudice degree of prejudice is a function of the correlation between prejudice scores and social pressure scores, while for belief rejection the corresponding Louis Lajos Mezei relationship does not exist. Adjustment of the observed race and belief rejection scores for the operation of social pressure, by the application of regression calculations, has resulted in the reversal of the observed relationship between the relative influence of race and belief prejudice for the interactions marry, date, and accept as kin. The conclusion was reached that perceived social pressure to race prejudice can be used as an explanation for the observed race rejection for marital interactions. The joint operation of perceived belief congruence and perceived social pressure explain the data on race and belief rejection reported in the literature. Rokeach M., Smith P. W., Evans R. I. Two kinds of prejudice or one? In Rokeach M. The open and closed mind. Basic Books, New York, 1960 Triandis H. C., Davis E. E. Race and belief as determinants of behavioral intentions. Journal of Personality and Social BEYOhOlOgYe 19659 3. 715’726 RACE, BELIEF, AND PERCEIVED SOCIAL PRESSURE AS DETERMINANTS OF INTERPERSONAL BEHAVIORAL INTENTIONS By Louis Lajos Mezei A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Psychology 1970 {65> Copyright by LOUIS LAJOS MEZEI 1970 ACKNCMLEDGMENTS During the past twelve years a number of psychologists at the University of Toronto, and at Michigan State University have significantly contributed to my professional education and growth, for which I take this opportunity to express not appreciation. The late Professor William Line, among others, has greatly assisted me in sustaining any direction, by illuminating the relation- ship between humnism and science in our numerous discussions. Professor Rokeach has shown to me by example, how you can be a scientist with the highest standards, and at the same time work on exploring significant and difficult human problems. He has assisted in making this dissertation much clearer than it would have been without his many helpful suggestions. I wish to express 1w appreciation to w wife, Tara, for her continuous encouragement and for her assistance in coding and tabulating the data. Dr. Ned Papania and Mr. John Turner have assisted the completion of this study by making subjects available. Dr. Alice Eagly, a member of mythesis Committee at the start, has assisted by suggesting helpful changes in the initial design of the study, such as the suggestion of using more than one belief variables. A member of my thesis Committee Professor T. M. Allen provided a number of helpful suggestions for the design of the study, and the analysis of the data. I I Some of the ideas discussed in a seminar by Professor 'William Faunce about the relationship between social structure and personality structure influenced the systematic approach used. Pro- fessor Eaunce has also been helpful in improving the clarity of the presentation by asking incisive questions about possible sta- tistical interactions between the independent variables. I appreciate the assistance of’Mr. Charles Poland in communicating with the computer for the analysis of the data. I grate- rely acknowledge the assistance of the Michigan State University Computer Laboratory by providing free computer time. III TABLE OF CONTENTS INTRODUCTION....................................................... Systematic orientation............................................. Racial and ethnic prejudice........................................ Phenomenological explanation of prejudice, stereotyping............ Dynamic explanation of prejudice and stereotyping.................. Race prejudice as a special case of belief prejudice............... Stereotyping as belief rejection................................... Selection of marriage partners on the basis of belief congruence... Importance of belief............................................... Type of social interaction and race and belief effects............. Belief congruence and acceptance of support by prejudiced persons.. Social pressure explanation of race rejection...................... HYPOTHESES......................................................... PROCEDURE AND EXPERIMENTAL DESIGN.................................. Subject............................................................ Instruments"...........u......................................... Interpersonal choice scale....................................... Stimulus persons................................................. Perceived Social Pressure........................................ Treatment of the data.............................................. Grouping of rejection scores..................................... PTGJUdiCG SCOI'GSeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee (DO\\J'IC'\IO 10 11 15 18 19 19 22 22 23 23 24 25 26 RESULTS.OOOOOOOOOOOOOOOOO...OOIOOOOOOOOOOOOO0000......OOOOOOOOOOOIOO. 29 Rejection scores and social pressure scores.......................... 29 Prejudice BOOPOSeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee 35 Interactions for which race prejudice is stronger than belief mejwiIOOOOOOOOOO....0.O...00......0.0.0....000......00.000.000.00... no Interactions for which belief prejudice is higher than race prejudice #2 Interaction with no difference between race and belief prejudice..... “3 Comparison of race and belief prejudice scores adjusted for social pressweOOOOOOOOOOOCOO0.....0OOOOOOOOOOOOOOOOOOOCOOOOOOIIOOOOOOOOOOOO M The significance of the adjusted race and belief prejudice scores.... “9 Relationship between social pressure and prejudice scores across all interactionseeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee 52 Social pressure attributed to parents and friends.................... 56 DISCUSSION AND IMPLICATIONS....o....o.o....o..o............o......o.. 57 Effective social pressure and race rejection......................... 59 Comparison of the beliefs used....................................... 60 Prejudice, social pressure to prejudice, and balance theory.......... 61 Intergroup contacts and intergroup attitudes......................... 6“ CONCLUSIONS.......................................................... 66 BIBLIOGRAPHY......................................................... 67 AEPENDIX A: Sample questionnaire for white communist stimulus person 70 APPENDIX B! Appendix Tables 15-19eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee 73 TABLE 10 11 LIST OF TABLES Distribution of subjects in the experimental groups............ 20 Comparison of rejection scores for communist belief stimulus person by sex and r‘c. Of SubJBCtseeeeeeeeeeeeeeeeeeeeeeeeeeeee 21 Mean rejection scores for stimulus persons by subjects, mean social pressure scores of parents and friends, and coefficients of correlation between the rejection and social pressure scores for the two experimental grOUPB................................ 30 Mean rejection scores for communist belief stimulus persons, mean social pressure scores of parents and friends, and correlation of rejection and social pressure scores............ 31 Mean rejection scores for civil rights belief stimulus persons, mean social pressure scores of parents and friends. and correlation of rejection and social pressure scores............ 32 T-test comparisons of rejection scores for four pairs of sti- mulus persons in the communist and civil rights belief exp°r1m9nt3eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee 3“ Race and belief difference scores (RD, B2), race and belief social pressure scores (LSP, BSJ), and the correlation co- efficients between R2 and LSP, and B2 and BSJ in the comet belief studyCCCOO-j-Cz.OOOITUCOCOOC-T-OCOCCCOOCOOCOOCOOQ. 36 Race and belief difference scores (RD, B2), race and belief social pressure scoresQigg, BSJ), and the correlation co- efficients betweenLRD and RSJ, “and B2 and BSJ in the civil rights beliaf BtudYeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee 37 Race and belief difference scores (RD, B2), race and belief social pressure scores RSP, ESP and the correlation co- efficients between R_I_)_ and ESP, and _B2 and BS3 in the two experiment! combinedeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee 38 Mean race and belief prejudice scores after adjustment for social pressure, and t-tests between them for the communist experimenteeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee “6 Mean race and belief prejudice scores after adjustment for social pressure, and grtests between them for the civil rights “xporimfinteeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee “7 VI TABLE 12 13 1h 15 16 17 18 19 Summary of grtests for significance of the social pressure adjusted prejudice score deviations from sero.................. 51 Rank order correlation coefficients between mean prejudice and mean social pressure scores across all interactions........ 53 Rank order correlations between mean prejudice scores and coefficients of correlation between prejudice and social pressure .crOII 811 interactions............................... 53 Standard deviations of rejection and social pressure scores in the communist bOliOf experiment............................. 73 Standard deviations of rejection and social pressure scores in the 01711 rights belief experiment.......................... 7“ Standard deviations of race and belief difference scores, and of social pressure race and belief difference scores in the communist belief experimenteeeeeeeeeeeeeeeeeeeeeeeeeeeeeeee 75 Standard deviations of race and belief difference scores, and of social pressure race and belief difference scores, in the civil rights belief experiment.......................... 76 T-test comparisons of rejection and difference scores for communist and 01V11 rights beliefseeeeeeeeeeeeeeeeeeeeeeeeeeeee 77 VII FIGURE LIST OF FIGURES Scattergram of coefficients of correlation between prejudice scores and social pressure scores toward prejudice for the ten interactions in the two experiments.......................u55 Balance theory model of belief congruence theory between person (2) and other (2)eeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeeéo Balance theory model of belief congruence theory between person (p) and other (2) modified by social pressure...........61 Balance model for race prejudice score.........................62 VIII INTRODUCTION Systematic orientation Social Psychology originated in the same year as a bio- logically oriented psychological system.(McDougall, 1908), and as a sociologically defined system (Ross, 1908). Due to this historical background, three sets of systematic approaches can be employed in the prediction and explanation of human social behavior.‘He can use a purely sociological system, a purely psychological system, or an interacting socio-psychological system.for describing the structure and dynamics of social interaction. Depending on the nature of the problem area studied, one of the three approaches results in the most parsimonious and comprehensive explanation. Because the social scientist attempts to describe and predict relationships between real life variables, there are correspondances, similirities, and occasionally even isomorphism.between sociological and psychological variables. However, serious conceptual and methodological problems arise when theoreticians and experimenters directly apply socio- logical concepts to the psychological level of description, and vice versa, without adjusting for the essentially different meaning of the concepts in the two fields. In real life situations, during the process of formation of social groups the interaction model is the most suitable. Newcomb (1965) studied the acquaintance process in groups of students, who lived together in a university residence. He found that in addi- tion to the development of mutual attraction in pairs of individuals 2 mainly as the result of perception of commonly shared values, another important variable for explaining strong attraction is a mmlti-personal microsociological variable, defined by’a tendency of the died to change evaluation of other group members. In this situation intra-personal psychological and multi-personal micro- sociological interdependency exists, and both sets of variables are necessary for accurate behavior prediction and explanation. In the problem area of determining the relative in- fluence of race congruence and belief congruence on interpersonal choice behavior for different social interactions purely sociological, purely psychological, and the interaction.models can all be useful depending on the degree of social flexibility of the particular social interaction in question. For highly formalised and well defined social interactions the sociological variables by themselves are accurate predictors of the fully socialised -average- individual's behavior. For example Bogardus (1967) reports the strong consistency of the degree of acceptance and rejection of ethnic groups by members of various groups over a forty year period, showing that group variables,can be highly useful concepts.'Where there are non- existent, or non-enforced group norms psychological variables become better predictors of social behavior. In most real life situations both sets of variables interact, and for a comrehensive explanation we need to use all three sets of the variables discussed above. The major focus of the present study is psychological in character because it attempts to reconcile contradictory explanations of findings about the relative influence of race and belief congruence 3 on behavioral intentions in various social situations. Conceptual clarification is one of the tools used, because inappropriate use of sociological variables in the psychological explanation is one of the sources of contradiction. As the result of the conceptual clarification eXperimental techniques are applied to illustrate the operation of perceived social pressure, the psychological equivalent of social norms, in interpersonal behavior choices. Racial and ethnic prejudice Racial and ethnic prejudice has been characterised in the literature as rejection or acceptance of members of different races or ethnic groups as the result of their membership in their groups. A typical definition is presented by'Allport (1958, p. 8): "Prejudice is an avertive or hostile attitude toward a person who belongs to a group, simply because he belongs to that group, and is therefore presumed to have the objectionable qualities ascribed to the group." Bogardus (1928, p. 1h) provides a similar, but more general definition of racial attitudes: "A racial attitude is an established or acquired tendency to act in a social situation involving persons of a race different from one's own. The members of each race possess attitudes friendly and antipathetic toward the members of all other races of whom they have ever heard." Bogardus (1928) has pioneered the application of a social distance scale to study the degree of acceptance and rejection of ethnic group members on the basis of their group membership. The 14 original scale, which has been used with slight modifications in subsequent studies, consisted of a series of interaction items: admission to "close kinship by marriage, to my club as personal chums, to my street as neighbours, to employment in my occupation, to citizenship in my country, as visitors only to my country, and 'would exclude from my country." The findings over a forty year period (Bogardus, 1967) indicate that persistent racial and ethnic preferences exist. In addition to the national pattern persistent regional differences in ethnic prejudice are also present. Bogardus (1928, p. 230) has hinted at a social pressure explanation of race prejudice: ".. in a real sense the nature of racial problems at any particular time depends on public opinion." The sociological studies show that modal race prejudice can be measured and can be explained by the adherence of individuals to social norms. Social norms pro- hibit many intimate interactions with members of different ethnic groups. In psychological terms perceived social pressure against marrying a Negro by a white person is the determining variable in interracial marriage rate. Phenomenological explanation of prejudiceI stereotyping ‘We can translate the findings on rejection of ethnic groups from the sociological to the psychological level of explanation by the application of a phenomenological frame of reference. Katz and Braly (1933) have demonstrated that correspon- ding to the degree of rejection on the social distance scale un- diserable characteristics are assigned to members of rejected groups, and the stereotyped descriptions of ethnic group members are 5 consistent. The findings indicate that prejudiced racial attitudes are more involved than a simple set of conditioned responses to the race name. According to Katz and Braly: "The student is prejudiced against the label Negro, because to him it means a superstitious, ignorant, shiftless person of low social status." The stereotyped perceptions found by Katz and Braly have persisted over time the same way as the social distance patterns persist. Gilbert (1951) has repeated the earlier study to find that Princeton undergraduates have given the same stereotypes for ethnic and racial groups in 1951 as in 1933, although there was a reduction in the degree of extreme prejudice observed. Dynamic explanation ofrprejudice and stereotyping Bettdheim and Janowitch (1950) have reported an interesting dynamic explanation of stereotyping in their study of prejudice in veterans. They reported that Jews were ascribed qualities related to unfair economic competition and control, while to Negroes qualities related to induflgence in primitive, socially unacceptable sex behavior were attributed. The authors point out that both sets of these qualities were used to characterise Jews in Nazi Germany. The findings of the study have also demonstrated a direct relation- ship between degree of perceived frustration and amount of prejudice expressed. The observed facts indicate that stereotyped prejudice serves dynamic needs of the prejudiced person by permitting reduction of anxiety through projection. Unacceptable feelings and desires inside the person are avoided by attributing them to others as a defense against lowered self-esteem. 6 There is a relationship between extreme prejudice and psycho-pathology of the prejudiced person. However the psycho- pathological explanation can not fully explain the incidence of socially supported prejudice observed by Bogardus (1967). Race prejudige as a special case of belief prejudice Rokeach (1960) has proposed that race prejudice actually operates through the mechanism of ascribing unacceptable beliefs to undifferentiated members of discriminated racial groups, and rejecting them on the basis of belief differences, but that there is no discrimination on the basis of race per se. The belief dif- ference explanation is consistent with the findings on the relation- ship between race prejudice and stereotyping discussed above. Rokeach, Smith, and Evans (1960) have reported two studies in which race and belief, and ethnic membership and belief were compared as determinants of friendship acceptance. When subjects were given an opportunity to express degree of acceptance of stimulus persons, they chose more on the basis of belief than on the basis of race or ethnic membership. The hypothesis that: "insofar as psychological processes are involved belief is more important than ethnic or racial membership as a determinants of social discrimination" was confirmed. Contrary to common sense expectations conventionally measured prejudice scores against Negroes did not relate to the observed differences in acceptance and rejection of people. The Rokeach, Smith, and Evans study has used the technique of varying stimulus persons in race and belief and asking subjects, whose «0'0 7 race and belief were known, to indicate degree of acceptance as a friend for stimulus persons whose race and belief were specified. The procedure compared degree of acceptance of pairs of stimulus persons when only race varies, when only belief varies, and when both race and belief vary. Anderson and Cote (1966) using the same stimulus person technique have confirmed that friendship rejection between English and French Canadians operates on a belief, rather than an ethnic basis. Smith, Williams, and Willis (1967) have extended‘the Rokeach et a1. study to identifying stimulus persons by sex, race and belief and to utilization of both white and Negro subjects from the North and the South. Their findings show that belief is more important than race is, which in turn is more important than sex is in interpersonal friendship choices. A notable exception was a Louisiana white sample, where race was by a slight margin more important than belief was. Regional differences are especially strong indicators that perceived social pressure may be operating, to make race rejection stronger than belief rejection. Rokeach and Mezei (1966) have extended the comparison of race and belief to acceptance as coffee partners in a university setting experiment and to acceptance as working partners in a real life employment situation. Instead of stimulus persons described on paper, two white and two Negro confederates provided for both race and belief variation. Both white and Negro applicants for employment in a mental hospital setting tended to choose the two 8 confederates, one white and one Negro, who agreed with them on work related issues. for potential working partners. Similar results ‘were obtained in the coffee partner choice experiment, carried out. on the Michigan State University campus, where even race prejudice, measured by the California E scale, did not influence the choice pattern. Stein, Hardyck, and Smith (1965), Stein (1965). and Stein (1966) have demonstrated that, when belief is manipulated in terms of values of teenagers and therefore the stimulus person is more real, belief is a stronger determinant of friendship and a number of other social distance choices than race is. In addition significant racial prejudice was also observed on a number of interactions. Stein et a1. (1965) concluded that 3 "At present all we can state from our empirical findings is that a belief effect is strong on all items, whereas a race effect occurs on items that appear to involve publicly visible relationships that are 'sensitive' or controversial by prevailing cultural standards." Stereotyping as belief rejection A series of studies carried out by Byrne and his coworkers, (Byrne, 19613 Byrne and Wong, 1962; Byrne and McGraw.196#z and Byrne, 1965) show that even in Texas, attitude similarities are more im- portant in interpersonal friendship choices than race similarities. Rokeach et a1. (1960) have also reported that both in the South and in the North belief differences are more important than race differences in influencing interpersonal acceptance. Byrne and his coworkers have contributed to the understanding 9 of the process of prejudiced behavior by comparing how low and high prejudiced subjects perceive Negroes. Highly prejudiced white subjects, according to Byrne and Wong (1962), assume a dissimilarity of attitudes of Negro strangers to their own. In spite of the initial differences, both high and low prejudiced whites have found Negro stimulus persons with similar attitudes to their own attractive, and Negroes with dissimilar attitudes unattractive. Byrne and McGraw (196u) in a more refined study have con- tinuously manipulated the degree of attitude similarity of the sti- mulus persons. They have found differences in acceptance of Negroes by high and low prejudiced subjects. Highly prejudiced persons accepted a Negro in terms of belief similarities, provided there was complete similarity of attitudes of the stranger about a relatively large number of objects. It appears that to overcome the assumed dissimilarity of attitudes, resulting from prejudice, more similarity of the Negro stranger is needed. The results are consistent with the statement that race rejection operates through the mechanism of assumed belief dissimilarity. Selection of marriage partners on the basis of belief congruence One interesting validation of the belief congruence principle in real life was reported by Newcomb (1963), who worked in the theoretical framework of balance theory and did not attempt to validate the belief congruence principle. He found that girls, who changed from previously conservative to liberal values, married an atypically liberal group of men. Belief congruence resulted in self-selection of a belief validating environment, which was very 10 different from the parental belief patterns. The study provides an example of marriages to socially acceptable people in terms of socio-economic class, as predictable from sociologically de- fined factors, but to an atypically liberal group of men as de- termined by psychological variables. Importance of belief One relevant variable in comparing the relative influence of race and belief on interpersonal choice is importance of the belief..A number of studies have directly compared the effectiveness of beliefs of varying importance. Nelson (1965) has demonstrated that for friendship acceptance the most important determinant is value, the next factor in importance is belief similarity, and the next important variable is race. She found that values accounted for Sufi, beliefs for 39%, and race for 6% in the variance of the friendship choices made. The values used included honesty, sincerity, patience, sociability, and efficiency, the belief variable was communism, and Negrodwhite comparisons were used as the race variable. Triandis (1961) has found in comparing acceptance of Negroes as the race variable, with philosophy of life as the belief variable that social distance was influenced more by race than by belief. Hokeach (1961) has pointed out that the common philosophy belief variable was too sketchy and of insufficient saliency to override the racially stereotyped behavior expected. 11 Martin (196“) has demonstrated that the belief ”same philosophy'is less salient than concrete beliefs are. Martin employed ethnic group stimulus persons of English Canadian against . French or Indian Canadian ethnic origins, and three sets of beliefs, communism, philosophy of life, and religion for determining acceptance by English Canadians on a social distance scale. Communism was found to be the strongest determinant of behavioral intentions, followed by same philosophy, Religion was the weakest belief determinant, probably due to the fact that with the ethnic groups employed religion and group membership are confounded. Type of social interaction and race and belief effects Triandis and Davis (1965) reported that the effectiveness of belief and race congruence in determining behavioral intentions depends on the particular social interaction contemplated. The behavioral intentions were obtained for items in the behavioral difference scales reported by Triandis (196a), and Triandis (1967). Departing from the usual conceptualisation of the social distance scale as unidimensional Triandis and Davis (1965) have identified and used five factors as follows: formal social acceptance vs. formal social rejection (I would admire the ideas of), friendshjaacceptance (I would accept as an intimate friend. I would eat with, I would go fishing with), marital acceptance (I would marry, I would date), social distance (I would exclude ‘ from my neighbourhood, I would accept as a close kin by marriage, I would treat as an equal), and subordination (I would be com- manded by). 12 The pattern of factors obtained depends on the similarities of reponses among several social interactions, which in turn depend on the stimulus person used. For example, marital interaction has been obtained as a separate and independent factor only because sex was one of the attributes of the stimulus persons, and both male and female stimulus persons were presented to male subjects. ‘Without the obvious reponse consistency of males choosing females as marriage partners, the marital interaction would probably appear in the social distance factor. Sex is an important determinant for the emergence of the friendship factor, which could also disappear as an independent factor if only race and belief of the stimulus persons are specified. Insko and Robinson (1967) have found no separation of the items on the friendship and social distance fac- tors, demonstrating that the factors are unstable. Triandis and Davis reported that the social distance and the marriage factors are responded to in terms of race prejudice, the evaluative and formal social rejection are responded to in terms of belief prejudice, while two other factors subordination and friendship rejection are reponded to both in terms of race and belief. Stein (1966) has also reported that white, gentile teen-agers were willing to interact with likedvalued Negroes! in situations such as: "sit next in class", "eat lunch with", "work on a committee with", and "have 33 a 91039 personal friend". In contrast on such interaction items as "invite home for dinner", "live in the same apartment house", "date brother", "have a close 13 .relative marry“, and ”have as a neighbour on the street" white gentile subjects tended to react in racial terms. Stein suggested the variables of social pressure, the presence of others in the situation, and intimacy of the interaction as possible explanations for the differences found. Insko and.Robinson (1967) have presented data which tends to refute the intimacy explanation of race discrimination. They used Stein's technique of belief manipulation to test the effects of the semantic differential evaluative, the friendship, and the social distance factors on behavioral intentions as a function of race and belief similarity. They found that the hypothesis that the more intimate the interaction is the more important will be the race consideration could not be supported..Another interesting finding was that they reported a much higher overall race effect, than found in other studies. The high race effect could be‘thecreault of high social pressure to race prejudice in North Carolina. Willis and Bulatoe (1967) have partially confirmed that marriage acceptance depends to a larger extent on race than belief as compared to friendship acceptance. Using Filipino subjects and Filipino, Chinese, Japanese, American, and Spaniard stimulus persons they found that on the average belief accounts for thirty times as much variance as ethnicity in the friendship ratings, while in the marriage ratings belief is twenty times more power- ful in predicting choice behavior than race. However in absolute terms belief was found to be more powerful than race in both the friendship and marriage ratings. 14 Triandis, Loh, and Levin (1966) have explored the relative effect of the variables: race, belief, type of clothing worn, and quality of spoken English on behavioral intentions. Language was . found to account for 70% of the choice variance on the friendship, evaluation and admiration factors, while on the social distance factor race explained 57% and language accounted for 32% of the variance. The civil rights belief variable employed had a signifi- cant, but small, effect only on two sets of behavioral intentions; evaluation and admiration. Because the language variable probably signified social class differences to the subjects, it can be conceptualised as a belief variable. Belief congruence and acceptance of support by prejudiced persons In the absence of social pressure for increasing race rejection the opposite effect can be observed as the need for belief validation increases, and a person of a different race is available for the purpose. Malof and Lott (1962) have found no significant differences in the acceptance of support by high and low ethnocentric white subjects from a Negro confederate in an.Asch type of experiment. In the experiment a number of white confederates showed consensus on erroneous judgments of the length of lines, while a Negro confederate supported the actual perception of the length by the white subject. Berg (1966) using an ambiguous per- ceptual situation obtained similar results. He reported no relation- ship between the influence of a Negro confederate on autokinetic judgments and prejudice expressed by white subjects on the California § scale, authoritarianism measured by the §_scale, and social distance judgments. 15 Social ppessure egplanation of race rejection Triandis and 13m. (1965) proposed that intimacy of behavior determines the relative influence of race and belief factors in interpersonal choice. Intimate behaviors are responded to in terms of race, while less intimate behaviors are responded to in terms of belief congruence. One major problem with the intimacy explanation arises from the fact that the five behavioral differential scales are not ordered according to degree of intimacy. There are wide variations in degree of intimacy within each scale. For example "I would accept as kin by marriage" and "I would exclude from my neighbourhood" are representing very different degrees of intimacy, but both appear in the social distance factor. In addition there is no compelling reason for translating the observed social distance scale into a psychological intimacy scale. It has been clearly demonstrated that certain social interactions, especially marital intentions, are determined more by race than by belief in the United States. Interestingly, as discussed above (Willis and Bulatao, 1967) in the Philippines belief is more important than race is in determining marital intentions. The fact of regional variations in the relative ins fluence of race and belief in interpersonal choice behavior for the same social interaction has also been pointed out above. The national and regional variations can not be explained by assuming a simple race effect, or by using the intimacy hypothesis. An.explanation of the empirical facts is possible by the extension of the belief congruence principle. Behavioral 16 intentions depend not only on the congruence of beliefs with the person to be chosen for interaction, but also on beliefs about the consequences of the intended inter-racial or inter-ethnic interaction in terms of reference group relationships. Not only congruence of beliefs with the person to be interacted with, but also congruence of beliefs with those of significant reference persons about the interaction need to be considered. One study designed to examine promotion practices in industry (University of Michigan Survey Research Center, 196“) demonstrated that ethnic discrimination by supervisors is related to a perceived bias in higher management. Racial or ethnic discrimination is not necessarily an individual personality, or evaluative variable, but is partially determined by perceived social pressure. Beliefs about social pressure operate only if they are perceived by the subjects as originating from significant people. ‘When subjects do not perceive social pressure to take race into account thewaould choose interaction partners on the basis of belief congruence, provided sufficient information is available for overcoming stereotyped belief perceptions. The present study was undertaken because no study in the literature demonstrates and explores the possible dynamic interplay of perceived social pressure with perceived belief and race congruence in interpersonal choice behavior. The design of the study was directed at providing an explanation of the data in the literature on the relative influence of race and belief on behavioral intentions. 17 In designing the study, a new set of measures to estimate perceived social pressure were devised..All the other variables used have been employed in the studies reviewed above. Two belief variables, co-unism and beliefs about civil rights legislation were employed in order to reduce the risk of confounding a possible statistical in- teraction between race and the belief variables. The selection of two belief’variables was a compromise between the type of studies which use a large number of belief variables and a single social interaction, such as the Rokeach et a1. (1960) study, and studies which employ only one belief variable and a large number of social interactions, such as the Triandis and Davis (1965) study. For the current research ten of the 15 social interactions used by Triandis and Davis (1965) were selected to represent all of the five factors previously isolated. Because a statistical interaction was expected between race, belief, and type of social interaction, in the analysis of the data the magnitudes of race and belief rejection were compared se- parstely for each social interaction. In the present study each social interaction‘was analysed separately as contrasted.with the use of five factors employed by Triandis and Davis (1965) as the unit of analysis. HYPOTHESES 1. There is a statistical interaction between the relative influence of race and belief congruence and the type of social inter- action in the acceptance and rejection of people. The statistical interaction applies to both the absolute degree of rejection of the stimulus persons and to differential treatment of the stimulus persons on the basis of race and belief differences. 2. Belief congruence is a stronger determinant of inter- personal choice intentions than race congruence, provided no strong perceived social pressure operates against the interpersonal choice. 3. For race prejudice to operate, when belief congruence is perceived, perceived social pressure to discriminate on the basis of race has to be stronger than perceived social pressure to discriminate on the basis of belief. h. For belief prejudice the effectiveness of the perceived social pressure, as measured by the correlation between perceived social pressure and size of belief rejection, is uniform.for all interactions. For race prejudice the effectiveness of perceived social pressure is a function of the degree of prejudice expressed. 18 PROCEDURE AND EXPERIMENTAL DESIGN Subjects During the spring and summer of 1967 students in several sections of a large Introductory Psychology class and two smaller summer classes have served as subjects, during regular class time. In order to control for possible systematic effects related to the class sections used, every second subject received the alternative form of the two questionnaires used. In each section one half of the subjects filled out the questionnaire using communism as the belief variable, and one half of the subjects filled out the questionnaire describing a stimulus person with beliefs about civil rights legislation. A total of 166 subjects participated in the study. The data given by seven subjects could not be used due to incomplete questionnaires turned in. Table 1 shows the distribution of the remaining 159 subjects by sex, race, and belief in the experimental groups..A test for possible race and sex differences in the communist experimental group was carried out by performing forty analyses of variance of the rejection scores. Because the civil rights expe- rimental group contained only seven Negro subjects, the preliminary analysis was carried out only for the communist group, which contained 21 Negro subjects. Table 2 presents the rejection scores for the four stimulus persons on the ten interactions used by race and sex of the subjects, and the indication of significant differences in the com- munist experiment. As there were only three significant sex, and four 19 20 Table 1 Distribution of subjects in the experimental groups C o m m u n i s t B e l i e f A ge a iwfin s t F o r GRAND Male Female Total Male Female Total TOTAL White 31 28 59 0 0 O 59 Negro 11 10 21 0 0 O 21 Total #2 38 80 0 O O 80 C i v i~ l R i g h t s B e l i e f A g a i n s t F o r "“*" GRAND Male Female Total Male Female Total TOTAL White 11 11 22 23 27 50 72 Negro 00 00 00 h .3 1L, 7 Total 11 11 22 27 3O 57 79 SAME RACE White Negro f subjects P e r s o n OTHER RACE ANTI-COMMUNIST ANTI-COMMUNIST White Negro S 21 Table 2 COMMUNIST White Negro OTHER RACE rison of rejection scores for communist belief stimulus persons by sex and race 0 S t i m u l u SAME RACE Compa S u b j e c t S u b j e c t S u b j e c t S u b j e c t COMMUNIST action Sex White Negro M Inter- Date Kin Friend M 63 22 21.. 22 68 32 63 23 67 .43 95 3h. 25 33 62 3h. M F Fish with 65 ee 11 with §,p<.05 for race difference §.p<;05 for sex difference, Exclude M Admire M 22 significant race differences out of the possible 40, the data for the two sexes and the two races were combined. Instruments lnterpersonglachoice scale Ten sets of interpersonal behaviors were presented to each subject in a random order, as shown in the questionnaire presented in .Appendix.l. The ten interaction items were chosen to represent the five factors outlined by Triandis (1967). The personal interaction items used were: merry,, date, accept as kin by marriage, accept as an intimate friend, go fishing with, and eat with. The four more ims personal interaction items consisted of be commanded by, admire the ideas of, treat as an equal, and exclude from the neighbourhood, all items in this class indicated more formalised type of social inter- actions. Nine of the interactions were stated in positive or ac- ceptance terms such as "I would marry". The only negatively stated interaction was "I would exclude from my neighbourhood." In order to make all observations comparable the raw data obtained for this interaction were transformed by reversing the scale. The complete scale consisted of ten behavior items used to determine to what extent the subject would interact with a stimulus person, presented at the top of the page. Behavioral intentions were measured by a seven point scale, presented as seven discrete points in a continuum from'would to would not, using a format identical to that employed in semantic differential measurements. {h- e ”"18"! 23 Stimulusipersons In order to evaluate the relative influence of race and belief four stimulus persons were presented to the subjects for in- dicating their behavioral intentions on each of the ten social inter- actions as follows: A WHITE PERSON WHO IS A COMMUNIST A NEGRO WHO IS A CCMMUNIST A NEGRO WHO IS ANTI-COMMUNIST A'WHITE PERSON WHO IS.ANTI-COMMUNIST A second set of stimulus persons were presented to a different group of subjects as follows: A WHITE PERSON WHO IS FOR STRONG CIVIL RIGHTS LEGISLATION A NEGRO'HHO IS FOR STRONG CIVIL RIGHTS LEGISLATION A NEGRO NHO IS.AGAINST STRONG CIVIL RIGHTS LEGISLATION A'WHITE PERSON WHO IS AGAINST STRONG CIVIL RIGHTS LEGISLATION Perceived Social Pressure After the subjects indicated all their behavioral intentions, they were asked to indicate degree of approval or disapproval expected from their parents and their friends if they engaged in each of the social interactions with each stimulus person. Separate judgments were made for parental reactions and the reactions of friends, using the same seven point scale the subjects used for stating their behavi- oral intentions. .After completing the behavioral intention and the social pressure scales subjects were asked to give information about their race, sex, and belief positions on the belief variable used. They were 2h also aked about the race and beliefs of their parents, and about the race and beliefs of most of their friends. Treatment of the data The raw data obtained consists of rejection scores by individuals for a specified stimulus person in a particular social interaction. As there were four stimulus persons and ten interactions a total of forty rejection scores were obtained for each subject. Because on the scale employed seven signifies maximum rejection, and one denotes maximum acceptance, the des- criptive term, rejection score, is used with the understanding that we are dealing with an acceptance-rejection continuum..A mean score of below four is acceptance, while a mean score above four signifies rejection. Corresponding to each of the forty behavioral intention judgments, made by each subject, are perceived social pressure scores obtained for parents and friends, a total of eighty perceived social pressure scores. Seven on the social pressure scale signifies the strongest social pressure toward rejection of the stimulus person, one signifies approval of interacting with the stimulus person, while four is the neutral point. The rejection scale and the perceived social pressure scale were constructed in a similar way in order to make easy for subjects to express their rejection judgments and perceptions of social pressure, by using only one frame of reference. The similarities in the scale structure also facilitate the calcu— lations of coefficients of correlations between the rejection and the social pressure scores. 25 Groupingggf;rejection scores In the absence of major race and sex differences in the rejection scores, combinatim of the data into two groups was judged to be appropriate, one group using communist and the other employing civil rights belief stimulus persons. Before the data could be com- bined transformations were necessary in order to allow for the race Hand belief differences of the subjects, as compared to the race and belief of the stimulus person used. In the communist belief experiment there were no belief differences between the subjects, as all subjects declared them- selves anti-communists. Therefore for combining the rejection scores only the race differences between the subjects needed to be considered. Instead of tabulating the rejection scores for each stimulus person it was more useful to tabulate the scores as follows: Different Belief,Same Race (BZBI) Different Belief, Different Race (B:B:) Same Belief, Different.Race (Bsz) Same Belief, Same Race (B131) For example, to arrive at the rejection score for a stimulus person ‘with the same belief and of the same race as the subject (BfBfi) the rejection scores of the 59 white subjects for the anti-communist white stimulus person were combined with the rejection scores of the 21 Negro subjects for the anti-communist Negro stimulus person. In the civil rights belief group there were 22 white subjects against strong civil rights legislation, while the other 57 subjects were for strong civil rights legislation, and because 26 in addition the sample included 7 Negro subjects, both belief and race adjustments were made in combining the data. The perceived social pressure scores were also grouped, using the same transformations as described above. In practice all that needed to be done was to make sure that, for each individual subject , the social pressure scores accompanied the appropriate rejection. scores into the new groupings. Prejudice scores The rejection scores obtained measure degree of acceptance and rejection of each stimulus person for a given behavioral intention. It is an absolute measure of the degree of rejection expressed. Even when there is acceptance of a stimulus person, differential acceptance as the result of race or belief prejudice is possible. As our major interest is to compare the differential influence of race and belief on prejudice, a more useful set of derived scores can be computed following Rokeach et al. (1960). A race difference socre was computed by obtaining the arithmetic sum of the two pos- sible rejection score differences due to race: (1343‘ - B‘R") + (B‘R‘ - s‘R")........Equation1 The numerical value, the difference between the rejection scores for members of the other race and for members of the subjects' race, is a prejudice score due to race differences. Because the study employed a seven point scale, each of the maximum possible rejection score differential is six, and therefore the race difference scores can range between -12 and +12. A positive value, which is significantly different from zero in- 27 dicates a rejection due to race alone, while a negative value indi- cates acceptance due to the race of the stimulus person. A similar belief prejudice score can be computed using equation 2: (371* -B*R+) + (B‘R‘- B‘R‘)..........equation2 The computed belief difference, or belief prejudice, score has the same properties as the race prejudice score discussed above. Because of the similarity of the structure of the race prejudice and belief prejudice scores they are directly comparable, and are useful for testing of the relative contributions of race and belief to rejection. Using the social pressure scores social pressure race difference and social pressure belief difference scores can be computed by the substitution of the appropriate perceived social pressure difference scores into equations 1 and 2, respectively. The meaning of the social pressure race difference score is analogous to the meaning of the race rejection score, except the social pressure race prejudice score measures the amount of social pressure for acting in a prejudiced manner as the result of race. Similar meaning is attributed to the social pressure belief difference scores. Coefficients of correlation can be computed between the computed prejudice scores and the social pressure difference scores for each interaction. Comparison of the coefficients of correlation for race and belief prejudice can reveal which effective social . pressure is higher in each interaction, the one toward race rejection, or the one toward belief rejection. we can also compare the size of 28 race and belief rejection for each interaction. Computations from.the prejudice scores and the social pressure toward prejudice scores can provide a direct test of the hypothesis that for higher race rejection than belief rejection to take place, higher social pressure toward race prejudice than belief prejudice is needed. 'We can compute adjusted race and belief prejudice scores froa'which the effects of the operation of perceived social pressure have been eliminated by regression calculations. The statistical operations involved are outlined in the results section. The adjusted prejudice scores are the mean race and belief prejudice scores ex- pected under the condition of no perceived social pressure. Direct comparison of the adjusted race and belief prejudice scores for each interaction can provide information about the relative influence of race and belief on interpersonal acceptance under the condition of no perceived social pressure toward prejudice. RESULTS Rejection scores and social pressure scores Table 3 presents a summary of the mean rejection scores and the mean social pressure scores for the four stimulus persons across the ten social interactions. As expected, the largest rejection scores were obtained for the stimulus person with different belief'and of’different race, h.8 and h.” in the communist and civil rights experiments, respec- tively. The next largest rejection was found to be for the stimulus per- son of the same race and different belief, h.“ and 3.7 in the communist and civil rights experiments, indicating a slight degree of rejection of the communist stimulus person and a slight acceptance of the civil rights belief stimulus person. Highest degree of acceptance was indi- cated for the stimulus person of the same race and same belief’as the subjects, 2.6 and 2.” in the two experiments, respectively. The social pressure scores are similar to the rejection scores. The coefficients of correlation between the mean rejection.and mean social pressure scores range from 0.26 to 0.60 for the eight mea- sures reported in Table 3. Rejection scores are related to perceived social pressure scores for all stimulus persons, indicating that both race and belief rejections are related to perceived social pressure. Tables ”land 5 present information for each of the ten social interactions in the two experiments. The data include the mean rejection scores for the four stimulus persons, and the corresponding mean social pressure scores attributed to parents and friends. Tables 29 30 Table 3 Mean rejection scores for stimulus persons by subjects, mean social pressure scores of parents and friends, and coefficients of correlation between the rejection and social pressure scores for the two experimental groups Group Per- 512+ B‘R" B " B‘Ri' i a I .1; 2i .1; 2i. 1'. Commu— S 4.h “.8 3.7 2.6 nist P 5.0 .u2** 5.6 .41" 4.5 .33" 2.5 .47“ F 4.6 .49“ 5.1 A9" 14.1 A6" 2.6 .60** Civil S 3.7 u.u 3.6 2.4 rights P 3.“ .38** 4.9 .38** “.7 .39** 3.3 .26* F 3.8 .50** h.8 4.2 .n9** 2.9 .38** .49** *p*Hx or X' e “I - beH! ...................... Equation “ Equation “ is the formula used to calculate the adjusted rejection 8601’“, in order to obtain the value of the mean rejection score “5 expected under the condition of no perceived social pressure. This score is the same score we would obtain if we adjust the scores for each individual for the operation of perceived social pressure, and then compute the mean adjusted score for each interaction. Before carrying out tests of significance between the race and belief prejudice scores the standard deviations of each were adjusted for the reduction in variability accounted by the linear relationship of predicting prejudice scores from the perceived social pressure scores. As the amount of variance explained by the relation- ship between the tam sets of scores is 312, the adjusted variance is: s'2 w s2(1 - r2 )........................ Equation 5 Using the adjusted prejudice scores and the adjusted standard deviations B—tests were carried out between the mean race and mean belief prejudice scores for each interaction. The degrees of freedom used were reduced by one, in order to allow for the additional restraint imposed by the linear regression. Table 10 presents the adjusted mean race prejudice scores, the adjusted mean belief prejudice scores, the standard deviations of the adjusted scores, and the results of _t-tests for testing the significance of the differences betwun the adjusted mean race and belief prejudice scores for each of the ten interactions in the communist belief experiment. Table 11 presents identical information for the civil rights experiment. In both experiments nine out of the ten interactions resulted in numerically higher adjusted belief rejection as compared to the adjusted race rejection scores. The one interaction for which 0 T 1 Interaction harry Date Kin Friend Fish with Eat with Commended by Admire Equal Exclude Nean race and belief prejudice scores after adjustment for social pressure, “6 Table 10 and Brtests between them for the communist experiment Mean race prejudice 1.3 1.0 0.8 0.3 0.8 0.6 0.1 0.5 0.2 Mean belief Standard prejudice 1.9 1.7 1.5 2.5 1.2 0.7 1.1 2.7 1.3 .0.7 *p<,05, **p<.01 (one tailed tests) deviation 2.“ 3.0 2.“ 1.5 1.5 1.5 1.1 2.1 3.0 Standard deviation 2.3 2.7 2.5 1.0 1.0 2.0 1.2 1.2 2.8 1.3 3': 1.57 1.83* 1.69* 9.23100! 1.9“* 0.18 3e68** 1“.72** 1e96* 1.26 “8 race prejudice is numerically higher than belief prejudice is exclude from my neighbourhood for which all prejudice scores are small, 0.2 and 0.6 for race, and -0.7 and 0.6 for belief prejudice. All stimulus persons are accepted as neighbours in absolute terms. The most dramatic result of adjusting for the operation of social pressure takes place in the interactions marry, date, and acgept as kin, for which all unadjusted race prejudice scores are numerically higher than the corresponding belief prejudice scores, as outlined in Tables 7 and 8. Elimination of the effect of perceived social pressure results in a reversal of the relationship: all belief prejudice scores for the three interactions are numerically higher than the race prejudice scores, as reference to Tables 10 and 11 indicates. Without adjustment for social pressure the Baggy, 8222: and accept as kin interactions resulted in significantly higher race than belief rejection in the communist belief experiment, as outlined in Table 7. Inspection of the results of the Bytests for the communist belief eXperiment in Table 10 shows that for the interactions B332 and accept as kin belief prejudice became significantly higher than race prejudice, and the significant difference for the m§3£y_inter- action vanished after adjusting the scores for the operation of perceived social pressure. In the civil rights experiment race prejudice was significantly higher for the interactions Baggy and accept as kin, but after adjusting for social pressure belief rejection for marry became significantly higher than race rejection, and the significant difference for accept as kin vanished. “9 To sum up we find that elimination of perceived social pressure influences from the prejudice scores for the interactions marry, date, and accept as kin, which showed a higher apparent race than belief prejudice, results in a reversal of the relative influence of race and belief prejudice. Before adjustment all six race re- jection scores calculated in the two experiments were numerically higher than the belief prejudice scores, however they all became lower after elimination of the effect of the operation of social pressure on race and on belief prejudice. Five of the race prejudice scores were significantly higher than the belief prejudice scores before, but after correcting for social pressure none of the race difference scores are higher than the belief difference scores and three of the belief prejudice scores are significantly higher than the race prejudice scores. The previously reported findings that for a number of interactions such as marry, date, and accept as kin race prejudice is stronger than belief prejudice can be explained by the effect of perceived social pressure, because when the effect of perceived social pressure is statistically eliminated belief prejudice becomes significantly larger than race prejudice. The significance of the adjusted race and belief prejudice scores The next question of interest is how much race and how much belief prejudice remains after the elimination of the effect of perceived social pressure. Inspection of the race rejection scores in Tables 10.and 11 shows a range of -0.2 to +1.3 and that 18 out of the 20 race rejection scores have a positive algebraic 50 sign, indicating that some race prejudice remained after elimination of the perceived social pressure effects. The adjusted belief prejudice scores range from -O.7 to “.0 and 19 of the 20 belief rejection scores have a positive sign, indicating that after ad- justment for social pressure high belief rejections still remain. Using the mean rejections scores, and the corresponding standard deviations presented in Tables 10 and 11, we can test for the significance of the deviation of each rejection score from zero, in order to find out if the adjusted prejudice scores signifies significant amounts of prejudice. Table 12 presents the summary of the teratios calculated for the mean race and belief rejection scores for each interaction in the two experiments. The results show that belief rejection is significant for all interactions in both experiments, with the exception of exclude from neighbourhood in the communist belief group, in which persons of the opposite belief are accepted as neighbours. Four of the interactions; accept as friend, admire, be commanded by, and exclude from neighbourhood show no significant race rejection in both experiments, in addition fish with, and eat with resulted in no significant race rejection in the civil rights belief experiment. Four interactions; marry, date, accept as kin, and treat as equal show small, but significant, race rejections in both experiments, even after the adjustments were made for the operation of perceived social pressure. Although elimination of the effect of perceived social pressure resulted in greater belief than race rejection in all but one interaction, for which the race and belief rejection scores Table 12 Summary of t: tests for significance of the social pressure adjusted prejudice score deviations from zero Communist Experiment Civil Rights ;xperiment _t_ Race _t_ Belief 1; Race 1; Belief prejudice prejudice prejudice prejudice Interaction scores scores scores scores Iarry 5.04** 7.51** 3.97** 5.98** Date 2.80** 5.54** n.o1** 4.72** Kin ’ 3.01M 5.13** 2.11* 11.51?" Friend 1.H6 7.07** 0.08 5.93** Fish with h.68** 12.19** 1.14 4.23** Eat with 3.31** 2.96** 1.76 4.15** Commanded by 1.73 8.02** 1.27 7.43** Admire 0.94 21.20** 0.67 2.62* Equal 2.32* u.11** 2.16‘ 3.09** Exclude 0.51 -3.25** 1.80 1.80 *p<0059 **p<001 52 were not significantly different, a null amount of race rejection remained in at least' four of the ten interactions. All race rejection scores after elimination. of the perceived social pressure effects are call and do not exceed 1.3 units as colpared with a maximum of “.8 units before the adjustment. The results show that perceived social pressure accounts for a large proportion of race prejudice expressed for the interactions m, dg_t_e_, and accept as kin, but not for the full amount of the observed prejudice scores. Rektionship betwoen social Eessure and Amjudice scores across all interactions One way to measure the degree of association between prejudice and perceived social pressure for each experiment as a whole is to compute rank order correlation coefficients behreen the prejudice and the social pressure to prejudice scores across the ten interactions, instead of across persons for each interaction. Table 13 presents the rank order correlations computed from the ten pairs of prejudice and social pressure mean scores, or one set of scores for each interaction. All the rank order correlations, with one exception, are substantial ranging from .81 to .596. The only low rank order correlation of O.l&1 was between belief prejudice and per- ceived social pressure of parents in the civil rights experi-ent. The results indicate that the degree of prejudice expressed is associated with perceived social pressure. On the whole the ngnitude of the rank order correlations, reported in Table 13, confirm the \ previous findings based on the analysis of the correlations between the rejection scores and the perceived social pressure scores in Tables 1} and 5, that both for belief and for race prejudice there 53 Table 13 Rank order correlation coefficients between mean prejudice and mean social pressure scores across all interactions Race prejudice Belief prejudice Experiment Self-parent Self-friend Selfbparent Selfbfriond Communion 0.81 0.85 0.96 0.93 Civil.Rights 0.91 0.92 0.h1 0.85 Combined 0.80 0.89 0.83 0.90 Table 10 Rank order correlations between.mean prejudice scores and coefficients of correlation between prejudice and social pressure across all interactions Race prejudice Belief prejudice Experiment Selfbparent Selfbfriend Selfbparent Self-friend 5“ is a significant correlation with perceived social pressure. The essential difference between the social pressure- prejudice relationship is pointed up by inspection of the data pre- sented in Table 1‘}. Table 1‘! carries the analysis performed in Table 13 one step further, by presenting rank order correlations bemen the prejudice scores on the one hand and the correlation coefficients bet‘teen the prejudice scores and the perceived social pressure scores on the other hand. All the rank order correlations computed for race prejudice are substantial, while all the six correlations computed for belief prejudice are mull. The differences in the relationship between the sise of prejudice and strength of social pressure between race and belief prejudice are illustrated in Figure 1. Figure 1A is a scattergram of the mean prejudice scores for each of the ten interactions against the corresponding correlation coefficient betwungthe prejudice scores and the social pressure scores for race rejection. Figure 13 is the corresponding scattergram for belief rejection. In order to simplify the illustrations in Figure 1 the scattergrams are presented for the average correlations for parents and friends, instead of separately as presented in Table 9. Figure 1 shows that the three interactions, my, date, and ggcept as Q, which gave high race prejudice responses are associated with high correlation coefficients, while the four high belief rejection interactions, _ad__m_ir_g, carded by, accept as friend, and m are associated with considerably lower correlations. For the belief rejection scores the smgnitude of the corr‘elations is level regardless of the sise of the belief prejudice, while for race Coefficient of correlation between prejudice and social pressure scores 55 .13 x o.6__ _ x x o . 0.5__, .. o x o o.u _ 0 ° ‘ o X 0 o 0'37 a x 0.2h- us- x x x o x Oel _- i— l I 1, i l l l 1 O 1 2 3 h 0 1 2 3 1+ 5 Mean Race Prejudice Score Mean Belief Prejudice Score Figure 1A Figure 1B Figure 1 Scattergrams of coefficients of correlation between prejudice scores and social pressure scores toward prejudice for the ten interactions in the two eXperiments 56 prejudice the sise of the correlation depends on the ngnitude of the prejudice. . £30021. Eessure attributed to pgents and friends Inspection of the correlation coefficients in Table 9 reveals one consistent difference between the effectiveness of social pressure attributed to parents and friends. There are no lawful differences for race rejection, but all ten of the belief difference social pressure scores are larger. for friends than parents in the two experiments combined. If we consider the two experiments separately we find that the differences arise from the fact that for the civil rights group friends' social pressure is much stronger than parents' social pressure toward belief rejection. Subjects discriminate on the basis of civil rights belief more in the direction of approval by friends than in the direction of approval by parents. DISCUSSION AND IMPLICATIONS The results show that when the effect of perceived social pressure on both race and belief prejudice is statistically eliminated ' belief prejudice is stronger than race prejudice even for intimate social interactions. The observations of Triandis and Davis (1965) that friendship acceptance operates according to the belief congruence principle, but behavioral intentions for small social distance inter- actions, such as marriage or dating, are governed more by racial considerations were confirmed. However their proposal, that degree of intimacy of the interaction will determine whether race or belief is the more important determinant of behavioral intentions, can.be replaced by the more general explanation that differential perceived social pressure toward race and belief prejudice accounts for the data in the literature. The operation of perceived social pressure is consistent with the findings of regional and national variations in prejudice, as we can expect regional and national variations in effective social pressure against interracial interactions. It is of interest to note that the perceived social pressure variable operates differently for race and for belief prejudice. The present study provides the first empirical demonstration of the fact that both perceived social pressure and effectiveness of the perceived social pressure, as measured by the coefficient of correlation between the prejudice and social pressure scores, are significant determinants of race rejection. For belief rejection social pressure is a significant factor, but the effectiveness of the perceived social 57 58 pressure is level for belief rejection across the ten interactions studied. The explanation of the previously reported observation that race prejudice is higher for marital interactions than belief prejudice is the highly effective operation of social pressure toward increased race rejection in some interactions such as, marry, date, and accept as kin. One possible problem in interpreting the results arises from the consideration that the perceived social pressure judgments were made by the same subjects, who have also provided the rejection scores. The possibility exists that the social pressure judgments may have been contaminated by rationalisation of the rejection score judgments. The risk of contamination was reduced by construction of the questionnaires such a way that the social pressure judgments were made independently of the rejection judgments. Internal com- parisons of the data support the contention that the judgments were made independently. On the civil rights belief prejudice social pressure scores friends showed significantly higher correlations with the belief prejudice eXpressed by the subjects than parents, in spite of the fact that both social pressure judgments were made on the same sheet of paper. Uniform correlations between the prejudice scores and the social pressure scores for both, parents and friends, would be expected under the assumption that the social pressure scores are the result of rationalisation. Two studies, reported by Epstein and Komorita (1966a, 1966b, ) support the hypothesis that since prejudice was learned from parents we can expect correlations between parental social L. “apt 59 pressure as perceived by the subjects and their own prejudiced choices of stimulus persons. Epstein and Komorita haveeobtained social distance scores from 9-12 year old children and three weeks later they obtained from the same children the perceived social distance scores of the parents for the same group of stimulus persons. In both studies, using white and Negro children respectively, they repor- ted high correlations between the perceived parental social distance and the childrens' own social distance. Using a different design, degree of approval and disapproval by parents and friends for the interactions with the stimulus persons could be measured independently, however even if there is divergence between the actual and perceived social pressure scores, the latter are of greater interest, because the perceived social pressure is the effective psychological variable. Effective social pressure and race rejection is the study demonstrated the degree of race prejudice is directly related to the effective operation of the perceived social pressure to race rejection. The effectiveness of social pressure is measured by the sise of the coefficient of correlation between the social pressure scores and the rejection scores. The finding that for high race rejection to take place there is a high effective social pressure present, while for high belief prejudice there is no high effective social pressure, indicates that over and above the perception of parental approval and disapproval of interactions with a different race stimulus person another, yet unmeasured, variable operates to make the social pressure more effective in race rejection for certain interactions. ‘111111 60 One possible explanation is that in addition to the degree of approral and disapproval, measured in the current study, expectations about more severe sanctions against some of the inter-racial behavioral intentions make the observed social pressure score mere effective for the situations for which there is higher race than belief rejection. Further research can answer the question: why does social pressure against dating a Negro by'a white person result in a higher degree of conformity than social pressure against dating a communist. Cogpgrison of the beliefs used In order to reduce the danger of undetected statistical interaction between the race and the belief variables employed two belief variables were used in the present study. On the whole the communist belief was found to be more powerful for belief discrimi- nation than beliefs about civil rights legislation, as belief rejection -was significantly higher in the communist than the civil rights study in nine interactions, (Appendix Table 19). In contrast to the trend for belief rejection, there are no significant differences between the race prejudice scores obtained in the two experiments. The fact that there are differences in belief rejection, but no differences for race rejection in the two experiments indicates that race and belief performed as unrelated variables. For both experiments statistical elimination of belief effects resulted in equal race rejection, expected only if there is no differential interaction between the type of belief used and degree of race rejection. The possibility that there is equal confounding of the race and belief variables in the two experiments can not be ruled out by the data. 61 However, it is more likely, that if confounding of the race and belief variables exists, it would be different for the two beliefs, one race related and the other not involved with race, and this possibility is ruled out by the analysis of the data outlined above. Prejudice, socia;_pressure to_prejudice, and balance theopy So far this report emphasised the social pressure expla- nation of the race rejections found in the three interactions for which race rejection was observed to be higher than belief rejection. There were seven other interactions, where in the absence of absolute rejection due to race, there were relatively smaller acceptance of members of the other race . we can use the notation employed by Heider (1958) in his discussion of stable triads, to present the belief congruence theory of acceptance and rejection of people as modified with perceived social pressure attributed to significant others. The simplest state- ment of belief congruence theory is illustrated in Figure 2 for person 2 who has a belief or attitude toward an object x, which is either positive (+) believes in, likes, or negative (-) dislikes, does not believe in, person 2 also perceives the relationship of another person g.to x.as positive or negative. x x x x ++ -- +- -+ P+6 P+0 p-c p-o 2a 2b g3, gg_ Figure 2 Balance theory model of belief congruence theory between person (p) and other (9) 62 Given the relationships as indicated in Figure 2, 2 will accept g_in the situations outlined in Figures 2a and 2b and will reject g_in the situations outlined in Figures 2c and 2d. The race of g,in relation to 2 does not need to be considered; it could enter the picture only if it results in misperception of the relationship of g_to x,as in racial stereotyping, or in attributing unacceptable beliefs to an undifferentiated member of another race. To consider the operation of the social pressure variable in the same system a simplification needs to be made in the relation- ship between E and g, the significant other, for purposes of illustration only. waho likes §_or is connected in some other positive way with g perceives that g rejects g_( and also wants p,to reject g, which will not change the structure of the relation- ship). Then corresponding to Figures 2a, 2b, 2c, and 2d the resulting configurations are illustrated by Figures 3a, 3b, 3c, and 3d. X X X x + + - - + - - + p +(-) 0p +(-) 0p ~(-) 0p -(-) ° 4- + - + - + - S S S 5 3a 32 32 3.01 Figure 3 Balance theory model of belief congruence theory between person 2 and other 2 modified by social pressure In situations 3a and 3b the pgx relationships lead to acceptance of o by p, while the Egg relationships will operate in the direction of rejecting g_(shown in brackets), provided the pg bond is strong. 63 Whether _o_ will be accepted or rejected depends on the relative strength of the two forces operating in opposite directions. In the situations described by Figures 3c and 3d, both sets of forces operate in the direction of rejecting g by p and as the result of social pressure the already present belief rejection will become higher. We can use the relationships in the model above to illus- trate how perceived social pressure toward race prejudice has resulted in the observed race difference scores. Equation 1, used for computing race prejudice scores for analysis of the data above, refers to the following operations with the rejection scores for the four stimulus persons: (B‘R' - 3412*) 4- (311" - B'R")..........Equation 1 Each of the four terms of equation 1 can be expressed as a balance pattern as illustrated in Figure in x x x x + +1 +» + - + - + p + (-) o p + o p - (-) o p - o + - + - s s “a - up. + 42 - 1a Figure h Balance model for race prejudice score In Figures '43, hp, 1&3, and “g the relationship between 2 and 2 resulting from the pg interaction is shown without brackets, and the relationship between 2 and g as the result of the 252 interaction is shown with brackets. 64 Since the expression (ha - hb) + (no - Md) corresponds to Equation 1, we can substitute the four corresponding p9 relationships into Equation 1, in order to obtain the resultant pg value equivalent to the race prejudice score: (+(-) - +) + (-(-) - -) = 2- (ba -4b) + (4c - 4d ) we end up with 2-, or twice the race rejection due to social pressure for a stimulus person. Because of the operation of perceived social pressure the pgx balance model needs to be modified to take into account the Egg relationship, especially when the social pressure is highly effective. The combined pggx model is a better predictor of the direction and strength of the pg_relationship than the simple pg§_model, usually employed. In more general terms whenever conformity to social pressure is a significant determinant of behavioral intentions the balance model for determining the acceptance or rejection of another person can be, and needs to be, expanded to take into account both the belief congruence and the social pressure variables. Intergpoup contacts and intergroup attitudes Amir (1969) has presented a review of the literature on the relationship between intergroup contact and changes in intergroup attitudes. He lists six favorable conditions which tend to reduce prejudice. The six conditions are: equal status contact, contact between members of the majority group and higher status members of the minoritytgroup, when an authority is in favor of the contact, when the contact is more intimate as compared with more casual contacts, when the contact is rewarding, and when members of both 65 groups interact in situations which involve important common goals. All the six factors can be characterised as conditions under which there is an opportunity for belief congruence perception and in which, if social pressure is present it is in the direction of increased intergroup interaction. The unfavorable conditions listed by.Amir are: competetive contact, when prestige of one group is lowered as the result of the contact, when members of a group are in a state of frustration, when the groups' moral or ethical standards are objectionable to each other, and when the minority groups are given a lower status position. All the contacts which are unfavorable to reduction of prejudice provide conditions for the perception of belief incongruence and social pressures for increasing prejudiced behavior are also present. It appears that to achieve more harmonious intergroup contacts between Negroes and whites integration, and the resulting contacts, by themselves are not sufficient. Two major conditions need to be met, which at times due to conflicting interests can not be achieved. There is a need for perception of belief congruence on salient issues, especially on issues which are instrumental in character and lead to achievement of commonly shared goals. Under these conditions not only the absence of social pressure to race prejudice is assured, but it is likely that social pressure toward mutual acceptance would develop. Legislation against discrimination, and the elimination of discrimination by significant others would reduce the effective Social pressure to prejudice, and would improve intergroup attitudes. ill-I'lli‘l CONCLUSIONS 1. Belief congruence is a stronger determinant of inter- personal behavioral intention than race congruence, provided no strong perceived social pressure operates against the behavioral intention. 2. For race prejudice to be stronger than belief prejudice under the condition of perceived belief congruence on a salient issue perceived social pressure to discriminate on the basis of race has to be stronger than perceived social pressure to discriminate on the basis of belief. 3. For belief prejudice the effectiveness of perceived social pressure, as measured by the correlation between perceived social pressure and size of belief rejection, is uniform for all inter- actions, but for race prejudice the effectiveness of perceived social pressure is a function of the degree of prejudice expressed. 66 BIBLIOGRAPHY BIBLIOGRAPHY Allport G. W. The nature of prejudice. New York: Anchor Books, 1958 Anderson C. C., Cote A. D. J. Belief dissonance as a source of dis- affection between ethnic groups. Journal of Personality and Social bychology, 1966, [1, 1107-1152 Amir Y. Contact hypothesis in ethnic relations. Psychological Bulletin, 1969. A. 319-342 Berg K. R. Ethnic attitudes and agreement with a Negro Person. Journal of Personality and Social Psychology, 1966, E, 215-219 Bettelheim B., Janowitz M. Social change and prejudice includinggdynamics of prejudice. New York: Free Press, 1960 Bogardus E. S. Immigration and race attitudes. Boston: Heath, 1928 Bogardus E. S..A forty year socggl distance study. Los‘Angeles: University of Southern California Press, 1967 Byrne D. Interpersonal attraction and attitude similarity. Journal_pf Abnormal and Social ngghology, 1961, 62, 713-715 Byrne D..Authoritarianism and response to attitude similarity- dissimilarity. Journal of Social Psychology, 1965. 632. 251-256 Byrne D., wong D. J. Racial prejudice, interpersonal attraction and assumed dissimilarity of attitudes. Journal of.Abnormal and Social Psychology, 1962, 62, 246-253 Byrne D.,McGraw C. Interpersonal attraction toward Negroes. Human Relations. 1969, 12, 201-213 Gilbert G. M. Steretype persistence and change among college students. Journal of.Abnormal and Social Psychology, 1951, 36, 245-254 Epstein R., Komorita S. S. Childhood prejudice as a function of parental ethnocentrism and punitiveness. Journal of Personality and Social Psychology, 2, 1966, 259-264 Epstein R.. Komorita S. S. Prejudice among Negro children as related to parental ethnocentrism and punitiveness. Journal of Personality and Social Psychology, 1966, 3, 643-647 Hays W. L. Statistics for pgychologists. New York: Holt, Rinehart and Winston, 1963 67 68 Heider F. The psyphology of interpersonal relations. NeW'York: Wiley, 1958 —f Insko C. A., Robinson J. E. Belief similarity versus race as determi- nants of reactions to Negroes by Southern white adolescents: a further test of.Rokeach's theory. Journal of Personality and Social Psychology,lz, 1967. 216-221 Katz D., Braly K. W. Racial stereotypes of 100 college students. Journal of.Abnormal and Social Psychology, 1933. ZQ, 280-290 McDougall'W. Introduction to Social Psychology. London: Methuen and Co., 1908 Malof M., Lott.A. E. Ethnocentrism and the acceptance of Negro support in a group pressure situation. Journal ofHAbnormal and Social Psychology, 1962, 65, 254-258 Martin B. M. E. Ethnic group and belief as determinants of social distance. Unpublished M. A. thesis, University of'Western Ontaric , 7r“ 19 Nelson A. M. The relative importance of race,_belief and values as determinants of discrimination. Unpublished M. A. thesis, Michigan State University, 1965 Newcomb T. M. Persistance and regression of changed attitudes: long range studies. Journal of Social Issues, 1963, 12, 3-14 Newcomb T. M. Interpersonal constancies, Psychological and Sociological approaches, in Klineberg 0., Christie F. (Eds.) Perspectives in Social Psychology. New York: Holt, Rinehart, 1965, 38-h9 Rokeach M. The open and closed mind. New York: Basic Books, 1960 Rokeach M. Belief vs. race as determinants of social distance: Comments on Triandis's paper. Journaliof Abnormal and Sgcial Psychology, 1961, ég, 187-188 Rokeach M., Smith P. W}, Evans R. I. Two kinds of prejudice or one? in Rokeach M. The open and closed mind. Ibid, 132-168 Rokeach M., Mezei L., Race and shared belief as factors in social choice. Ross E. A. Socialrfsychology. New York: Macmillan, 1908 Stein D. D. Similarity of belief systems and interpersonal preferences: A test of.Rokeach's theory o:_prejudice. Unpublished Ph. D. dissertation, University of California, Berkeley, 1965 I 1 l lllll‘ll I'll 69 Stein D. D. The influence of belief systems on interpersonal preference. Psychological Monographs, 1966, 66, whole number 616 Stein D. D., Hardyck J. A., Smith M. B. Race and belief: An open and shut case. Journal of Personality and Social Psychology, 1965, Smith C. R., Williams L., Willis R. H. Race, sex, and belief as determinants of friendship acceptance. Journal of Personality and Social Psychology, 1967. 5, 127-137 Triandis H. C. A note on Rokeach?! theory of prejudice. Journal of Abnormal and Social Psycholggy, 1961, 62, 184-186 Triandis H. C. Exploratory factor analyses of the behavioral components of social attitudes. Journal of Abnormal and Social Psychology, 1960, 66, #20-030 Triandis H. C. Toward an analysis of the components of interpersonal attitudes. in Sherif C. F., Sherif’M. (eds.).Attitude,»ego- inv61vements and change. New York: Wiley, 1967, 227-270 Triandis H. C., Davis E. E. Race and belief as determinants of behavioral intentions. Journal of Personality and Social Psycho- 10 t 1965: .2_: 715-726 Triandis H. C., Loh W} D., Levin L. A. Quality of spoken English, and opinions about civil rights as determinants of interpersonal attitudes. Journal of Personality and Social ngchology, 1966, 2, 468-472 University of Michigan Survey Research Center. Discrimination without, ppejudice, a study10f promotion_practices in industry. Ann.Arbor: University of Michigan, 1964 Willis R. H., Bulatao R. A. Belief and ethnicity as determinants of friendship and marriage acceptance in the Phillippines. Paper presented at the 75th annual meeting of the.American Psychological Association, 1967 APPENDICES APPENDIX A Sample questionnaire for white communist stimulus person The purpose of this study is to find out to what extent, or how strongly, do people intend to interact with different persons. In making your choices please indicate how you would behave in a certain situation toward the person described. You are asked to make your choices known by using scales, which have seven possible positions. Please indicate one choice for each question in the order they are presented. Here is how to use the scales: I would : : : : : : I would not very quite slightly neutral slightly quite very strong strong strong strong strong strong ‘would would would would 'would would not not not As you can see the seven scale points from left to right represent decreasing strength of very strong positive intention (first scale point) through a neutral intention (middle scale point) to very strong negative intention (last i. e. seventh scale point). If’you feel that you‘would choose quite strongly a given person for the indicated interaction, as presented below, you should place a check mark as follows: I would_:_)_(_:__:__:__:_:_1 would not On the other hand if you feel that quite strongly you would not choose a given person for the indicated interaction, as presented below, you should make your check mark as follows: I would__:__:__:__:__:_X_:__I would not You should use the other five scale positions as applicable to your choice the same way as illustrated above. Remember that the neutral point means that you are as much for as against interacting with a person in a given situation, and therefore should be used only if you have nc~positive or negative preference. IMPORTANT 1. Please check your marks in the middle of the spaces, not on the boundaries. _:_§_:__:__:__I___:__ this not this 2. Be sure you check every interaction item for every person. 3. Never put more than one check mark on a single scale. h. Do not think too long about a given choice, but make sure that you are aware of both the person (described on the top of the page) and the particular social behavior (described on the scale). After questions about yourself you will find scales, which are built similarly, make sure you carefully answer the questions on them. 70 71 A WHITE PERSON WHO IS A COMMUNIST Iwould : : : : : : Iwouldnot ADMIRE THE IDEAS OF THIS PERSON I would__:__:__:_:_:_:_I would not BE COMMANDED BY THIS PERSON Iwould : : : : : : Iwouldnot GO FISHING WITH THIS PERSON Iwould:::: ::Iwou1dnot Iwould:::: ::Iwou1dnot Iwould : : : : : : Iwouldnot EXCLUDE THIS PERSON FRCM MY NEIGHBOURHOOD Iwould : : : : : : Iwouldnot ACCEPT THIS PERSON AS A CLOSE KIN BY MARRIAGE Iwould : : : : : : Iwouldnot TREAT THIS PERSON AS AN EQUAL Iwould : : : : : : Iwouldnot Iwould : : : : : : Iwouldnot ACCEPT THIS PERSON AS AN INTIMATE FRIEND 72 HOE uoc HOE HOE HOE no: HOE HOE HOE HOE Lo_>m£on >5 mo o>oLaam o_:oz ” u u u n u o_:oz Lo_>m;on >5 mo o>oLdam Lo_>lmwml>ll+mlwbm+mmm o_:oz " . u o_303 ao_>m;on >5 mo o>oLQdm II.II.:I.:I:.I:.I:.Ilo_:oz n_:oz ” H u u u u Lo_>c;on >5 E0 o>oLaam o_3011I" n n u u ” Lo_>lmNMI>5 mo o>onqam o _ :03 ” .Il....l.||"||o _ :03 Lo_>mcon >5 mo e>oLaam o_:oz ” u u u u " o_:oz Lo_>mnon >5 Ho o>odem Hm_z:zzoo < m. HOE HOE HOE HOE HOE HOE HOE HOE HOE HOE Lo_>m;on >5 mo o>0nadm o_:oz ” u n u u " p.303 Lo_>m;on >5 mo o>oLaam Lo_>lmwml>ll+mlo>odem o _303 n . .nlullzsoz Lo_>m;on >5 mo o>oLadm m h z m x < m > 2 0:3 zommmm mh_13 < ozm_mu mhm 2.x mmonu < m< zommmm m_:h hmmuu< _ m_ ooomeOmIo_mz >2 zomm zommmm m_:h moanuxm _ m. zommmm m_IH mhmmm omoz