I I II | I II I III‘ I I II 'ov—aI IMPLIED EVALUATION AS A MEDIATOR OF THE ATTITUDE SIMILARITY - ATTRACTION RELATIONSHIP Thesis for the Degree of M. A. MICHIGAN STATE UNIVERSITY JOHN W. CONDON 1977 II IIIIIIIZIIIIILIII IIIIIIIIIIIIIIIIIIIIIII 1/ LIB RA RY Q51g . . '. '3 I‘r-I=1»ICI1:E,:L:1 buts fl Umwcrsxty j ABSTRACT IMPLIED EVALUATION AS A MEDIATOR OF THE ATTITUDE SIMILARITY - ATTRACTION RELATIONSHIP BY John W. Condon In research that employed the hypothetical stranger paradigm (Byrne, 1971) it was predicted and demonstrated that the functional relationship between attitude similarity and attraction was mediated by implied evaluation. Previous work (Byrne and Rhamey, 1965) which reported independent effects of attitude similarity and personal evaluations on attraction was questioned on methodological grounds. These ”independent effects” were shown to have been a function of improper use of an analysis of variance design with psychologically dependent factors. Using partial correlational analysis of measures of perception of agreement with the stranger, subjects' expectations of strangers' liking for the subjects, and subjects' attraction to strangers, it was found that implied evaluation explained 43 percent of the variance in agreement induced attraction, while the partial correlation supporting any direct effect of agreement on attraction accounted for only 3 percent of the variance. It was suggested that the implied evaluation model, in which agreement leads to expected liking which results in attraction, might John w. Condon operate because of an expectation of mutual need gratification between persons who like each other. IMPLIED EVALUATION AS A MEDIATOR OF THE ATTITUDE SIMILARITY - ATTRACTION RELATIONSHIP BY John w. Condon A THESIS Sublitted to Michigan State University in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Psychology 1977 ACKNOWLEDGMENTS One of the most rewarding aspects of this work has been the continuing realization that the members of my thesis committee are genuinely interested in my development as a social psychologist. I would like to express my appreciation to Bill Crane for his encouragement, guidance, and friendship. Bill's help has been invaluable throughout this research. But, more than that, he pro- vides a.model of scholar and teacher that one can only hope to approximate. I would like to thank Joel Aronoff for continually insisting that I define my question and for comments and criticisms which have led me to a more thorough understanding of my study. I don't know whether to thank Larry Messé more for identifying problas or for helping to solve them. His ability and willingness to do both have contributed much to this research. Finally, thanks to Carol, Debbie, Irene. Kate, John, and Shirley for their friendship and what remains of my sanity. ii TABLE OF LI ST or TEES . C O C O O O O O 0 INTRODUCTION . . . . . . . . . . . Overview of the Problem . . . Review of Area . Evidence for a Confounding . Goals of the Present Research Overview of Subjects . . . . . . . . . . Design Procedure . . . . . . . . . . Measures . . . . . . . . . . mm C C O C O O O O O O O O O O Replication . . . . . . . . . Confounding . . . . . . . Implied Evaluation Model . . Sincerity and Believability . Olmss I“ O O O O O O O O O O O O Replication . . . . . . . . . Confounding . Sincerity and Believability . Implied Evaluation Implications for Further Research . . . . . . . . . . . . . CONTENTS Page a s s o s s o o o s s o o (:3qu 10 10 11 O O O O O O O O O O O O O O C 11 O O O O O C O O O O O O O O O 11 O O O O O O O O O O O O C O O 14 O O O O O O O O O O O O O O 16 O O O O O O O O O O O O O O O 16 O O O O O O O O O O O O O O O 17 O O O O O O O O O O O O C O O 20 O O O O O O O O O I O O O O O 23 O O O O O O O O O O O O O I O 25 O O O I O O O O O O O O O O O 25 O O O O O O C O O O O O O O O 25 O O O O O O O O O O O O O O C 26 O O O I O O O I O O O O O 27 28 30 31 iii Appm I CBS 0 . O O O O O O O O O I O C O 0 APPENDIX A. Survey of Attitudes . . . . . . . . B. Interpersonal Judgment Scale . . . C. Expected Liking and Perception of Agreement Measures D. Sincerity and Believability Measures iv Page 35 35 38 4O 42 LIST OF TABLES Table Page 1. Analysis of Variance on Attraction Scores by Attitude Sflmilarity, Evaluation. and Order Of "an.“es O O O O O O O O O O O O O O O O O O O O O O O 1 7 2. Analysis of Variance on Expected Liking Scores by Attitude Similarity, Evaluation, and Order of Measures . . . . . . . . . . . . . . . . . . . . 18 3. Analysis of variance on Perception of Agreement Scores by Attitude Similarity, Evaluation, and Order of Measures . . . . . . . . . . . . . . . . . . 20 4. Partial Correlations Between Perception of Agreement and Attraction with Expected Liking Held Constant, and Between Expected Liking and Attraction with Perception of Agreement Held Constant . . . . . . . . . . . . . . . . . 22 5. Analysis of Variance on Sincerity Scores by Attitude Similarity. Evaluation. and Order Of ”“8““, O O O O O O O O O O O O O O O O O O O O O O O 23 6. Analysis of variance on Believability Scores by Attitude Similarity, Evaluation, and “a“ o: mamas 0' O O C O O O O O I O O O O O O O O O O 24 CHAPTERI INTRODUCTION Overview of the Problem It is a truism in the social psychological literature that people consistently express liking for others who share their atti- tudes. Literally hundreds of studies could be cited in support of the agreement - attraction link (Newcomb. 1961; Berscheid and walster, 1969; Aronson. 1970; Byrne. 1969, 1971; Byrne and Griffitt, 1973). The research reported herein is addressed to the question of why this effect occurs. It has been demonstrated repeatedly that the greater the extent of agreement between two persons. the more one is attracted to the other. Furthermore. the generality of this relationship has been verified across a number of stimulus modes, including measurement of attraction to bogus strangers whose paper and pencil attitude surveys were actually prepared by the experimenter (Byrne. 1961. 1962, 1971). to tape recorded voices and video taped attitude expressions (Byrne and Clore. 1966). to experimental confederates whose attitude responses were prearranged (Aronson and Worchel, 1966; Byrne and Griffitt. 1966), between two persons who were expressing their true attitudes on a selected topic (Brewer and Brewer, 1968), and among two groups of college men who lived together for fifteen week periods 1 (Newcomb, 1961). Each of these studies has yielded a positive rela- tionship between attitude similarity and liking. While there is little question that individuals consider agreeing others to be more attractive than disagreeing others, there have been some lively debates concerning the dynamics of the agreement - attraction relationship (cf. Aronson and “orchel. 1966: Byrne and Griffitt. 1966: Kaplan and Anderson. 1973a, 1973b: Byrne, Clore, Griffitt, Lamberth, and Mitchell, 1973a, 1973b; Insko, Thompson, Stroebe, Shaud, Pinner, and Layton, 1973). Much of this controversy has centered on the nature and relative importance of different kinds of intervening variables. Byrne and his colleagues (Byrne. 1971; Byrne and Clore, 1967, 1970; Byrne and Lamberth, 1971) have maintained that agreement pro- vides a person with consensual validation and helps to satisfy, ”a learned drive to be logical, to interpret the environment correctly, and to function effectively in understanding and predicting events“ (Byrne and Clare, p. 101). Byrne and Clare label this drive the 'effectance motive" (after White, 1959) and they suggest that agree- ment. in satisfying this motive, takes on the properties of an uncon- ditioned stimulus in higher order classical conditioning and that it elicits an affective response which is paired with its source (i.e., the agreeing or disagreeing other). Other researchers (Aronson and worchel, 1966; Nelson. 1966; Hclhirter and Jecker, 1967) have suggested that the agreement - liking relationship might be mediated by an implied evaluation from.the other. It is one intent of the present research to demonstrate that an implied evaluation model provides a reasonable explanation of the agreement - attraction relationship. The thesis of this model is that people use attitudinal agreement as a signal that the agreeing or disagreeing others will like or dislike them. Increases in agreement lead to higher expectations of liking from the other. It is this expected liking that affects a person's attraction to another, with attraction being an increasing function of expected liking. This implied evaluation model assumes that people have reason to like others who like them. The sensibility of this assump- tion is supported by a plethora of evidence which suggests that people are inclined to expect to have rewarding interactions with those who like them (cf. Thibaut and Kelley, 1959). The notion that people prefer rewarding interactions to the unrewarding variety probably represents one of psychology's few approximations of a natural law. One benefit of the implied evaluation model is that it can account for individual motivational differences. While the effectance motive explanation assumes that all persons respond to agreement on the basis of the same need. no such assumption is necessary with the implied evaluation model. It is true that agreeing others who like us may provide us with consensual validation, but they may also feed us, bolster our self esteem, provide us with sexual gratification and so on. Response to attitude similarity need not be thought of as being unidimensional across persons. Byrne has dismissed the implied evaluation model, or, as he calls it, the approval-cue hypothesis, on both theoretical and empirical grounds. Theoretically, he argues that the approval-cue hypothesis fails to account for the association of similarity with liking and dissimilarity with disliking, in the subject's experience: There is not much gained in explanatory power by transporting the attitude - attraction effect from the experimental room to the scene of the subject's early childhood experiences (Byrne, 1971, p. 337). ‘ It is difficult to understand the aversion of a reinforcement theorist toward acceptance of the possibility of a simple learned association between agreement and reciprocated liking. Further, it should be noted that the manner in which agreement operates as an unconditioned stimulus in Byrne and Clore's effectance motive explanation depends entirely on events occurring outside of the experi- mental room and, quite possibly, during the subject's early childhood. On the empirical side, Byrne (1971) cites evidence that attitude similarity and personal evaluations of the subject.combine additively to affect people's attraction to others, with personal evaluations accorded three times the weight of attitudes in predicting attraction. Further, he maintains that these two types of reinforcers exert independent effects on the attraction response. However, these “independent effects" come about strictly as a function of the design employed by Byrne and his colleagues. Independence is never demon- strated, but is forced by the constraints of the experimental design. The purpose of the present study is to show that the implied evalu- ation model, which predicts a relationship between agreement and personal evaluation, has never been subjected to an adequate empirical test and that, when properly studied, the agreement - liking relation- ship is, in fact, mediated by anticipated evaluation. Review of Area Early research utilizing Byrne's attraction paradigm failed to uncover any differential effects on the attraction response attributable to the importance of attitude topics (Byrne and Nelson, 1964). This was a disturbing finding since it had been assumed that variations in agreement on more important topics should exert a greater influence on attraction than such variations on less sub- jectively important topics.1 The negative results of Byrne and Nelson's topic importance study coupled with a desire to identify stimuli which would produce greater effects on attraction than that of attitude similarity led Byrne and Rhamey (1965) to study the relative effects of personal evaluations and attitude similarity on liking for a stranger. Previous research (Backman and Secord, 1959: walster and Halster, 1963; Aronson and Linder, 1965) had established that people's attraction to others varied as a function of the positivity of personal evaluations supplied by those others. Byrne and Rhamey reasoned that people were highly motivated to ”maintain and enhance" their self concepts and, therefore, that positive and negative personal evaluations would act as potent rewards and punishments. They went on to suggest that even if the self is considered to be "another topic about which attitudes may differ” (p. 885), it should be of vastly greater importance than other impersonal topics. Factorial combinations of four levels of pro- portion of attitude similarity (1, 2/3, 1/3, and 0) and three levels of evaluation (positive, negative, and a no evaluation control group) yielded significant effects for both ”independent” variables and their interaction. Post hoc analysis allowed the respective assignment of 3:1 weighting coefficients to personal evaluations and agreement on impersonal topics in the prediction of attraction responses. Aronson and Norchel (1966) first suggested the implied evalu- ation model. They proposed that the attitude similarity - attraction relationship llmay be due, at least in part, to an implicit assumption that people who hold attitudes similar to our own will like us” (p. 157). They also asserted that the stronger influence of personal evaluations would be particularly evident in a face-to-face inter- action, and that, in such a setting, agreement would have no effect on attraction. The Aronson and “Orchel ”replication" of the Byrne and Rhamey study utilized two levels of agreement (71% and 29a) and two levels of personal evaluation (positive and negative) in a face-to-face situation involving a naive subject and a confederate. The Aronson and "orchel study left much to be desired on methodological and statistical grounds. Byrne and Griffitt (1966) reanalyzed Aronson and Nbrchel's data and were able to show that the results were quite compatible with those reported by Byrne and Rhamey, thus extending \ the three to one weighting of evaluations versus agreement to face- to-face interactions. This finding of the greater potency of evalu- ations over agreement has received wide support. The same pattern has been reported by Clore and Baldridge (1970), Gouaux and Summers (1973), Johnson (1974) and Singh (1974, 1975). Evidence for a Confounding2 There are compelling reasons to suspect that the ”independent“ effects associated with evaluations and agreement in the above mentioned studies were based on a logical error which led to the production of these results through a statistical artifact. First, as has been noted by Kaplan and Anderson (1973a), the assignment of relative weights in the Byrne and Rhamey study relied on an additive model. The presence of a significant interaction between agreement and evaluation raises serious questions regarding the additivity of effects of these two variables. Second, a number of experimenters have found that increases in agreement led to similar increases in subjects' anticipation of liking from.the other (Nelson, 1966; HcWhirter and Jeoker, 1967: Insko et a1., 1973). If this finding is veridical, then manipulation of the agreement variable is accompanied by uncontrolled variation in expected evaluation. Thus, these studies in which both agreement and evaluation are “independently“ manipulated, and their relative effects assessed (e.g., Byrne and Rhamey, 1965) may involve a con- founding which vitiates interpretation of their results. Any factorial combination of these two variables, which has treated them as if they were actually orthogonal and has not taken into account the probable high degree of relationship between them, was in violation of one of the assumptions of the statistical test on which the findings were based. In a somewhat related study Touhey (1975) has shown that controlling for interpersonal congruency (the extent to which self- ascribed traits are attributed to one by another) eliminated the attitude similarity - attraction effect. In addition, at least one investigator (Stephan, 1973) reported that subjects distorted the shailarity information to make it compatible with the evaluation condition. If, as is probable, the manipulation of attitude simil- arity is differentially perceived across levels of evaluation and the same is true for evaluation across levels of attitude similarity, then the reported results of omnibus P tests on uncontrolled attraction scores are clearly inappropriate for assessing any effects due to either factor. goals of the Present Research It should be evident at this point that the role of anticipated liking in the agreement - attraction relationship probably has been confused in previous research. The present study was conducted in order to lend support to the notion that the effect of attitude similarity on liking is mediated by implied evaluation. With these goals in mind, the following predictions were tested. 1. Replication of the Byrne and Rhamey study with the addition of measures designed to assess subjects' perceptions of the agreement and evaluation manipulations would support the contention of a specification confounding of these two variables in previous studies. 2. The agreement - attraction relationship is mediated by implied evaluation. That is, perception of agreement leads to an expectation of evaluation which, in turn, exerts an effect on attraction. CHAPTER II METHOD Overview of the Experiment The procedures employed in this study replicated those of Byrne and Rhamey (1965),3 with the addition of measures designed to assess subjects' perceptions of agreement with the stranger, and expected liking of the subject by the stranger. Subjects whose attitudes on a variety of topics had been obtained previously were provided with the responses of a stranger on the same topics. Bogus attitude surveys were prepared by the experimenter to represent the attitudes of the stranger. Agreeing and disagreeing responses on these surveys were arranged to correspond- with a predetermined proportion of attitude similarity for each subject, i.e., either 100 percent agreement, 67 percent, 33 percent, or 0‘ agreement. In addition to this, a number of subjects were presented with evaluations from the stranger which were ostensibly based on the stranger's reading of the subject's own attitude responses. These evaluations were also under the control of the experflmenter. Measures of the subject's probable liking for the stranger, expectation of stranger's liking for the subject, and the extent to which the subject perceived agreement between self and stranger were obtained. 10 11 Sub ects Attitude surveys were completed by the members of a large introductory psychology class and randomly assigned to levels of the similarity and evaluation variables. While there was some subject attrition due to nonattendance at the second experimental session, there was no evidence to suggest that loss of subjects was related, in any way, to experimental treatments. The data of this experiment consist of the responses of a sample of 226 students, that included approximately equal numbers of males and females. Design The design employed in this study was a 4 x 2 x 2 between- groups factorial. The first factor was proportion of attitude similarity (1, 2/3, 1/3, or O) . The second factor represented the evaluation variable (positive , or no evaluation supplied). The third factor, which was included to assess possible measurement order effects, was the order in which an expected liking measure and a perception of agreement measure were presented to subjects. Procedure The present study employed the “hypothetical stranger“ pro- cedure which has been used extensively by Byrne and his associates (Byrne, 1971) . A twelve item Survey of Attitudes (Appendix A) was selected fra Byrne's 56 item pool (Byrne, 1971) . Three items were chosen from each of the four levels of topic importance identified by Byrne and Nelson (1964) . Selection of items within each level of topic 12 importance was random with the exception that items lacking contem- porary relevance (e.g. , draft deferments) were excluded. Two days after completing the Survey of Attitudes, subjects participated in an experiment ostensibly concerned with the accuracy with which individuals can make interpersonal judgments on the basis of limited information. Ayaluation Hanialationn-One hundred and eleven subjects were in the positive evaluation group. They were given the following instructions. This study is concerned with the accuracy with which individuals can make interpersonal judgments on the basis of limited information. Your name was removed from the Survey of Attitudes which you completed the other day and your survey was given to another student of the same sex as yourself. We thought it might be interesting for you to know how he or she judged you and if this additional information would affect your judgments about her or him. Please read over this person's judgments of you (which follow) and try to form an impression of the person. Once you have read the judgments, do not return to this . section of the booklet. These instructions were followed by an Interpersonal Judgment Scale (138: Byrne, 1971: Appendix B) on which the “stranger" (actually the experimenter) had indicated that the subject was probably very much above average in intelligence , very mch above average in knowledge of current events, moral, well adjusted, that the ”stranger“ would probably like the subject very much, and would probably very much enjoy working with the subject in an experiment. As in the Byrne and Rhamey study, the attitude similarity manipulation followed. 13 An additional 115 subjects did not receive any evaluation from the ”stranger.“ These subjects were exposed only to the atti- tude similarity manipulation. Attitude Similaritx Manigulmtion.--A Survey Of Attitudes, representing those of the stranger , was completed by the experimenter for each subject. These surveys were constructed so as to correspond to one of four levels of proportion of similar attitudes (1 - agreement on twelve of twelve items: 2/3 - agreement on eight items, disagree- mant on four: 1/3 - agreement on four items, disagreement on eight: and O - disagreuant on all items). The strangers ' surveys were preceded by instructions relevant to the appropriate evaluation condition. For the positive evaluation group: On the next few pages you will find a Survey of Attitudes filled out by the same student whose judgments of you you have just read. Please read this person's responses to the twelve attitude scale items and try to form an impression of the person. Once you have read this person's responses, do not return to this section of the booklet. following this section you will be asked to judge this person on a few scales. for the no evaluation group a This study is concerned with the accuracy with which individuals can make interpersonal judgments on the basis of limited information. On the next few pages you will find a Survey of Attitudes filled out by another student of the same sex as yourself. Please read this person's responses to the twelve attitude scale items and try to form an impression of the person. Once you have read this person's responses do not return to this section of the booklet. following this section you will be asked to judge this person on a few scales. 14 leasures Immediately after reading the stranger's Survey of Attitudes, each subject responded to a number of measures designed to gauge perceptions of the evaluation and agreement variables, attraction to the stranger and the stranger's sincerity and believability. Expggtad Liking.--All subjects responded to two seven point scales which asked them to predict the strangers' liking for them (Appendix C). These items were the reverse of the last two items of Byrne's IJS, i.e., liking and desirability as a work partner. Data on these two items were summed to yield an expected liking score for each subject. This procedure is consistent with the customary combination of the final two IJS items as a measure of the subject's attraction to the stranger. It should be noted that the unbiased prediction of the stranger's liking for the subject was possible only for the no evaluation group. for the positive evaluation group, this measure served primarily as a manipulation check. Perception of Aggeement.--Subjects were asked to indicate on a seven point scale, bounded by complete agreement and complete dis- agreement (Appendix C), how'much they felt that they and the strangers were in agreement on the Survey of Attitudes. Since the present study was directed, in part, toward the demonstration of a confounding of attitude simdlarity and evaluation in previous research, the expected liking and perception of agreement questions preceded other measures. The perception of agreement and expected liking measures are of critical importance to this research. They were used both to establish the confounding of attitude similarity and evaluation and 15 to support implied evaluation as an intervening variable in the agreement - attraction relationship. Attraction to the Stranger.--Subjects completed Byrne's six item IJS (Appendix 3) consisting of seven point scales asking them to rate the stranger on intelligence, knowledge of current events, morality, adjustment, and the final two items--liking and desirability as a work partner-~which were combined to give a thirteen point attraction measure. Sincerity and Believability.--Subjects responded to two seven point scales which required them.to rate the strangers' sincerity and the believability of the information which they had received (Appendix D). Debriefing.--rollowing the experiment, subjects were debriefed. This included a complete disclosure of the experimental design and hypotheses and a survey of research employing the hypothetical stranger paradigm. CHAPTER I I I RESULTS Replication A three way analysis of variance, using attitude similarity, evaluation, and order of measures as independent variables, was per- formed on the attraction scores. Results of this analysis are pre- sented in Table 1. It may be seen that both the attitude similarity and evalu- ation variables exerted significant main effects on attraction, as measured by the last two items of the IJS. The mean attraction scores for the 1, 2/3, 1/3, and 0 proportions of agreement were 11.84, 10.53, 9.54, and 8,53 respectively (scale range 8 2-14). The mean attraction response for the positive evaluation groups was 10.74 versus 9.42 for the no evaluation groups. Unlike the Byrne and Rhamey results, the interaction between attitude similarity and evaluation was not significant. As has been ‘mentioned previously, the Byrne and Rhamey additive model neither predicts nor explains such an interaction. These results fit Byrne and Rhamey's theory better than do their own. Thus, at least con- ceptually, the present study replicated that of Byrne and Rhamey. 16 17 Table l.--Analysis of Variance on Attraction Scores by Attitude Similarity, Evaluation, and Order of Measures. Source of Variation 9_f_ g g Attitude Similarity (A) 3 110.148 28.885* Evaluation (8) 1 102.961 27.001* Order of neasures (O) 1 .532 .140 A x E 3 3.887 1.017 A x O 3 2.213 .580 B x O l .129 .034 A x E x O 3 6.345 1.664 Within Groups Error 209 3.813 *p_< .001 Oonfounding_ It has already been argued that the foregoing analysis does not provide a proper test of the relative effects of attitude simdl- arity and evaluation since these two variables were confounded. One reason for the inclusion of expected liking and perception of agree- ment measures in this study was to show that subjects' perception of the attitude similarity (agreement) variable would be influenced by the level of the evaluation variable to which they had been assigned. Similarly, it was expected that presentation of the various levels of attitude similarity would produce variation in expected liking between the levels of the evaluation variable. The following analyses are offered as evidence of this confounding. 18 First, scores on expected liking and perception of agreement were positively correlated. The Pearson product moment correlations between expected liking (e) and perception of agreement (a) were: for the positive evaluation group, £3 e - .678, g < .001: for the no evaluation group, r -ae .695, p_ < .001. These positive correlations raise doubts about the - .698, p < .001: and across both groups, 5” - psychological independence of agreement and evaluation. Second, an analysis of variance was performed on the expected liking scores with attitude similarity, evaluation, and order of measures as independent variables (Table 2). Table 2.--Analysis of Variance on Expected Liking Scores by Attitude Similarity, Evaluation, and Order of neasures. Source of Variation 9; gas 5 Attitude Similarity (A) 3 134.157 40.046* Evaluation (E) 1 45.418 13.5571- Order of Ileasures (0) l .048 .014 A x E 3 2.030 .606 A x O 3 11.316 3.378“ E x 0 1 12.393 3.699 A x E x O 3 4.026 1.202 within Groups Error 209 3.350 .2 < .001 “2 < -02 further analysis of the significant attitude similarity by order of measures interaction revealed that it reflects minor 19 differences in the 2/3 proportion of agreement cells. These differ- ences are theoretically uninteresting since there are no deviations from the predicted pattern of results. The significant main effect for attitude similarity should be noted. The mean expected liking scores for the levels of attitude similarity were 1 - 11.67, 2/3 - 10.28, 1/3 - 9.00, and 0 - 8.07 (scale range - 2-14). The greater the proportion of similar attitudes, the more the subject expected to be liked by the stranger. The evaluation manipulation was supposed to have been the subjects' source of information about the strangers' liking of the subjects. Yet, these data suggest that another source of such information, attitude similarity, produced systematic variation in expected liking. This result further weakens the case for independence of attitude similarity and evaluation. Third, an analysis of variance on perception of agreement scores with attitude similarity, evaluation, and order of measures as independent variables is presented in Table 3. The main effect for evaluation indicates that presentation of a positive evaluation increased subjects ' perceptions of agreement with the strangers on the Survey of Attitudes. [lean scores on this measure were 3.90 for the positive evaluation groups versus 3.52 for the no evaluation groups (scale range - 1-7) . Subjects who received positive evaluations consistently reported higher levels of agreement than did subjects who received no evaluations, in spite of the fact that the experimental manipulation of agreement was the same for both evaluation groups. 20 Table 3.--Analysis of Variance on Perception of Agreement Scores by Attitude Similarity, Evaluation, and Order of Measures. Source of Variation 9; _M§ g Attitude Similarity (A) 3 113.027 71.681* Evaluation (E) 1 9.001 5.708“ Order of Measures (0) l .000 .000 A x E 3 .413 .262 A x O 3 1.494 .948 E x O 1 1.169 .741 A x E x O 3 1.453 .921 Within Groups Error 209 1.577 *2 < .001 “E < .02 Taken together , the three preceding analyses provide a strong case for the nonorthogonality of attitude similarity and evaluation. glied Evaluation Model Given the confounding of attitude similarity and evaluation, how are we to determine the effects of these two variables on attraction? Clearly, any instance in which a subject receives independently manipulated information about agreement and evaluation is suspect. The following analyses, accomplished using data from the no evaluation groups (ll - 115), were designed as a test of the inlied evaluation model which predicts that expected liking should be an intervening variable in the agreement - attraction relationship. 21 If the implied evaluation model is veridical, then the partial correlation between expected liking (e) and attraction (1) with perception of agreement (a) held constant should be greater than zero. If implied evaluation does not mediate the agreement-attraction relationship then this partial correlation should be zero. In the present study _r_‘3 - .654, p < .001. 1-a Similarly, if, for these data, implied evaluation is the sole mediator of the similarity - attraction relationship, then the partial correlation between perception of agreement (a) and attraction (1) with expected liking (e) held constant should be around zero, but if agreement has an effect on attraction independent of implied evaluation then this partial correlation should be greater than zero. Data from this investigation yielded, 511.. addition, .654 is significantly greater than .185, p < .001. The - .185, g< .024. In partial correlation supporting implied evaluation as an intervening variable in the agreement - attraction relationship accounted for 43 percent of the variance in attraction scores while the partial r_ supporting any effect of agreement independent of that mediated by imlied evaluation accounted for only 3 percent of the variance. Partial correlations on these measures for all groups in this study are presented in Table 4. The same pattern discussed above was obtained in all cases, but these results should be interpreted cautiously since, in many cases, they may be biased by the presen- tation of positive evaluations . 22 Table 4.--Partia1 Correlations Between Perception of Agreement and Attraction with Expected Liking Held Constant, and Between Expected Liking and Attraction with Perception of Agreement Held Constant. Proportion of Similar Attitudes Across O 1/3 2/3 1 Levels No evaluation as. 26 25 29 23 112 Ea1-e .113 .193 .156 .099 .185+++ r .765* .365+++ .620* .687* .654* -el-a Positive Evaluation 9g_ 26 26 25 22 100 Eel-e .133 .236 .318 -.034 .202++ Ee1... .362+++ .807* .550* .471+ .557. 111 Subjects: g;_- 223, r -al-e . *0 .182 , £91 *p_< .001 **p_< .003 tp.< .01 ++p_< .02 +H'p_< .03 a - perception of agreement e - expected liking 1 - attraction 23 Sincerity and Believability An analysis of variance on sincerity scores by attitude similarity, evaluation, and order of measures yielded significant main effects for both attitude similarity and evaluation (Table 5). Agree- ing others were seen as more sincere with the means for the attitude similarity conditions being, 1 - 5.78, 2/3 - 5.57, 1/3 - 5.53, and 0 - 4.88 (scale range - 1-7). Subjects who received no evaluation judged the stranger as being more sincere than did those who received positive evaluations (5.62 versus 5.23). Table 5.--Analysis of variance on Sincerity Scores by Attitude Similarity, Evaluation, and Order of Measures. Source of Variation gg_ M§' §_ Attitude Similarity (A) 3 8.412 5.024* Evaluation (E) 1 8.461 5.053** Order of Measures (0) l .614 .367 A x E 3 1.865 1.114 A x 0 3 1.607 .960 E x 0 1 1.842 1.100 A x E x O 3 1.487 .888 l-lithin Groups Error 209 1.674 *p_ < .002 “p < .03 The same pattern of results was obtained in a similar analysis of variance on believability scores (Table 6). Again, significant main effects for attitude similarity (means: 1 - 5.73, 2/3 - 5.31, 24 Table 6.--Analysis of Variance on Believability Scores by Attitude Similarity, Evaluation, and Order of Measures. Source of Variation g 95 E Attitude Similarity (A) 3 16.207 7.141* Evaluation (E) 1 15.392 6.781** Order of Measures (0) 1 6.005 2.646 A x E 3 1.691 .745 A x O 3 1.291 .569 E x O 1 1.469 .647 A x E x O 3 .284 .125 Within Groups Error 209 2.270 *2 < .001 "p < .01 1/3 - 4.88, and 0 - 4.45; scale range = 1-7) and evaluation (positive - 4.79 versus no evaluation - 5.34) were obtained. Enhanced attributions of sincerity and believability might be expected as natural covariates of the attraction induced by attitude similarity. However, reduced sincerity and believability scores in the positive evaluation conditions suggest that these variables are not linked with attraction in such a simple fashion. Otherwise, positive evaluations would have produced relatively higher scores on the sincerity and believability measures. These data could be inter- preted as more evidence for the agreement - evaluation confounding. CHAPTER IV DISCUSSION Replication Byrne's reinforcement model of attraction (Byrne, 1969, 1971; Byrne and Rhamey, 1965; Byrne and Griffitt, 1966; Byrne and Clare, 1967, 1970; Clare and Byrne, 1974) predicts additive effects for agreement and personal evaluations. As Kaplan and Anderson (1973a) have indicated, the Byrne and Rhamey study, which reported a signifi- cant interaction between attitude similarity and evaluation, repre- sents a failure of the additive model. Data from the present investi- gation do fit such an additive model when subjected to the same analysis used by Byrne and Rhamey and, thus, provide a conceptual replication of their study. However, given the reasonably convincing evidence for a confounding of the similarity and evaluation variables, it is apparent that such analyses of variance are inappropriate with these data. Confounding The high positive correlations between perception of agreement and expected evaluation along with the evidence for differential perceptions of each of the independent manipulations across the levels of the other point to the conclusion that previous interpretations 25 26 resulting in the assignment of three to one weights for evaluation versus agreement in the prediction of attraction were in error. These weighting coefficients resulted from a statistical artifact which served to confuse the relationship among agreement, personal evaluations, and attraction. The findings of the present study suggest that merely pro- viding for statistical orthogonality, assigning subjects to factorial combinations of independently manipulated variables, does not ensure the psychological independence of these variables. The positive relationship between perception of agreement and expected liking was obtained in spite of the fact that the levels of attitude similarity and evaluation were orthogonalized. Sincerity and Believability The main effects for attitude similarity on these two measures may be indicative of a general expectation on the part of college students that fellow students share their attitudes (Williams, 1975). Possibly, violation of this expectation resulted in less acceptance of the information fromnthe strangers. The main effects for evalu- ation provide additional circumstantial evidence for the confounding of the attitude similarity and evaluation variables. It is likely that the combination of such extremely positive evaluations and some- what less compelling attitude information provided most subjects with an inconsistent stimulus situation resulting in suspicion. 27 ggplied Evaluation As predicted, the preponderance of evidence supports the notion that the attitude similarity - attraction relationship is mediated by implied evaluation, with agreement having a significant but trivial direct effect on attraction. Attitude similarity acts as a cue to a person that the agreeing other will positively evaluate or like the person or that a disagreeing other will be less likely to be favorably disposed toward the person. The person's response of either attraction or repulsion to the other is based on this expectation of liking or disliking from.the other. In simple terms, we like others who agree with us because we expect them to like us, and conversely, we dislike disagreeing others because of an assumption that they will dislike us. Our reason for being attracted to other persons whom.we expect to like us may be that we expect these persons to gratify our needs. Centers (1975) pointed out that the need for consensual validation is not an adequate and sufficient explanation for all attraction between persons. The present research suggests that the attraction induced by similarity of attitudes is neither sufficiently nor adequately explained by a need for consensual validation. Rather, the agreement - attraction relationship is better explained by an expectation of mutual need gratification based on reciprocal liking, where the particular needs to be gratified are determined by the persons involved in the situation. This formulation suggests an idiographic approach to the attitude similarity - liking phenomenon in which the presentation of information indicating the likelihood of gratification of a person's predetermined needs would account for a greater proportion of 28 variance in attraction than either agreement or personal evaluations. The Byrne, London, and Griffitt (1968) study, which reported an effect for topic importance on agreement induced attraction, provides evi- dence that such individual differences may be very important deter- minants of liking. ggplications for Further Research As a way of understanding the differential importance of antecedents of interpersonal attraction, it seems reasonable to assume that people attempt to use the best possible information available to tha when they are judging others. Research in this area should be directed at the identification of people's most preferred categories of information and then, toward an understanding of how that infor— mation which is provided is utilized. It is probably not a bad guess that these forms of data that are most personally relevant to the individual (e.g., evaluations of self and indications of mutual reward) are among the most preferred judgmental cues. It is also likely that, in the absence of preferred information, those cues that are supplied are used as imperfect indicators of the desired knowledge. Just such relationships have been identified previously between race and values (Hokeach, Smith, and Evans, 1960) and between physical attractiveness and expected acceptance of a date (Herscheid, Dion, walster, and Halster, 1971). The present investigation has demonstrated that a similar pattern exists between attitude similarity and expected liking. Future research should be aimed at discovering the relative potency and causal links among those variables which appear to have 29 effects on people's attraction to others. As indicated in the present research, researchers should take care not to confuse statistical orthogonality with psychological independence. W8 n I»... FOOTNOTES 1In a later investigation, Byrne, London, and Griffitt (1968) found a topic importance effect using a different design. However this was subsequent to the developments which led to the Byrne and Rhamey (1965) study of the effects of personal evaluations and attitude similarity. 2Throughout this study, the term.confound refers to a specification confound (Crane and Brewer, 1973, p. 27) which involves problems in the correct identification of causal factors. In this particular case, the effects of agreement cannot be separated from those of personal evaluations. 3The Byrne and Rhamey study included a negative evaluation group. for the purposes of the present study, it was felt that the possible harmful effects of informing a number of undergraduate sub- jects that each was considered by a peer to be unintelligent, behind the times, immoral, slightly maladjusted, and that the peer would probably dislike the subject very much and find the subject to be an undesirable work partner, far outweighed the possible benefits to be derived from.such a manipulation. It should be noted that the partial replication accomplished for this study allows for the necessary comparisons. 30 REFERENCES Aronson, E. Some antecedents of interpersonal attraction. In W. J. Arnold S D. Levine (Bds.), Nebraska symposium.on motivation: 1969 (vol. 17). Lincoln, Nebraska: University of Nebraska Press, 1970. Aronson, B., & Linder, D. Gain and loss of esteem as determinants of interpersonal attractiveness. Journal of Experimental Social Aronson, 8., & Horchel, P. Similarity versus liking as determinants of interpersonal attractiveness. Psychonomdc Science, 1966, £5 157-158. Backman, C. W., 8 Secord, P. F. The effect of perceived liking on interperson attraction. Human Relations, 1959, $3! 379-384. Berscheid, 8.: Dion, X.) walster, 3.: & Walster, G. W. Physical attractiveness and dating choice: A test of the matching hypothesis. JOurnal of Experimental Social Psychology, 1971, 1' 173-189 e Berscheid, B. , 8 walster, B. Interpersonal attraction. Reading, Mass.: Addison-Wesley, 1969. Brewer, R. B., & Brewer, N. B. Attraction and accuracy of perception in dyads. Journal of Personality and Social Psychology, 1968, g, 188-193. Byrne, D. Interpersonal attraction and attitude similarity. Journal of Abnormal and Social Psychology, 1961, 93, 713-715. Byrne, D. Response to attitude similarity - dissimilarity as a function of affiliation need. Journal of Personality, 1962, 83, 164-177 . Byrne, D. Attitudes and attraction. In L. Berkowits (Bd.), Advances in experimental social psycholggy (vol. 4). New York: Academic Press, 1969, 35-89. Byrne, D. The attraction paradigg, New York: Academic Press, 1971. 31 Byrne, Byrne 3 Byrne, Byrne: mm: Byrne . Byrne, m0. Centers, R. Clare, 32 D., 8 Clore, G. L. Predicting interpersonal attraction toward strangers presented in three different stimulus modes. Psychonomic Science, 1966, 2, 239-240. D., 8 Clore, G. L. Effectance arousal and attraction. Journal of Personality and Social Psychology, 1967, 6. (4, Whole No. 638). D., & Clore, G. L. A reinforcement model of evaluative responses. Personality: An International Journal, 1970, 1, 103-128. D.: Clore, G. L.; Griffitt, W.: Lamberth, J.: & Mitchell, K. B. when research paradigms converge: Confrontation or integration? Journal of Personality and Social Psychology, 1973a, 38, 313-320. D.: Clore, G. L.: Griffitt, W.) Lamberth, J.: 5 Mitchell, H. B. One more time. Journal of Personality and Social Psychology, 1973b, 38, 323-324. D., S Griffitt, W. Similarity versus liking: A clarification. Psychonomic Science, 1966, g, 295-296. D., s Griffitt, W. Interpersonal attraction. In P. M. Mussen a M. R. Rosenzweig (Eds.), Annual review of psychology (Vel. 24). Palo Alto: Annual Reviews Inc., 1973. D., & Lamberth, J. Reinforcement theories and cognitive theories as complementary approaches to the study of attraction. In B. I. Murstein (Bd.), Theories of love and attraction. New York: Springer, 1971. D.: London, 0.: & Griffitt, M. The effect of topic importance and attitude similarity-dissimilarity on attraction in an intrastranger design. Psychonomic Science, 1968, 11, 303- 304. D., 8 Nelson, D. Attraction as a function of attitude similarity-dissimilarity: The effect of topic importance. Psychonomic Science, 1964, 1, 93-94. D., & Rhamey, R. Magnitude of positive and negative reinforce- ments as determinants of attraction. Journal of Personality and Social Psycholggy, 1965, 3, 884-889. Sexual attraction and love: An instrumental theory. Springfield, Illinois: Charles C. Thomas, 1975. G. L., S Baldridge, B. attitude-attraction research. Psycholggy, 1970, 9, 177-186. The behavior of item weights in Journal of Experimental Social 33 Clore, G. L., e Byrne, D. A reinforcement-affect model of attraction. In T. L. Huston (Ed. ) , Foundations of interpersonal attraction. New York: Academic Press, 1974. Crane, w. D., & Brewer, M. B. Principles of research in social psychology. New York: McGraw-Hill, 1973. Gouaux, C. , & Smers, K. Interpersonal attraction as a function of affective state and affective change. Journal of Research in Personality, 1973, 1, 254-260. Insko, C. A.: Thompson, V. D.; Stroebe, W.) Shaud, K. F.; Pinner, B. E.; & Layton, B. D. Implied evaluation and the similarity attraction effect. Journal of Personality and Social Psycholgy, 1973, .2_5, 297-308. Johnson, C. D. Competence motivation and interpersonal evaluation. Bulletin of the Psychonomic Society, 1974, _4_, 199-200. Raplan, M. P. , 5 Anderson, N. H. Information integration theory and reinforcement theory as approaches to interpersonal attraction. Journal of Personality and Social Psychology, 1973a, 23, 301-312. Kaplan, M. P., & Anderson, N. H. Cement on "When research paradigms converge : Confrontation or integration?” Journal of Personality and Social Psychology, 1973b, 32, 321-322. McNhirter, R. M., & Jecker, J. D. Attitude similarity and inferred attraction. Psychonomic Science, 1967, 1, 225-226. Nelson, D. Attitude similarity and interpersonal attraction: The approval-cue hypothesis. Paper presented at the meeting of the Southwestern Psychological Association, Arlington, 1966. Newooeb, T. M. The actmaintance process. New York: Bolt, Rinehart, a Ninston, 1961. Rokeach, M.: Smith, P. It; 5 Evans, R. I. Two kinds of prejudice or one? In M. Rokeach (Ed.), The open and closed mind. New York: Basic Books, 1960. Singh, R. Reinforcement and attraction: Specifying the effects of affective states. Journal of Research in Personality, 1974, 8, 294-305. Singh, R. Reinforcement, affect and interpersonal attraction. 34 Stephan, C. Attribution of intention and perception of attitude as a function of liking and similarity. Sociometry, 1973, 66, 463-475. Thibaut, J. W., & Kelley, H. H. The socia1_psygholggy of groups, New York: John Wiley 5 Sons, Inc., 1959. Touhey, J. C. Interpersonal congruency, attitude similarity, and interpersonal attraction. Journal of Research in Personality, 1975, 2, 66-73. Walster, E., & Walster, G. W. The effects of expecting to be liked on choice of associate. Journal of Abnormal and Social Psycholggy, 1963, 61, 402-404. White, R. W. Motivation reconsidered: the concept of competence. Psychological Review, 1959, 66, 297-333. Williams, D. C. Race and beliefs as determinants of perceived belief similarity. Unpublished doctoral dissertation, Michigan State University, 1975. APPENDICES APPENDIX A SURVEY OF ATI‘ITUDES APPENDIX A SURVEY OF ATTITUDES Please answer every question 1. Strict Discipline (check one) am very much against strict disciplining of children am.against strict disciplining of children am.mildly against strict disciplining of children an mildly in favor of strict disciplining of children am.in favor of strict disciplining of children am very much in favor of strict disciplining of children HHHHHH lllll N . College Education (check one) I strongly believe that it is very important for a person to have a college education in order to be successful I believe that it is very important for a person to have a college education in order to be successful I believe that perhaps it is very important for a person to have a college education in order to be successful I believe that perhaps it is not very important for a person to have a college education in order to be successful I believe that it is not very important for a person to have a college education in order to be successful I strongly believe that it is not very important for a person to have a college education in order to be successful 0 . Belief in God (check one) I strongly believe that there is a God I believe that there is a God I believe that perhaps there is a God I believe that perhaps there is no God I believe that there is no God I strongly believe that there is no God Hllll YOUR.NAHE 35 36 4. Political Parties (check one) U‘ s m I am a strong supporter of the Democratic party I prefer the Democratic party I have a slight preference for the Democratic party I have a slight preference for the Republican party I prefer the Republican party I am a strong supporter of the Republican party Smoking (check one) am very much in favor of smoking am in favor of smoking am mildly in favor of smoking am mildly against smoking am against smoking am very much against smoking In general, In general, In general, In general, In general, In general, HHHHHH . Financial Help from Parents (check one) I strongly believe that parents should provide financial help to young married couples I believe that parents should provide financial help to young married couples I feel that perhaps parents should provide financial help to young married couples I feel that perhaps parents should not provide financial help to young married couples I believe that parents should not provide financial help to young married couples I strongly believe that parents should not provide financial help to young married couples 7. Social Aspects of College Life (check one) In general, I am very much against an emphasis on the social aspects of college life In general, I am against an emphasis on the social aspects of college life In general, I amhmildly against an emphasis on the social aspects of college life In general, I amLmildly in favor of an emphasis on the social aspects of college life In general, I am in favor of an emphasis on the social aspects of college life In general, I am very much in favor of an emphasis on the social aspects of college life 37 8. Premarital Sex Relations (check one) In general, I am very much opposed to premarital sex relations In general, I am opposed to premarital sex relations In general, I am mildly opposed to premarital sex relations In general, I am mildly in favor of premarital sex relations In general, I am in favor of premarital sex relations In general, I am very much in favor of premarital sex relations 9. Creative Werk (check one) enjoy doing creative work very much enjoy doing creative work enjoy doing creative work to a slight degree dislike doing creative work to a slight degree dislike doing creative work dislike doing creative work very much HHHHHH H 0. Classical Music (check one) dislike classical music very much dislike classical music dislike classical music to a slight degree enjoy classical music to a slight degree enjoy classical music ‘ enjoy classical music very much HHHHHH p... l. Welfare Legislation (check one) am very much opposed to increased welfare legislation am opposed to increased welfare legislation amxmildly opposed to increased welfare legislation am mildly in favor of increased welfare legislation am in favor of increased welfare legislation am very mach in favor of increased welfare legislation HHHHHH 12. Foreign Movies (check one) I enjoy foreign movies very much I enjoy foreign movies I enjoy foreign movies to a slight degree I dislike foreign movies to a slight degree I dislike foreign movies I dislike foreign movies very much Hllll APPENDIX B INTERPERSONAL JUDGMENI‘ SCALE APPENDIX B INTERPERSONAL JUDGMENT SCALE Please use the following six questions to rate the person about whom you have just read. 1. Intelligence (check one) I believe that this person is very much above average in intelligence I believe that this person is above average in intelligence I believe that this person is slightly above average in intelligence I believe that this person is average in intelligence I believe that this person is slightly below average in intelligence I believe that this person is below average in intelligence I believe that this person is very much below average in intelligence Knowledge of Current Events (check one) I believe that this person is very much below average in his/her knowledge of current events I believe that this person is below average in his/her knowledge of current events I believe that this person is slightly below average in his/ her knowledge of current events I believe that this person is average in his/her knowledge of current events I believe that this person is slightly above average in his/her knowledge of current events I believe that this person is above average in his/her knowledge of current events I believe that this person is very much above average in his/her knowledge of current events ‘YOUR.NAHB 38 39 3. Morality (check one) This person impresses me as being extremely moral This person impresses me as being moral This person impresses me as being moral to a slight degree This person impresses me as being neither particularly moral nor particularly immoral This person impresses me as being immoral to a slight degree This person impresses me as being immoral This person impresses me as being extremely immoral b . Adjustment (check one) I believe that this person is extremely maladjusted I believe that this person is maladjusted I believe that this person is maladjusted to a slight degree I believe that this person is neither particularly maladjusted nor particularly well adjusted I believe that this person is well adjusted to a slight degree I believe that this person is well adjusted I believe that this person is extremely well adjusted llll 5. Personal Feelings (check one) I feel that I would probably like this person very much I feel that I would probably like this person I feel that I would probably like this person to a slight degree I feel that I would probably neither particularly like nor particularly dislike this person I feel that I would probably dislike this person to a slight degree I feel that I would probably dislike this person I feel that I would probably dislike this person very much O . Working Together in an Experiment (check one) I believe that I would very much dislike working with this person in an experiment I believe that I would dislike working with this person in an experiment I believe that I would dislike working with this person in an experiment to a slight degree I believe that I would neither particularly dislike nor particularly enjoy working with this person in an experiment I believe that I would enjoy working with this person in an experiment to a slight degree I believe that I would enjoy working with this person in an experiment I believe that I would very much enjoy working with this person in an experiment APPENDIX C EXPECTED LIXING AND PERCEPTION OP AGREEIENT HEASURES APPENDIX C EXPECTED LIKING AND PERCEPTION OF AGREEMENT MEASURES PLEASE DO NOT REREAD EARLIER SECTIONS OF THIS BOOKLET Answer the next two questions on the basis of the information you have received about this other person. We are interested in how 1' cu think this other person would evaluate you. Personal feelings (check one) This person would probably like me very much. This person would probably like me. This person would probably like me to a slight degree. This person would probably neither particularly like nor particularly dislike me. This person would probably dislike me to a slight degree. This person would probably dislike me. This person would probably dislike me very much. Working together in an experiment (check one) I believe that this person would very much dislike working with me in an experiment. I believe that this person would dislike working with me in an experiment. I believe that this person would dislike working with me in an experiment to a slight degree. I believe that this person would neither particularly dislike nor particularly enjoy working with me in an experiment. I believe that this person would enjoy working with me in an experiment to a slight degree. I believe that this person would enjoy working with me in an experiment. I believe that this person would very much enjoy working with me in an experiment. I!!! .Q.Oi...Ottttiitfiiifiititttitittifliiifltiitmfiitittiifitiimtifliiiiittiiit. 4O 41 How much do you feel that you and this other person would be in agree- ment on the Survey of Attitudes which you completed recently? (check one) complete disagreement disagreement slight disagreement neither disagreement nor agreement slight agreement agreement complete agreement APPENDIX D SINCERITY AND BELIEVABILITY MEASURES APPENDIX D SINCERITY AND BELIEVABILITY MEASURES we would like your estimate of the other person's sincerity. Having read all of the information supplied, do you feel that this person was: very sincere sincere slightly sincere neither sincere nor insincere slightly insincere insincere very insincere found the information to be: H very unbelievable unbelievable slightly unbelievable neither unbelievable nor believable slightly believable believable very believable 42 II I II I”!!! I” I"! III Will H“ I!!! III! III! II I!!! III!!! II I!!! (ll 00453707