_ -__r_—r-w “mu-m— CGMMETMENT, SELF-:EVALUATIW AND COMMQNICAWON AeflVSTY EN A M$$QNANT fii'FUAHGN: A STEPEW Q? mama»: EMBENTE WETH ENGLWH LANGUAGfi DEFECIENCY That: for “to Damn elf pk. D._ MICHIGAN STATE UNIVERSE“ Yoahihiro Akutsu '2 9 6 9 —T We umvsn ,T ' ' A! .‘ W’! 31'! '.?:’f"""* * q. ‘v III/INTI”! , 5w} ' !C m) \- 11117 "1 ‘~‘ Jiiy 1. '- -.~ MVP M/‘CHIGA W17] 3 M 129 W THES‘S K‘- m. Q- m “J This is to certify that the thesis entitled COMMITMENT, SELF-EVALUATION AND CCMMUNICA'I‘ION ACTIVITY IN A DISSONANT SITUATION: i A STUDY OF FOREIGN STUDENTS WITH . ENGLISH LANGUAGE DEFICIENCY : presented by Yoshihiro Akutsu has been accepted towards fulfillment of the requirements for Ph 0 D 0 degree. in COmmunica t 1011 fiflt/Z" ( Major professor _. __._ ._.-_ _ ~‘ 0-169 - -—. ~¢-—. ‘~.-,., --—-—~ H .. n, ,7,“ ABSTRACT COMMITMENT, SELF-EVALUATION AND COMMUNICATION ACTIVITY IN’A DISSONANT SITUATION: A STUDY OF FOREIGN STUDENTS WITH ENGLISH LANGUAGE DEFICIENCY By Yoshihiro Akutsu The Problem. A primary objective of this study is to determine the relation of ambiguity of commitment and positiveness of self-evaluation to selectivity of participation in potentially dissonance-increasing communication activities. The two independent variables, ambiguity of commitment and positiveness of self-evaluation, are employed with an attempt to explain the disconfirmation of tendencies to avoid exposure to dissonance-increasing sources. The following hypotheses are tested with foreign students who are actually committed to study at U.S. universities and who have objective language deficiency. For such foreign students,participation in communication activities which requires some English language proficiency is potentially dissonance-increasing because in such participation they may experience their language deficiency and thus such experience would add further cognitive elements dissonant with their commitment. Hypothesis 1: Of the foreign students with unambiguous commitment to study at U.S. universities, those Yoshihiro Akutsu with more positive self-evaluations are less selective in their participation in social communication activities than those with less positive self-evaluations. Hypothesis 11: 0f the foreign students with ambiguous commitment to study at U.S. universities, those with more positive self-evaluations are less selective in their participation in social communication activities than those with less positive self-evaluations. Hypothesis III: 0f the foreign students with more positive self-evaluations, those with unambiguous commitment to study at U.S. universities are less selective in their participation in social communication activities than those with ambiguous commitment. Hypothesis IV: 0f the foreign students with less positive self-evaluations, those with unambiguous commitment to study at U.S. universities are more selective in their participation in social communication activities than those with ambiguous commitment. Method. 572 foreign students who were enrolled in full time intensive English at ten universities comprise the subjects in this study. These subjects were relatively homogenized by sex, marital status, age, length of stay in the United States, and home country, with the resultant number of 275. The questionnaire method was used to gather the data. The questionnaires were administered at ten Yoshihiro Akutsu universities during the period of April 21 to May 21, 1969. The measures of three variables, positiveness of self-evaluation, ambiguity of commitment, and selectivity of participation in communication activities, are determined by the semantic differential technique, the questionnaire items dealing with the reality to which subjects are committed, and the Guttman scale of participation in communication activities, respectively. For analyzing the data, the Mann-Whitney U test is used. Results and Conclusions. The data point to the conclusion that positiveness of self-evaluation is negatively related to selectivity of participation in potentially dissonance-increasing communication activities. However, the evidence for the negative relationship between positiveness of self-evaluation and selectivity of participation within the ambiguous commitment group is weak as compared with the evidence for that relationship within the unambiguous commitment group. This may indicate that the effect of commitment on the resistance to change of the corresponding cognitive element is lessened if the reality to which the commitment corresponds is ambiguous. The evidence for the positive relationship between ambiguity of commitment and selectivity of participation in communication activities within the more positive self-evaluation group and for the negative relationship betwe is qu direc obvio part1 commu of se Yoshihiro Akutsu between them within the less positive self-evaluation group is quite weak. But the data show generally the predicted directions of relationships. we must come to the conclusion that it is obviously incorrect to assert a simple avoidance of participation in potentially dissonance-increasing communication activities without regard to positiveness of self-evaluation and ambiguity of commitment. COMMITMENT, SELF-EVALUATION AND COMMUNICATION ACTIVITY IN’A DISSONANT SITUATION: A STUDY OF FOREIGN STUDENTS WITH ENGLISH LANGUAGE DEFICIENCY By Yoshihiro Akutsu A THESIS Submitted to ZMichigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Communication 1969 Con Ste rec Gui Accepted by the faculty of the Department of Communication, College of Communication Arts, Michigan State University, in partial fulfillment of the requirements for the Doctor of Philoso hy degree. M5: 3 Director of ' si Guidance Committee: 7 Y - Chairman mix/EWW 4 fit? -LW m flogg- ii To my father, who died in my absence from Japan. 111 II t} Ur su co f0 th to CD Na 00 ACKNOWLEDGMENTS It is a great pleasure to acknowledge my indebtedness to Dr. Hideya Kumata, my major professor in the course of my graduate trainning at Michigan State University, for his long encouragement and generous support of my efforts, for the most s0phisticated conceptual and empirical criticisms he has provided, and for his reading of the draft again and again through the various stages of the preparation. I would like to express my profound appreciation to members of my Guidance Committee, Dr. Bradley S. Greenberg, Dr. Everett M. Rogers, and Dr. Frederick B. Haisanen, for their helpful and constructive comments on this study which led to many corrections and improvements, for which I am grateful. I have greatly benefited from discussions with Prof. Shigeo Imamura, Director of the English Language Center of Michigan State University, and further I am indebted to him for his great help in collecting the data needed in this study. I am greatly indebted to the following persons for their assistance in surveying foreign students iv 11'. th Ur. su c0} fo: the to Gre Wai 0n imp Pro. Gem inde date fer ACKNOWLEDGMENTS It is a great pleasure to acknowledge my indebtedness to Dr. Hideya Kumata, my major professor in the course of my graduate trainning at Michigan State University, for his long encouragement and generous support of my efforts, for the most saphisticated conceptual and empirical criticisms he has provided, and for his reading of the draft again and again through the various stages of the preparation. I would like to express my profound appreciation to members of my Guidance Committee, Dr. Bradley S. Greenberg, Dr. Everett M; Rogers, and Dr. Frederick B. Whisanen, for their helpful and constructive comments on this study which led to many corrections and improvements, for which I am grateful. I have greatly benefited from discussions with Prof. Shigeo Imamura, Director of the English Language Center of Michigan State university, and further I am indebted to him for his great help in collecting the data needed in this study. I am greatly indebted to the following persons for their assistance in surveying foreign students iv under their jurisdictions, without which this study could not have been done: Prof. Allis R. Bens, Director of American Language Institute, San Francisco State College; Prof. John Bordie, Associate Director of Foreign Language Education Center, University of Texas; Dr. Edward T. Erazmus, Director of Intensive English Center, the University of Kansas; Dr. Gordon Ericksen, Director of the American Language Institute, New York University; Dr. Joseph H. Friend, Director of Center for English as a Second Language, Southern Illinois University; Dr. Robert B. Kaplan, Director of English Communication Program for Foreign Students, University of Southern California; Dr. Leonard R. Marelli, the American Language Institute, New York University; Prof. Robert Parslow, Coordinator of English Language Institute, University of Pittsburgh; Prof. Theodore H. Plaister, Chairman of the Department of English as a Second Language, University of Hawaii; Dr. Clifford H. Prator, Vice Chairman of the Department of English, University of California, Los Angeles; Prof. John H. Rogers, Director of Intensive English Program, University of Miami; and Dr. Andrew H. Yarrow, Director of English Language and Orientation Program, Louisiana State University. My debt is also to a number of anonymous individuals who administered the questionnaire to the subjects and to the foreign students whose cooperation furnished the data on which this study isb tom Univ! Mich for Dr. is based. In connection with the data collection, I wish to mention my gratitude to Dr. Donald Knapp of Columbia University, Dr. George E. Luther of the University of Michigan and Dr. Robert L. Saitz of Boston University, for their special interests in this study. I am also grateful to Dr. Mitoji Nishimoto and Dr. Takeo Furu, who encouraged my completing the study. vi TABLE OF CONTENTS INTRODUCTION 0 O O O O O O 0 0 O O O O O O 0 Chapter I. THE PROBLEM O O O O O O O O O O O O I‘Dissonance Theory and Self Theory Hypotheses A . Review of Literature IL mmw............... Subjects Measurement of Variables Data Gathering Data Analysis III. RESULTS 0 O O O O O O O O O O O O 0 IV. INTERPRETATION AND CONCLUSIONS . . . EIBLIOGRAPHY................ APPEme O. O O 0 O O O O O O O O O O O O O 0 vii 34 52 69 78 84 Table l. 10. 11. LIST OF TABLES Page Distribution of 572 subjects among the ten universities chosen. . . . . . . . . . . . . . Distribution of 572 subjects among sex and marital status . . . . . . . . . . . . . . . . Male-single subjects by age. . . . . . . . . . . Distribution of 296 subjects homogenized by sex, marital status, age, and length of stay among regional clusters of home countries. . . . . . . . . . . . . . . . . . . Number of male-single subjects within each regional cluster of home countries . . . . . . Guttman scale analysis: participation in communication activities . . . . . . . . . . . Distributions of the subjects on the commitment- ambigUity score 0 O O O O O O O O O O O O O O O Distributions of the subjects on the self- positiveness score . . . . . . . . . . . . . . Distributions of the subjects on the Guttman scale score of participation . . . . . . . . . Distribution of the subjects with unambiguous commitment/high or low self-positiveness on the Guttman score of participation . . . . . . Distributions of the subjects with unambiguous commitment/high or low self-positiveness on the Guttman score of participation for the three regional categories. . . . . . . . . . . viii 35 46 46 49 52 54 55 55 56 58 S9 Table 12. 13. 14. 15. 16. 17. Distribution of the subjects with ambiguous commitment/high or low self-positiveness on the Guttman score of participation . . . Distributions of the subjects with ambiguous commitment/high or low self-positiveness on the Guttman score of participation for the three regional categories . . . . . . . Distribution of the subjects with high self-positiveness/unambiguous or ambiguous commitment on the Guttman score of participation . Distributions of the subjects with high r self-positivenesslunambiguous or ambiguous commitment on the Guttman score of participation for the three regional categories. Distribution of the subjects with low self-positivenesslunambiguous or ambiguous commitment on the Guttman score of participation . Distributions of the subjects with low self-pasitiveness/unambiguous or ambiguous commitment on the Guttman score of participation for the three regional categories. ix 0 Page COED! witl CID not ass in V3! has es; Ce- SOB Ste "C' .4 .C\Un. 0.. par. % m\%\ INTRODUCTION After WOrld war II, an increase in international communication, travel, and contact has been accompanied with a steady growth of empirical research on problems of cross-cultural communication. Such research might help not only in testing the generality and limitations of assumptions developed in one cultural setting but also in finding the hitherto unanticipated interactions among variables.1 One of the areas in which much research has been stimulated is the study of foreign students, especially in the United States. It might be assumed that language competence is a sine qua 223 for extensive communication activity. In some studies involving foreign students in the United States and American students abroad,2 it was found that 1E. Jacobson, H. Kumata, and J. E. Gullahorn, "Cross-Cultural Contributions to Attitude Research," Public Opinion Quarterly, 24(1960), 204-23. - 2R. T. Morris, The Two-Way Mirror: National Status in Foreign Students' Adjustment (Minneapolis: Univ. of Minnesota Press, 1960); W. H. Sewell and O. M. Davidsen, Scandinavian Students on an American Campus (Minneapolis: Univ. of Minnesota Press, 1961); J. E. Gullahorn and J. T. Gullahorn, "American Students Abroad: Professional versus Personal Development," The Annals of the American Academy of Political and Social Science, 368(1966), 43-59. ’ l 2 language competence is important for meaningful interaction between foreign students and host nationals. In some other studiea’,3 however, it was pointed out that a foreign student's confidence in his language ability has a more important influence on the development of social relations than his actual command of the language. The subjective assessment of language competence may include not only confidence in language ability but also efforts at realistic self-evaluation. As Gullahorn and Gullahorn have suggested, "confidence in language ability may be a special instanCe of a more general confidence in one's ability to interact with others in a rewarding manner."4 Many positive evaluations of self seem to give the individual a feeling of general confidence. Research evidence concerning the relationship between a foreign student's confidence in his language ability and its influence on the development of social relations require fhrther study. Foreign students with some language proficiency, even if the level of language proficiency is relatively low, may be able to participate in communication activities 3C. Selltiz, J. R. Christ, J. Havel, and s. w. Cook Attitudes and Social Relations of Forei n Students ig_t e United States (Minneapolis: Univ. oflMinnesota Press, 19635; Gullanrn and Gullahorn, “American Students Abroad." 4GuIIahom and Gullahorn, "American Students Abroad," 49. 3 with some effectiveness. Such communication experiences could lead to the perception of language skills as satisfactory, so that they might feel confident in their language skills. However, those with objective deficiency in language skills may be unable to participate in communication activities with the same effectiveness as those with some proficiency may have. For these students, therefore, communication experiences would not lead to the perception of language skills as satisfactory. In previous studies, Such levels of objective English proficiency have not been controlled. Pool characterized the foreign student as follows: The foreign student is a purposive traveler, usually responsible to someone at home for accomplishing certain objectives . . . . He is ; willing if nOt an eager traveler; Often but not always with unrealistic hOpes. Usually intelligent and ambitious, he suffers some competitive handicaps in winning recognition in an unfamiliar environment, . . .5 This notion suggests that objective language deficiency is a serious problem for foreign students who are committed to study at U.S. universities. Foreign students matriculating in either graduate or undergraduate programs at 5I. de 8. Pool, "Effects of Cross-National Contact on National and International Images," in International Behavior, ed. by H. C. Kelman (New York: 'Holt, Rinehart and Winston, 1965), p.111. 4 universities or colleges are screened for their English proficiency. If found deficient in their command of English, they are usually required to correct the deficiency before starting their programs by enrolling in full time intensive English. Such students are often advised to participate in communication activities in order to develOp their skills of English. According to dissonance theory,6 however, it can be assumed that for such foreign students, participation in communication activities which requires some English proficiency is potentially dissonance- increasing because in such participation they may experience their language deficiency and then this experience would add further cognitive elements that are dissonant with the English language requirement for their academic work at U.S. universities. Thus they would tend to avoid participation in such potentially dissonance-increasing communication activities. 0n the other hand, it also seems reasonable to suppose that those students who have more positive self-evaluations will actively participate in communication activities so as to help in .the development of their language skills. Such participation would be perceived by the "positivé’ student as challenging opportunities to test his mettle. 6L. Festinger, A Theo of Co nitive Dissonance (Stanford, Calif.: Stafifdffi'UETVT‘TfiEES, 1957) 5 This problem requires study involving objective assessments of language proficiency and measures of self- evaluation, commitment to study at U.S. universities, and participation in a variety of social communication activities. CHAPTER I THE PROBLEM On American campuses, are foreign students with objective English language deficiency likely to participate in a variety of social communication activities? How is their commitment to study at American universities or colleges related to this participation? Do those who are unambiguously committed to study tend to be less selective in their participation in social communication activities than those who are ambiguously committed? Furthermore, how do their self-evaluations influence this participation? Do those who have more positive self-evaluations tend to I be less selective in their participation in these activities? On the other hand, do those with less positive self-evaluations tend to be more selective in their participation in these activities? A primary objective of this study is to determine the relation of ambiguity of commitment and positiveness of self-evaluation to selectivity of participation in communication activities in a dissonant situation. The two independent variables, ambiguity of commitment and positiveness of self-evaluation, are employed 6 7 with an attempt to draw together two research traditions, dissonance studies and self studies. The insights to be gained from linking dissonance theory with self theory should help explain the disconfirmation of tendencies to avois dissonance-increasing information in a dissonance reduction process and should in turn become a part of the general fund of knowledge about individual behavior of participation in communication activities. FUrther, a question has been raised concerning the manipulation of commitment in dissonance studies, that is, the adequacy of the dissonance formulation for understanding the consequences of commitment in important life situations. The manipulation of commitment has been achieved almost entirely through instructional procedures in previous studies. Critical is whether the subject is influenced by these procedures. A real-life situation should be sought for its advantages in testing the dissonance formulation of commitment. Therefore, this study is done with those foreign students who are actually committed to study at American universities or colleges and who have objective English deficiency-~which is dissonant with their commitment. I In studies of foreign students, specific ideas and directions for future research have been brought to our attention. However, specific findings from studies of fereign students are seldom related to or integrated 8 with existing theoretical positions. What is needed in research in this area is some sort of theoretical rationale for hypotheses and predictions. In this sense, this study should also help in predictions of communication behavior of foreign students with objective language deficiency. Dissonance Theory and Self Theory According to Festinger's theory of cognitive dissonance,1cognitive elements or bits of knowledge which may be related to oneself, to one‘s behavior, or to the environment exist or may exist in relation to one another. When the obverse of one element follows from that element, dissonance arises. The term "dissonance" refers to "the existence of nonfitting relations among cognitions."2 For those foreign students who are required to correct English deficiency before starting their academic work, the two cognitions, the English language requirement for academic work at a university and English language deficiency, are in a dissonant relation. It seems reasonable to consider English language deficiency as the obverse of the English language requirement. According to Festinger's analysis, dissonance is aroused in post-decisiOnal situations. If the person has 1'L. Festinger, A Theory of Cognitive Dissonance (Stanford, Calif.: StanffirdIUhiv. Press, 1957). ZIbIdo’ P030 A: 9 already made a decision between two or more alternatives, he is post-decisional. Foreign students are clearly post- decisional because they have already decided to study in the United States rather than in their home countries or other foreign countries. Brehm and Cohen, however, emphasized the role of commitment in generating dissonance. They concluded that "under conditions where commitment is present, dissonance does clearly occur and produce its effects on cognition and, perhaps, on perception or behavior."3 Festinger also accepted this emphasis on commitment and, with respect to its conceptual meaning, noted that "a decision carries commitment with it if the decision unequivocally affects subsequent behavior."4 If a foreign student has decided to study at a U.S.fl university and now he is making arrangements in accordance with his decision, we can say that he is committed to the behavior of studying at a U.S. university. Thus if he is not permitted to start his study in spite of this commitment because he does not have the required English language proficiency, we can say that he is in a dissonant state 0 At the core of the theory of cognitive dissonance 3J. W. Brehm and A. R. Cohen, Explorations in Co nitive Dissonance (New York: John Wiley & Sons, 1962), p. 00. . . 4L. Festinger, Conflict, Decision and Dissonance (Stanford, Calif.: Stanford Univ. Press, 1964), p.156. 10 is the postulate that cognitive dissonance gives rise to a motivational tension state. Just as hunger is motivating, cognitive dissonance is motivating. Cognitive dissonance will give rise to activity oriented to reducing or eliminating the dissonance. Successful reduction of dissonance is rewarding in the same sense that eating when one is hungry is rewarding.5 According to Festinger, there are three major ways in which dissonance may be reduced: (a) changing one or more of the cognitive elements involved-in dissonant relations; (b) adding new consonant elements; and (c) decreasing the importance of the elements involved in diSsonant relations. When the two cognitions are dissonant with each other and are highly resistant to change and to decrease their importance, a mode of dissonance reduction would be through the addition of consonant elements. In this way, the prOportion of relevant elements which are dissonant would be reduced. With respect to the relationship between commitment and mode of dissonance reduction, Brehm and Cohen pointed out that "commitment, . . . increases the resistance to change of the corresponding cognitive element, for it increases the resistance to change of 5L. Festinger, "The Motivating Effect of Cognitive Dissonance," in Assessment of Human Motives, ed. by G. Lindzey (New York: Rinehart, 1958), p.70. 11 the reality to which the element corresponds."6 If the person is committed to a given behavior or decision, he would reduce the resulting dissonance by change in elements other than those involved in that commitment. However, if the reality involved is ambiguous, this effect of commitment would be lessened. Thus it can be said that ambiguity of the reality to which commitment corresponds has a considerable influence upon the mode of dissonance reduction. If a foreign student with objective English deficiency is unambiguously committed to study at a U.S. university, he would be likely to reduce the resulting dissonance by change in elements other than those involved in his commitment to study at a U.S. university. One of the clear implications of the theory concerns selective exposure to information: if dissonance exists, the person should seek out information that reduces dissonance and avoid information that increases dissonance. Brehm and Cohen, however, after careful review and analysis of the results of the relevant studies, concluded that ". . . all failed to find evidence confirming the derivation that dissonant information will be avoided in prOportion to the amount of dissonance produced. In each of these studies, while subjects sought out 6Brehm and Cohen, Explorations in Cognitive Dissonance, pp.8-9. 12 dissonance-reducing information, they did not necessarily aggig'dissonance-increasinginformation."7 Festinger interpreted this weakness of the results produced by studies oriented toward the problem of selective exposure to information as follows: . . . if the person has some confidence in his ability to deal effectively with the concrete aspects of the new information, he will overcome his reluctance to read it and will expose himself to these details with a critical attitude in order to reduce the dissonance by counterarguing.8 If the person does not feel confident of his ability to reduce dissonance in a specific situation, he would be expected to show a reluctance to expose himself to dissonance-increasing information. Therefore, it seems reasonable to suppose that if a person has much self- confidence to reduce dissonance, he would show a preference for potentially dissonance-increasing activity so as to help in the dissonance-reduction process. For foreign students With objective deficiency in the English language, participation in communication activities for which some English proficiency is required seems to increase dissonance because in such communication activities they might experience their language deficiency. 71bid., p.93. 82 8Festinger, Conflict, Decision, and Dissonance, p. . 13 Among them, however, those who have much confidence to cOpe with such dissonance would participate in communication activities so as to help in the dissonance-reduction process. Since a self-evaluation which includes many positive self definitions gives the individual a feeling of confidence,9 Festinger's interpretation concerning self-confidence may implicate positive self-evaluation. Feeling fundamentally self confident, the "positive" person is less ruffled by unhappy events. LThus he finds it possible to take criticism and to evaluate it clearly. In terms of positive self perceptions, Combs and Snygg State . . . negative percepts are evaluated against the large mass of basically positive experience, in which perspective they seem far less important or overwhelming . . . . The possession of a large reservoir of positive experience of self provides the individual with a vast security to be used as a base for adventure and a firm foundation for meeting even the more difficult aspects of life with courage.10 Combs and Snygg's notion of positive self perceptions has parallels in Lecky's theory of self-consistency and 3A. W. Combs and D. Snygg, Individual Behavior, rev. ed. (New York: Harper & Row, 1959), p.241. 1°Ib1d., p.242. 14 in Rogers' self theory.11 Lecky, in his theory of self-consistency, postulated a need for self-consistency as a basic need of the organism. For Lecky, the necessity to maintain the unity and integrity of the organism is the source of motivation and the achievement of a unified and self- consistent organization is the develOpmental goal. Thus the individual tends to seek the type of experience which confirms and supports the unified organization. Then it seems reasonable to say that the consistency of the pervading self-evaluation is seen as the most important source of motivation for social behavior, which is in turn selective in character. Rogers, in his self theory, placed great emphasis upon the notions that (a) the self strives for consistency, (b) the organism behaves in ways that are consistent with the self, and (c) experiences that are not consistent with the self are perCeived as threats. Then he formulated a series of nineteen propositions, some of which are: (1) "As experiences occur in the life of the individual, they are either (a) symbolized, perceived and organized into some relationship to the self, (b) ignored because there is no 11P. Lecky, Self-consistency: A Theo of Personality(New York: Island Press, 1945); C. R. Rogers, CIient-centered Therapy; Its Current Practice,_lmplications, and Theory(Boston: Houghton-Mufflin, 1951). 15 _ perceived relationship to the self-structure, (c) denied symbolization or given a distorted symbolization because the experience is inconsistent with the structure of the self."12 According to this prOposition, the primary criterion for selectivity in perception is whether the experience is consistent with the self-evaluation at the moment. Rogers pointed out that persons will often stoutly maintain and enhance self-evaluations which are completely at variance with reality. The person who feels that he is competent will exclude from awareness evidence that contradicts this evaluation or he will reinterpret the evidence to make it consistent with his sense of competence. If, on the other hand, the person feels that he is incompe- tent, he may even do poorly to prove to himself and to others that he is no good. These are elaborated in the following propositions. (2) "Most of the ways of behaving which are adopted by the organism are those which are consistent with the concept of the self."13 This prOposition is the theme of Lecky's theory of self-consistency. (3) "Any experience which is inconsistent with the organization or structure of self may be perceived as a threat, and the more of these perceptions there are, 12;§1d., p.503. 13Ibid., p.507. 16 the more rigidly the self-structure is organized to maintain itself."14 The person builds up self-defenses against threatening experiences. If the self-evaluation becomes less consistent with experiences, more defenses will be built up by selecting experiences which are consistent with the self-evaluation. According to Combs and Snygg's notion of positive self perceptions, however, those who have positive self- evaluations would have far less need to defend the self against threatening experiences. These experiences, if they are minor, may be perceived by the "positive" person as challenging opportunities to test his mettle. For persons with positive self-evaluations, therefore, self- testing can be an exhilarating experience to be met with commitment. In terms of self-evaluation, Festinger argued that a drive for self-evaluation or a desire to know precisely that one is and is not capable of doing would require social behavior for its resolution.15 Festinger pointed out that the importance of communication would be enhanced when no direct or simple test can be used to judge the validities of attitudes, opinions, beliefs, or 14Ib1d., p.515. 15L. Festinger, "MOtivations Leading to Social Behavior," in Nebraska Symposium on Motivation, 1954, ed. by M. R, Jones (Lincoln: Univ. of Nebraska Press, 1954). - l7 abilities because ". . . the person 'tests himself' against, or more Specifically, compares himself with other persons."16 Cofier and Appley generalized this by saying that"uncertainty about one's characteristics may lead to communication and affiliation, when data cannot be obtained from nonsocial sources."17 However, these notions would not be applicable to the settings where such communication may increase dissonance unless the notion of positive self-evaluation is taken into account. In general, it would be reasonable to assume that in a dissonant situation, the more positive an individual's self-evaluation, the less selective his participation in social communication activities that are potentially dissonance-increasing. Foreign students who are committed to study at a U.S. university but not permitted to do so because of their objective English deficiency would be in a dissonant state--a motivational tension state. They would then be motivated to reduce the dissonance by maximizing cognitive elements that would be consistent with commitment. If they have more positive self-evaluations, they might be less selective in their participation in social communication —__ 16Ib1d., p.195. 170. N. Cofer and M. H. Appley, Metivation: _Iheorygand Research (New York: Jo Wiley'&ISons, 1967), p.787. - 18 activities to reduce the dissonance by reevaluating them- selves or by seeking out evidence that is consistent with commitment. If, on the other hand, they have less positive self-evaluations, their participation might be more selective. Hypotheses A central kernel of dissonance theory, as discussed in the preceding section, is the notion that where a person is committed to a given behavior, information that is clearly inconsistent with that commitment should create dissonance and he should manifest attempts to reduce that dissonance. The crucial role of commitment in the theory of cognitive dissonance is to aid the specification of the point at which a dissonant relationship occurs and to aid the specification of the ways in which attempts at dissonance reduction are made. Little attention has been given to commitment as an explicit variable in previous research in spite of its crucial role in the creation of dissonance. It seems reasonable to hold that the arousal of dissonance is unequivocal in situations that involve commitment and any information inconsistent with that commitment. Thus, commitment facilitates the specification of what is (iissonant and what is consonant. Where a foreign student With commitment to study at a U.S. university is informed 19 that he cannot start his study because of his English language deficiency of which correction is required for study at a U.S. university, dissonance should be aroused in connection with his commitment and then he should manifest attempts to reduce it. For such a foreign student who has objective English language deficiency as well as commitment to study, the information that he is not permitted to start his study without correction of his language deficiency is clearly inconsistent with his commitment. In terms of the reduction of dissonance, the theory of cognitive dissonance is not capable of predicting the specific manner in which consonance among cognitions will be achieved. Among the crucial defects of the theory is an inadequate specification of the conditions that will lead to a certain form of dissonance reduction. The important factor controlling the various modes of dissonance reduction would be the resistance to change of relevant cognitions. As Brehm and Cohen pointed out,18 a guideline for the understanding of resistance to change is based on the notion of commitment. That is, 18Brehm and Cohen, Explorations in Cognitive :2issonance, p.65. 20 commitment increases the resistance to change of the reality to which the relevant cognition corresponds and thereby affects the modes of dissonance reduction. Hence dissonance is likely to be reduced by change in elements other than those involved in the commitment. If the reality involved is ambiguous, however, the resistance against making the corresponding cognitive element nonveridical with reality would be lessened. Unfortunately, there are no data bearing directly on this aspect of dissonance reduction. At this point it should be noted that the cognitive elements which are dissonant with the commitment will not always be easily changeable. For example, the cognitive element corresponding to English language deficiency, which has been determined objectively by reliable tests of English language proficiency, is quite difficult to change in actuality. Under this condition dissonance reduction would take some form other than change in dissonant elements. One such possible form is the addition of further cognitive elements which must be selected so that they are all consonant with the commitment. In this way they will reduce the prOportion of dissonant elements. Where dissonance can be reduced by adding new cognitive elements, the person would expose himself voluntarily to sources which would add new elements that ‘would reduce dissonance but would certainly avoid exposure 21 to sources which would increase dissonance. This selectivity in exposure to sources, however, remains questionable with regard to the avoidance of exposure to dissonance-increasing sources. This is also considered as an apparent disconfirmation of dissonance theory. Self theory may be a useful way of improving predictions of avoidance Of exposure to dissonance-increasing sources. According to self theories, as discussed in the preceding section, it seems reasonable to suppose that those who have more positive self-evaluations will actively participate in social communication activities which are potentially dissonance-increasing so as to help in the process of attempting to reduce dissonance. Avoidance of participation in potentially dissonance- increasing communication activities would be useful in the dissonance reduction process only if the person has less positive self-evaluation. Unfortunately, again, there are no data bearing directly on the relationship between positiveness of self-evaluation and participation in potentially dissonance-increasing communication activities. The notion of positive self-evaluation would help explain the disconfirmation of tendencies to avoid exposure to dissonance-increasing sources. Ambiguity of commitment and positiveness of self-evaluation would set limits on the avenue of dissonance reduction and thus they would provide the 22 theory with predictive power in specifying a mode of dissonance reduction. In this study, therefore, the following hypotheses are tested with foreign students who have objective language deficiency and who are actually committed to study at U.S. universities or colleges. For such foreign students, participation in communication activities which requires some English language proficiency is potentially dissonance-increasing because in such participation they may experience their language deficiency and then such experience would add further cognitive elements that are dissonant with their commitment. Hypothesis I: Of the foreign students who are unambiguously committed to study at U.S. universities, those with more positive self-evaluations are less selective in their participation in social communication activities than those with less positive self-evaluations. Hypothesis II: Of the foreign students who are ambiguously committed to study at U.S. universities, those with more positive self-evaluations are less selective in their participation in social communication activities than those with less positive self-evaluations. Hypothesis III: Of the foreign students with morengSItive self-evaluations, those who are unambiguously committed to study at U.S. universities are less selective in their participation in social communication activities than those who are ambiguously committed. 23 Hypothesis IV: Of the foreign students with less ypositive Self-evaluations, those who are unambiguously committed to study at U.S. universities are more selective in their participation in social communication activities than those who are ambiguously committed. Review of Literature Research on selective exposure to dissonant information. Although a critical analysis of the theory of cognitive dissonance has been ably done in relation to other theories of cognitive consistency,19 selectivity of participation in potentially dissonance-increasing communication activities has not been clarified thoroughly. Ehrlich ggflal. and Engel, assuming that dissonance would exist following the purchase of a new automobile, found that those who purchased a new car read more ads about that car than about other cars.20 19For example, C. E. Osgood "Cognitive Dynamics in the Conduct of Human Affairs, Puhlic Opinion Quarterly, 24(1960), 341-65; R. B. Zajonc, "The Concepts of—Balance, Congruity, and Dissonance,‘ Public Opinion Quarterly, 24 (1960), 280-96; Brehm and Cohen, Explorations in Cognitive Dissonance; R. Brown, "Models of Attitude Change,"in Néw Directions in Psychology, ed. by R. Brown, E. Galanter, E. H. Hess, and G. Mandler (New York: Holt, Rinehart and Winston, 1962); S. Feldman, ed., Cognitive Consistency (New York: Academic Press, 1966). 20D. Ehrlich, I. Guttman, P. Schonbach, and J. Mills, "Post-decision Exposure to Relevant Information," Journal of Abnormal and Social Psychology, 54(1957), 98-102; JTWF. Engél,"Are Automobile Purchasers Dissonant . Consumers?" Journal of Marketing, 27(1963), 55-58. 24 But this finding is equivocal in terms of selective exposure to information, because they did not avoid reading ads about other cars. In experiments reported by Mills, Aronson, and Robinson and by Rosen,21 it was shown that there is a significant preference for articles stressing positive aspects of the chosen alternative over articles stressing positive aSpects of the rejected alternative. However, there was no indication of a preference for thoSe emphasizing negative aspects of the rejected one over those emphasizing negative aspects of the chosen. While it is possible to interpret the results as evidence for the existence of selective exposure to consonant informa- tion, it is also possible to interpret them as simply indicating a preference for the information which explained how to do well on the chosen alternative. The results of the study by Adams, on mothers' exposure to information regarding child-rearing practices, were not entirely confirmatory, either.22 Mothers who heard a view Opposing their own showed more interest in the future 21.1. Mills, E. Aronson, and H. Robinson, "Selectivity in Exposure to Information " Journal of ormal and Social Psychology, 59(19595, ; S. Rosen Post-decision Affinity for Incompatible Information," Journal of Abnormal and Social Psychology, 63(1961), 188-90. 22J. S. Adams, "Reduction of Cognitive Dissonance by Seeking Consonant Information," Journal of Abnormal and _§ocial Psychology, 62(1961), 74-78. 25 talk supporting their own view than in the Opposing talk, but so did the mothers who heard a view supporting their own. ‘Mbreover, the mothers who heard the Opposite side showed more interest in hearing one talk or the other than the mothers who heard their own side. This might be interpreted simply as an arousal of curiosity on the issue. In this regard, Brehm and Cohen point out that "the Opportunity of exposing oneself to dissonant information is, at least under some circumstances, accepted as some sort of challenge."23 In the study reported by Feather, it was demonstrated that subjects expose themselves equally to both dissonant and consonant information but are more critical in reading the dissonant material than in reading the consonant material.24 This might suggest that if a subject is confident of his ability to deal with the dissonant information, avoiding dissonant information would not be expected. An experiment by Canon examined the effects of self-confidence on selectivity of exposure to information.25 23Brehm and Cohen, Explorations in Cognitive Dissonance, p.93. 24N. T. Feather, "Cognitive Dissonance, Sensitivity, and Evaluation," Journal of Abnormal and Social Psychology, 66(1963), 157-63. p. 25L. K. Canon, "Self-Confidence and Selective Exposure to Information," in Festinger, Conflict, Decision, ppd Dissonance. . 26 The results showed that the less confidence the subject had in being able to deal with dissonant material, the greater was his preference for consonant material, and that those who were highly confident had a greater preference for dissonance-increasing information. In the light of the data obtained by Canon, Festinger came to the following conclusion. 5 When the person perceives that dissonance will be effectively reduced by exposing himself to and coping with the details of the dissonant information, one will certainly not observe avoidance of it. Avoidance of potentially dissonance-increasing information would be useful in the service of dissonance reduction only if the person feels unable to cape with the new information in its details.26 Research on commitment and mode of dissonance reduction. Little research has been done toward the delineation of the factors that determine resistance to dissonance reduction. Brehm and Cohen pointed out the guideline based on the notion of commitment for the understanding of resistance to change. Given commitment to the discrepant act, the subject can generally reduce his dissonance only by some realignment which makgs his cognitions consistent with his behavior. The experiment reported by Davis and Jones showed that those subjects who can uncommit themselves do 26Festinger, Conflict, Decision and Dissonance, p.96. . 27Brehm and Cohen, Explorations in Cognitive Dissonance, p.67. 27 not change their attitudes whereas those who commit them- selves change their attitudes to come in line with their behavior as a way of reducing the dissonance engendered by that behavior.28 The experiments by Zimbardo, by Cohen, and by Brock and Buss also provided similar illustrations of the relationship between commitment and mode of dissonance reduction--that is, the particular commitment sets limits on the process of dissonance reduction.29 Greenwald found, in his study of belief and behavior changes following a persuasive communication, that communications produce changes in belief and in behavior except when the subject is clearly committed to a position contrary to the one presented in the communication.30 When such commitment is present, belief changes but behavior does not. He suggested that commitment immunizes the subject against behavioral change. Kiesler and Sakumura also suggested that behavioral acts 28K. Davis and E. E. Jones, "Changes in Inter- personal Perception as a Means of Reducing Cognitive Dissonance," Journal of Abnormal and Social Psycholo , 61(1960), 402910. 29P. G. Zimbardo, "Involvement and Communication Discrepancy as Determinants of Opinion Change," Journal of Abnormal and Social Psychology, 60(1960), 86-94; A. R. Cohen, "A Dissonance Analysis of the Boomerang Effect," Journal of Personality, 30(1962), 75-88; T. C. Brock and A. H. Buss, "Dissonance, Aggression and Evaluation of Pain," Journal of Abnormal and Social Psychology, 65(1962), 197-202. 30A. G. Greenwald, "Behavior Change Following a Persuasive Communication," Journal of Personality, 33(1965), 370-91. 28 involving commitment are difficult to change and attitudes consistent with the behavioral commitment are strengthened but if attitudes are inconsistent, they are likely to change toward consistency with the commitment.31 Research on self-evaluation and dissonance. Studies concerning cognitions about ability indicate that an individual's cognitions about his abilities have a considerable effect on dissonance reduction. It is consistent for an individual to make a good performance if he believes that he has high ability, whereas a good performance is inconsistent with perceived ability if an individual believes he is incompetent.32 Vroom hypothe- sized that “if a person believes himself to possess an ability and believes that successful performance of his task requires that ability, he will prefer performing the task effectively to performing it ineffectively."33 This prOposition was supported by the studies of'Kaufmann 310. A. Kiesler and J. Sakumura, "A Test of a Medel for Commitment, " Journal of Personality and Social Psychology, 3(1966), 349 -53. 3211. B. Gerard, s. A. Blevans, and T. Malcolm, "Self-Evaluation and the Evaluation of Choice Alternatives," Journal of Personality, 32(1964), 395- 410. 33V. H. Vroom, WOrk and Motivation (New York: Wiley, 1964), p. 247. 29 and Bachman.34 Aronson and Carlsmith showed,furthermore, that if an individual's outcomes are inconsistent with his cognitions about his abilities, he attempts to modify the outcomes.35 In terms of cognitions about oneself, Deutsch gt Elm demonstrated that "postdecisional dissonance will only occur if the decision is perceived to reflect or measure aspects of the individual's self . . . ."36 Festinger and Bramel discussed the use of defensive projection as a means of dissonance reduction and suggested that there is a similarity between certain dissonance-reduction mechanisms and certain defense mechanisms in spychoanalytic theory.37 A given defensive behavior may reflect itself in a given mode of dissonance reduction. The Bramel experiment showed that there is more projection where dissonance is high than where it 34H. Kaufmann, "Task Performance, Expected Performance, and Responses to Failure as Functions of Imbalance in the Self-Concept," unpublished doctoral dissertation, Univ. of Pennsylvania, 1962; J. G. Bachman, "Motivation in a Task Situation as a Function of Ability and Control over the Task,” Journal of Abnormal and Social Psychology, 69(1964),272-81. 35E. Aronson and J. M. Carlsmith, "Performance Expectancy as a Determinant of Actual Performance," Journal of Abnormal and Social Psychology, 65(1962), 178-82. 36M. Deutsch R. M. Krauss, and N. Resenan "Dissonance or Defensiveness?" Journal of Personality, 30(1962), 21. 37L. Festin er and D. Bramel "The Reactions of Humans to Cognitive issonance " in e Experimental Foundations of Clinical Ps cholo , ed. By A. Bachrach lNEw YorE: Basic Books, I562). A 30 is low.38 It seemed to appear that projection would Occur in order to reduce dissonance. It was found, moreover, that the major dissonance-reducing effects of projection tend to occur when self-esteem is exceptionally high. This might suggest that dissonance would occur when the Subject's behavior follows from the Opposite of his motivation to maintain a positive self-evaluation. Rosearch on self-evaluation andyparticlpation in communication activity Brownfain demonstrated that stability of the self-conception is a dimension of personality serviceable to understanding and predicting behavior.39 According to Brownfain, terms such as Lecky's "self-consistency," Rogers' "organized self-structure" or "wholeness of personality," Combs and Snygg's "differentiation" are all suggestive of the same variable which Brownfain called "stability of the self-concept." For Brownfain, the self-concept is a configuration of possible self-definitions and the stability of the self-concept derives from interrelations among various ways of defining the self. Stability of the self-concept was measured in terms of the 38D. Bramel, "A Dissonance Theory Approach to Defensive Projection," Journal of Abnormal and Social Psychology, 64(1962), 121-29. 39J. J. BrOwnfain, "Stability of the Self-Concept as a Dimension of Personality," Journal of Abnormal and Social Psychology, 47(1952), 597-606. 31 discrepancy between positive and negative definitions of the self. The major findings which favor the subjects with more stable self-concepts were these: (a) they have a higher level of self-esteem (higher self-ratings), (b) they are freer of inferiority feelings and nervousness,~ (c) they are better liked and more p0pular in the Suhject group, (d) they view themselves more as they believe others view them, (e) they have more active social participation, and (f) they Show less evidence of defensive behavior.’ These findings might give support to the assumption that positiveness of self-evaluations may be negatively related to selectivity of participation in social communication activity. This assumption might be also supported, in a sense, by the Brown study which showed that the high and low participants in formal organizations develOp self-judgments which correspond closely to their actual participant behavior, and they feel expected to participate accordingly.40 Research on foreign students' language proficiency and their participation in communication activity. Research with foreign students in the United States has focused in particular on the effects of contact with Americans and the processes of interaction but not on selectivity of participation in communication activities. 40E. J. Brown, "The Self as Related to Formal Participation in Three Pennsylvania Rural Communities," Rural Sociology, 18(1953), 313-20. 32 From different points of view, Smith, Pool, and Mishler reviewed some of this research.41 However, the dissonance aspect of language deficiency was not taken into account. In terms of language deficiency, its negative relation to participation in communication activities has been suggested by some studies of foreign students in the United States. Morris reported a positive relation between language facility and the extent of social interaction for the foreign students at UCLA.42 Among Scandinavian students at a midwestern university, those rated higher in English language proficiency by interviewers also scored higher on measures of participa- tion in American life than those rated lower.43 It was also found that foreign students at Cornell who scored higher on measures of contact with Americans tended to report no difficulty in understanding American 41M. B. Smith, ed., "Attitudes and Adjustment in Cross-cultural Contact: Recent Studies of Foreign Students," Journal of Social Issues, 12-l(l956); I. de S. Pool, "Effects of Cross- National Contact on National and International Images," in International Behavior, ed. by H. C. Kelman (New York: Holt, Rinehart and Winston, 1965); A. L..Mish1er, "Personal Contact in International Exchanges," in International Behavior. 42R. T. Morris, The Two- -WsylMirror: National Status in Foreign Students' Adjustment (Minneapolis: Univ. of Minnesota Press, 1960). . 43W. H. Sewell and 0.1M. Davidsen, Scandinavian Students on an American Campus (Minneapolis: Univ. of Minnesota Press, 1961). 33 English.44 A more extensive study of foreign students in the United States also reported a positive relationship between interviewers' ratings of their English language skills and the extent of their social relations with Americans.45 In this study, furthermore, it was found that a student's confidence in his ability to Speak English has a more important influence on the development of social relations than his actual command of the English language. Further Deutsch and Wen indicated that, for foreign trainees, English language facility is an important factor in satisfaction with the total social experience in the United States.46 44C. Selltiz, J. R. Christ, J. Havel, and S. W’. Cook, Attitudes and Social Relations of Foreign Students in the United States (Minneapolis: Univ. of Minnesota Press, 1963). 45Ibid. _ 46S. E. Deutsch and G. WOn, "Some Factors in the Adjustment of Foreign Nationals in the United States," Journal of Social Issues, 19(1963), 115-22. CHAPTER II METHOD Subjects Foreign students who were enrolled in full time work in an intensive English program at ten universities during the survey period of April 21 to May 21, 1969, comprise the subjects in this study with the resultant total number of 572. In this number of subjects, those low level students who would not be able to handle the questionnaire used in this study intelligently as judged on the basis of their English proficiency scores, those who did not complete all the questionnaire items, and those who had no plan to study at American universities or colleges after completing study of English are excluded. The foreign students comprising the subjects in this study were screened for their English proficiency through tests administered by universities in which they were enrolled and then they were required to correct their English deficiency before starting their academic programs at American universities or colleges. 34 35 The ten universities involved in this study were chosen among the seventeen universities which offer an intensive English program for foreign students in the United States on the basis of their cooperativeness and availability of apprOpriate group of students for this study. Table 1 presents the distribution of subjects used in this study among the ten universities. Table 1.--Distribution of 572 subjects among the ten universities chosen. University Frequency Eggt University Frequency Eggt California 58 10 Michigan 71 12 (UCLA) State Hawaii 34 6 New York 90 16 Kansas 67 12 San Francisco 55 9 . State(College) Louisiana 64 11 Southern Cal.‘ 27 5 State Miami 74 13 Southern Ill. 32 6 36 Measurement of Variables Positiveness Of self-evaluation. A number of methods have been developed for measuring self-evaluation. Subjects are required to reveal the content of self-evaluations, for example, by rating the self against a series of bipolar traits (the semantic differential technique)1; by rating the self on specified personal characteristiCs, relative to the others in a group of limited size (the self-other ranking technique)2; by responding to a standard personality inventory (the A self-rating inventory)3; or by writing out self-deScriptive Swords (the "twenty-statements").4 . 0n the criteria of usefulness to determine positiveness of the self-evaluation, ease of response, 1C. E. Osgood, G. J. Suci, and P. H. Tannenbaum, The Measurement of Meaniog (Urbana, 111.: Univ. of Illinois Press, 1957); R. L. Burke and W. G. Bennis, "Changes in Perception of Self and Others during Human Relations Training," Human Relations, 14(1961), 165-82. 2A. D. Calvin and w. H. Holtzman, "Adjustment and the Discrepancy between Self Concept and Inferred Self," Journal of Consulting Psychology, 17(1953), 39‘440 - - 3Brownfain, ”Stability of the Self-Concept as a Dimension of Personality." 4M. H. Kuhn and T. S. McPartland, "An Empirical Investigation of Self-attitudes," American Sociological Review, 19(1954), 68-76; T. S. McPartland and J. H. Cuming, "Self-Conception Social Class, and Mental Health," _Hpman Organization, 17(1958), 24-29. 37 and cross-cultural stability, the semantic differential technique is used in this study. Especially cross- cultural stability would be a very important criterion for this study. OSgOOd's semantic factors seem to be relatively stable across people differing widely in both language and culture.5 The instrument used to serve as a measure of positiveness of the self-evaluation consists of six bipolar adjectival seven-step scales which are selected and adopted from among Osgood's semantic differential scales with high loadings on the evaluative factor.6 The scales used are: friendly-unfriendly, good-bad, happy-sad, honest-dishonest, important-unimportant, successful-unsuccessful. 5H. Kumata, "A Factor Analytic Study of Semantic Structures across Three Selected Cultures," unpublished Ph. D. dissertation University of Illinois, 1958; H. Kumata and W. Sc ramm, "A Pilot Study of Cross-Cultural Methodology," Public Opinion Quarterly, 20(1956), 229-38; H. Maclay and E. E. Ware, "Cross-Cultural Use of the Semantic Differential," Behavioral Science, 6(1961), 185-90; G. J. Suci, "A Comparison of Semantic Structures in American Southwest Culture Groups " Journal 0 Abnormal and Social Psychology, 61(1960), 25-30; Y. Tanaka, T. Oyama, and C. E. Osgood "A Cross-Cultural and Cross-Concept Study of the Generality of Semantic Spaces," Journal of Verbal Learniog and Verbal Behavior, 2(1963), 392-405; H. C. Triandis and C. E. Osgood, "A Comparative Factorial Analysis of Semantic Structures in Menolingual Greek and American College Students," 8 a Journal of Abnormal and Social Psychology, 57(195 187-960 - 6Osgood, Suci, and Tannenbaum, The Measurement p§_Meaniog, pp.31-75. 38 Subjects were instructed to rate themselves against a series of six bipolar scales on the basis of their immediate feelings about themselves. To measure positiveness of subjects' self-evaluations, a digit was assigned to each of the seven positions on scales: the extreme positions in the direction of such adjectives as friendly, good, happy, honest, important, and successful were scored 7 and the other extreme positions were scored 1, with the neutral positions scored 4. These digit scores were summed and averaged over the scales to obtain a self- positiveness score for each respondent. Positiveness of self-evaluation is operationally defined as a self-positiveness score. The more positive the self-evaluation, the higher the score. Ambiguity of commitment to study at a U.S. university. Ambiguity of commitment refers to ambiguity of the reality to which subjects are committed. The four items dealing with the reality to which subjects are committed are used to determine individual differences on a continuum related to its ambiguity. For each item two response Options (yes or no) are available. The items used and the ordering of these items are: (1) Are you planning to study at an American university or College after completing study of English?; (2) Have you submitted an application for admission to that university or college?; 39 (3) Have you been approved for admission to that university Or college?; and (4) About your major work, have you consulted with your academic adviser in your major department at that university or college? In terms of a set of dichotomous items, five possible response patterns occur. That is, the j-th pattern is a set of responses such that the responses to the first j items are all positive and the responses to the other items are all negative. Each respondent was assigned a score corresponding to patterns 0 to 4. Subjects who responded positively to a given item have higher rank scores than those who responded negatively to that item. However, the score of 0 refers to uncommitment which implies no arousal of dissonance. Ambiguity of commitment to study at a U.S. university is Operationally defined as a rank score which is assigned to each respondent (commitment-ambiguity score). The more ambiguous the commitment, the lower the score. Selectivity of particlpation in social communication activity. Social communication activities taken into account in this study are restricted to those in which subjects are advised by teachers to participate in order to develOp their skills of English. The eleven items dealing with those activities are used to determine individual differences 40 on a continuum related to participatiOn. The items used are: talking with American friends; talking with friends from other countries; visiting American_homes; attending lectures; attending plays; attending movies; listening to the radio; watching television; reading American books; reading American magazines; and reading American news- papers. For each item four response Options ranging from "no, not at all" to "yes, frequently" are available. The subjects' task is to respond to each item on the basisof their participation during "last" week. To score the items in the simplest manner, scores ranging from 0 to 3 may be assigned to the response categories of each item, with the 0 being assigned to the "no, not at all" position. A respondent's score on the set of items is then calculated by summing the item scores. Such an arbitrary scoring technique is based on two assump- tions: the equality between similar responses to all items and the equality of the distance between adjacent response categories. Both items and responses may be differentially weighted. An approach is to use the standard deviations of the distributions of the items. However, it has been indicated that the converted sigma technique correlates about .99 with the arbitrary scoring.7 7G. Murphy and R. Likert, Public Opinion and the Individual (New York: Harper, 1938);W . J. Goode and P. K. Hatt, Methods in Social Research (New York: .McGraw-Hill, 1952). 41 A cogent critic of developing a single arbitrary index is Guttman who claimed that arbitrary indices may obfuscate the purpose of the research because of an intuitive assessment of unidimensionality involved in the summated scaling procedure.8 Guttman presented the procedure which adheres to the logical structure of an ordinal scale. According to Guttman, Scale analysis tests the hypothesis that a group of peOple can be arranged in an internally meaningful rank order with respect to an area of qualitative data. A rank order of peOple is meaningful if, from the person's rank order, one knows precisely his responses to each of the questions or acts included in the scale.9 With respect to the relationship between arbitrary summated indices and Guttman scale scores, it is pointed out that It is true that if an area is scalable, then the resulting scale scores will correlate very highly with any index obtained by arbitrary weights, provided the weights are in the right direction. Scale theory proves that there is no harm in obtaining an apparently arbitrary index from a scalable area, either for descriptive or predictive purposes.lo 83. A. Stouffer, L. Guttman, E. A. Suchman, P. F. Lazarsfeld, S. A. Star and J. A. Clausen, Measurement and Prediction (Princeton: Princeton Univ. Press, 1950), ch.6. - 9 Ibid., p.88. 10Ibid., p.175. 42 An index of some process reflecting a tendency to positive or negative behavior might be developed from responses to a heterogeneous collection of items. Thus subjects responding differently to the items may be comparable with respect to the strength and direction of this tendency. In fact, arbitrary indices seem to be frequently used as a means of dimensionalizing survey data, and evidence of the items' scalability in the Guttman sense seems to be scarcely found. Many researchers have been satisfied with the pragmatic consequences of summated scaling for the dimensions assessed in their studies. In the Lazarsfeld and Thelens study,11 however, summated scales were used after attempts to develop Guttman scales failed. In comparison with summated scaling, Guttman scaling seems to provide a relatively unambiguous approach to studying the organization among items. In this study, therefore, Guttman scaling is attempted to develOp a cumulative unidimensional scale for participation in social communication activities. Guttman's procedure makes possible a ranking of respondents in terms of their participation in communication activities, but does not provide a basis for determining how 11p. F. Lazarsfeld and w. Thelens, Jr., The Academic Mind (Glencoe, Ill.: Free Press, 1958). 43 much more one person participates than another. That is, the result of Guttman scaling is an ordinal scale. If a set of n dichotomous items forms a perfect scale, then only n + 1 scale types are obtained. The j-th scale type is a set of dichotomized scores such that the responses to the first j items are all positive and the responses to the other n-j items are all negative. Thus a scale score can be assigned to each respondent according to scale types 0 to n. A score of j means that the response pattern for a respondent is the j-th scale type. In practice the problem of scaling analysis is to determine whether the data can be treated as if they constituted a perfect scale. The degree to which the data fit the model of perfect Guttman scale is expressed in what is known as the reproducibility coefficient. The reproducibility coefficient for a given item is determined by the following formula: Number of Errors Total Number of Responses Errors are failures to fit the model. The over-all reproducibility of a Guttman scale is the average of the item reproducibilities. In this study, the scale is judged acceptable 44 if the following criteria are met:12 (1) The reproducibility coefficient is .900 . or more. (2) No item has less than .850 reproducibility. (3) No item has more error than non-error. (4) Item marginals are distributed over a wide range. (5) Items with marginals around 50% are included in the scale. (6) At least six items are included in - the scale. Selectivity of participation in social communica- tion activities is Operationally defined as a Guttman scale score of participation. The less selective the participa- tion, the higher the score. Data Gathering The questionnaire method was used to gather the data needed in this study. The forms of the questionnaire prepared for this study (see Appendix) were mailed to the directors or the coordinators of intensive English program at the ten universities involved in this study. During the period of April 21 to May 21, 1969, these questionnaires were administered at class meetings by letouffer,‘ t al., Measurement and Prediction; also see Gullahorn's rationale of these criteria, J. E. Gullahorn, "A Factorial Study of International Communication and Professional Consequences Re orted by Fulbright and Smith-Mundt Grantees, l947-l957,’ unpublished Ph.D. dissertation, Michigan State University, 1964. LQJL‘ | VIII-III” % 45 teachers at each university (except University of Hawaii where the questionnaires were handed out to subjects at class meetings and, after completed by subjects at homes, were turned in). The directors or the coordinators were also asked to report whether they had special events or anything unusual on their campuses during the survey period. Neither of them were reported. Data Analysis In analyzing the data, it seems necessary to control the variables such as sex, marital status, age, length of stay in the United States, length of enrollment in an intensive English program, and home country because of their relevance to subjects' participation in social communication activities. Sex and marital status. Table 2 presents the distribution of 572 subjects among sex and marital status. The number of subjects in three categories other than a Male-Single category is relatively small so that subjects in each category can be homogenized further by age, length of stay in the United States, length of enrollment in an intensive English program, and home country. Thus only male-single subjects were used to test the hypotheses. 46 Table 2.--Distribution of 572 subjects among sex and marital status. Male Female Single Married Single Married 365 71 124 12 (647.) (12%) (227.) (27.) Aa_~ 365 male-single subjects range in age from 18 to 40, but most of them are included in the age group of 18-27 (Table 3). Those male-single subjects who are over 27 in Ege were also eliminated from the subjects used to test the hypotheses. Table 3.--Male-single subjects by age. Range of ages 18-22 . . . 23-27 . . . 28-32 . . . Over 32 . . Total . . . Number of subjects a o o o o o o 201(5570) 1231342) 351 9%) 6( 21) 365(1001) 47 Length of stay in the United States and Length Of enrollment in an intensive English_program. Those students who have been approved for admission to a university or a college are usually required to take one or two terms of full time work in intensive English on the basis of their English proficiency scores and then most of them are permitted to start part time academic work in their major fields. In some universities an intensive English program follows the quarter or the trimester system and in some other universities it follows the semester system. This means course duration differs among universities. Whether a program follows the quarter system or the Semester system, however, all programs begin in September. For these reasons, 25 subjects who arrived in the United States before September 1968 were eliminated. Furthermore, 3 subjects who had not stayed in the United States at least for one month were also eliminated. Thus the resultant number of subjects used to test the hypotheses are 296. It can be said that they are relatively homogenized by Sex, marital status, age, length of stay in the United States, and length of enrollment in an intensive English program. However, they come from heterogeneous backgrounds, representing a wide variety of countries in which English is not the native language. In this study, these possible background 48 variables are represented by the variable of home country. Home country. The variable of home country was controlled according to the Farace factor analytic study of identify- ing regional clusters of countries.13 The Farace study defined six regional clusters of 109 countries on the basis of shared patterns over 54 variables which simulate eight aspects of life in each country: political factors, health and nourishment capabilities, agricultural productivity, climate, population characteristics, cultural factors, income and industrialization, and the mass media. Regional labels for the six clusters of countries are Latin America, North America/Western EurOpe, Central/South Africa, Asia, North Africa/Middle East, and Communist East Europe. Table 4 presents the distribution of 296 subjects who are relatively homogenized by sex, marital status, age, and length of stay in the United States among the six regional clusters of their home countries. As presented in Table 4, their home countries are distributed mostly over the three regional clusters of Latin America, North Africa/Middle East, and Asia. Therefore, the hypotheses will be tested for these three regional clusters. 13R. V. Farace, "Identifying Regional 'Systems' in National Development Research," Journalism Quarterly, 43(1966), 753-60. 49 Table 4.--Distribution of 296 subjects homogenized by sex, marital status, age, and length of stay among regional clusters of home countries. “nastiness equeqeee er Latin America 102 34% North Africa/Middle East 102 34 Asia 71 24 Communist East Europe 11 4 Western EurOpe 7 3 Central/South Africa 3 1 Statistical test. The subjects are categorized into the following four groups on the basis of their self-positiveness scores and their commitment-ambiguity scores: the unambiguous commitment (commitment-ambiguity scores 3 and 4) and high self-positiveness (above the median of self-positiveness scores) group; the unambiguous commitment and low self- positiveness (below the median of self-positiveness scores) group; the ambiguous commitment (commitment-ambiguity scores 50 l and 2) and high self-positiveness group; the ambiguous commitment and low self-positiveness group. To test the hypotheses, the following pairs of groups are compared on Guttman scale scores of participation in communication activities: (1) the unamblguous commitment and hlgh self-positiveness group and the unambiguous commitment and lgglself-positiveness group for Hypothesis I; (2) the amblguous commitment and hlgh self-positiveness grOup and the ambiguous commitment and log self-positiveness group for Hypothesis II; (3) the unambiguous commitment and hlgh self-positiveness group and the amblguous commitment and high self-positiveness group for Hypothesis III; (4) the unambiguous commitment and lo! self-positiveness group and the ambiguous commitment and log self-positiveness group for Hypothesis IV. Since each pair of groups constitutes two independent groups and the measure of participation- selectivity (a Guttman scale score of participation) constitutes an ordinal measure, the Mann-Whitney O_test is apprOpriate for analyzing the data.14 The method for determining the significance of the observed value of O depends on the size of n2, the number of cases in the larger of two independent groups. If n2 is 14S. Siegel, Nonparametric Statistics for the Behavioral Sciences (New York: MCGraw-Hill, 1956), pP0116-260 ‘ \‘J 51 between 9 and 20, the significance of the observed value of‘g is determined by reference to Table of Critical Values of.O in the Mann-Whitney Test.15 If n2 is larger than 20, the probability associated with a value as extreme as the observed value of'O is determined by computing the value of‘g and testing this 5 value by reference to Table of Probabilities Associated with Values as Extreme as Observed Values of 5:16 The value of.§ is determined by the following formula: nlnz 9' 2 (nl)(n2)(n1+nz+1) * - 12 where n1 - the number of cases in the smaller of two independent groups; n2 - the number of cases in the larger. 15Ibid., p.277. 16Ibid., p.247. CHAPTER III RESULTS Guttman scaling. A sample of 100 is recommended for Guttman scale analysis.. Therefore, Guttman scaling was attempted with all male-single subjects within each of the three regional clusters of home countries. The number of male-single subjects within each regional cluster is presented in Table 5. Table 5.--Number of male-single subjects , within each regional cluster of home countries. Regional cluster Number Of subjects Latin America . .. . . . . . . . 116 North Africa/Middle East . . . . 121 AS 18 O O O O O O 0 O O O O O O O 103 52 53 Table 6 presents the Guttman scale patterns for each regional cluster with reference to questionnaire items pertaining to participation in communication activities. For the three regional clusters, seven-item cumulative Scales were develOped. As these results illustrate, unidimensionality reflects the patterning of participa- tion in communication activities for each of the three regional clusters of home countries. Variations in the participation of the three regional categories of subjects resulted in different patternings of the items. Distributions of the subjects on the measures of the three variables. In Tables 7, 8, and 9 shown are the distributions of the subjects used to test the hypotheses within each of the three regional categories on the measures of commitment-ambiguity, self-positiveness, and participation-3electivity. On these measures of the three variables, the possible differences among the universities invloved in this study should be examined to determine any possible contamination which might be caused by geographical differences. Any significant difference among the universities on each measure was not found. Thus, it may be assumed that the possible contamination caused by geographical differences is negligible. 54 Table 6.--Guttman scale analysis: participation in communication activities. Itema Marginalb Cutoffc Reproducibility Latin America TV 93.1% 2 .966 Newspaper 86.2 2 .940 Radio 80.2 2 .905 American friend 70.7 2 .897 Magazine 58.6 2 .897 Movie 37.9 2 .888 American home 14.7 2 .966 Scale Reproducibility: .928 North Africa/Middle East TV 92.6% 2 .967 Radio 81.8 2 .917 Newspaper 70.2 2 .917 Magazine 57.9 2 .967 Book 45.5 2 .893 Play 28.1 2 .868 American home 19.0 2 .934 Scale Reproducibility: .923 Asia Radio 94.2% 2 .981 Newspaper 89.3 2 .951 TV 78.6 2 .913 Magazine 63.1 2 .854 Book 53.4 2 .913 Movie 35.9 2 .932 Play 16.5 2 .922 Scale Reproducibility: .924 8For full wording, see the questionnaire in Appendix. bThe marginals indicate the percentage of subjects responding positively to the items. cA subgect is considered to respond positively to the item if e checks a response category at or beyond the designated cutoff level. 55 Table 7.--Distributions of the subjects on the commitment- ambiguity score Latin America North Africa/ Asia Middle East Score Frequency Per Frequency Per Frequency Per cent cent cent 1 40 39.2 11 10.8 17 24.0 2 25 24.5 22 21.6 14 19.7 3 21 20.6 46 45.1 13 18.3 4 16 15.7 23 22.5 27 38.0 Total 102 100.0 102 100.0 71 100.0 Table 8.--Distributions of the subjects on the self- positiveness score. Latin America North Africa/ Asia A Middle East Score Frequency Per Frequency Per Frequency Per cent cent cent 201-300 0 0 0 0 1 1.4 301-400 7 6.9 9 8.8 10 1401. 401-500 1.4 1.307 20 19.6 18 2504 501-600 44 43.1 41 4002 22 31.0 6.1-7.0 37 36.3 32 31.4 20 28.2 Total 102 100.0 102 100.0 71 100.0 Table 9.--Distributions of the subjects on the Guttman 56 scale score of participation. Latin America North Africa/ Asia Middle East Score Frequency Per Frequency Per Frequency Per . cent cent 1 cent 0 2 1.9 4 3.9 1 1.4 l 6 5.9 10 9.9 2 2.8 2 11 10.8 15 14.7 10 14.1 3 12 11.8 14 13.7 8 11.3 4 16 15.7 18 17.7 13 18.3 5 26 25.5 15 14.7 16 22.5 6 17 16.6 17 16.6 12 16.9 7 12 11.8 9 8.8 9 12.7 Total 102 100.0 102 100.0 71 100.0 57 Hypotheses testiog. (l) Hypothesis I. The distribution of the subjects in each of the two groups, the unambiguous commitment and high self-positiveness group and the unambiguous commitment and low self-positiveness group, on the Guttman score of participation in communication activities is shown in Table 10. The median of Guttman scores for each group and the result of the O test for this pair Of groups are also shown in Table 10. Since z_2;1.65 has a one-tailed probability under Ho of p<:.05, our decision is to reject H0 at the 5 per cent level. The distribution of these subjects within each of the three regional categories is shown in Table 11. We can reject H0 at the 5 per cent level for the two regional categories of Latin America and North Africa/Middle East, but we fail to reject H0 at the same level for the Asia category. 58 Table 10.--Distribution of the subjects with unambiguous commitment/high or low self-positiveness on the Guttman score of participation. Unambiguous Commitment Score High Self-positiveness Low Self-positiveness 0 2 4 l 2 7 2 5 l7 3 10 10 4 7 16 5 16 10 6 11 10 7 l6 3 Total 69 77 Median 5.0 3.5 O test 5 = 4.01; p < .001 59 Table ll.--Distributions of the subjects with unambiguous commitment/high or low self-positiveness on the Guttman score of participation for the three regional categories. Latin America N.Africa/M.East .Asia Score Unambiguous Unambiguous Unambiguous Commitment Commitment Commitment High- Low- High- Low- High- Low- self self self self self self 0 l 0 l 3 0 l l 0 2 2 4 0 l 2 1 5 4 9 0 3 3 l 2 3 6 6 2 4 2 4 2 7 3 5 5 3 5 9 2 4 3 6 l 4 6 4 4 2 7 6 0 6 1 4 2 Total 15 22 33 36 21 19 Median 5.3 4.0 5.0 2.8 4.9 4.0 H teSt £32011; £30017 £33003; psoOOl 5.1052; 2.0064 60 (2) Hypothesis II. Table 12 presents the distribution of the subjects among Guttman scores of participation in the ambiguous commitment and high self- positiveness group and that in the ambiguous commitment and low Self-positiveness group. We can reject H0 at the 5 per cent level. If n1:l4 and n2-19, a‘g of 87 or less enables us to reject H0 at the 5 per cent level for a one-tailed test. If n1:15 and n2:16, a O_of 77 or less enables us to reject H0 at the same level. As shown in Table 13, we can reject H0 at the 5 per cent level only for the North Africa/Middle East category. 61 Table 12.--Distribution of the subjects with ambiguous commitment/high or low self-positiveness on the Guttman score of participation. S Amblguous Commitment core High Self-positiveness Low Self-positiveness 0 0 1 l 1 8 2 6 8 3 5 9 4 12 12 5 l8 l3 6 15 10 7 6 5 Total 63 66 Median 4.9 4.1 O test lg - 2.44; p - .007 62 Table l3.--Distributions of the subjects with ambiguous commitment/high or low self-positiveness on the Guttman score of participation for the three regional categories. Latin America N.Africa/M.East Asia Score Ambiguous Ambiguous Ambiguous Commitment Commitment Commitment High- Low- High- Low- High- Low- self self self self self self 0 0 1 0 0 0 0 l 0 4 0 4 l 0 2 1 4 1 1 4 3 3 5 4 0 5 0 0 4 6 4 4 . 5 2 3 5 13 5 l 3 4 5 6 7 5 6 1 2 4 7 2 4 2 0 2 1 Total 34 31 14 19 15 16 Median 4.9 4.1 5.7 3.4 4.6 4.9 U test _z_-1.41; p-.079 _I_J_-46; p<.001 _I_J_-106; p>.05 63 (3) Hypothesis III. For both the over-all groups (Table 14) and the three regional categories (Table 15), we fail to reject H0 at the 5 per cent level for a one- tailed test. The conclusion is that the data do not support Hypothesis III. 64 Table l4.--Distribution of the subjects with high self-positiveness/unambiguous or ambiguous commitment on the Guttman score of participation. High Self:positiveness Score Unambiguous Commitment Ambiguous Commitment 0 2 0 l 2 1 2 5 6 3 10 5 4 7 12 5 16 18 6 ll 15 7 l6 6 Total 69 63 Median 5.0 4.9 65 Table 15.--Distributions of the subjects with high self-positiveness/unambiguous or ambiguous commitment on the Guttman score of participation for the three regional categories. Latin America N.Africa/M.East Asia Score High self- High self- High self- positiveness positiveness positiveness Unambi. Ambi. Unambi Ambi. Unambi. Ambi. commit .a commit .13 commit commit . commit . commit . 0 l 0 1 0 0 0 1 0 0 2 0 0 l 2 1 l 4 1 0 4 3 l 5 3 0 6 0 4 2 6 2 4 3 2 5 3 13 9 l 4 4 6 l 7 6 6 4 2 7 6 2 6 2 4 2 Total 15 34 33 14 21 15 Median 5.3 4.9 5.0 5.7 4.9 4.6 U test §;1.18; pa .119 ‘§=.89; p: .188 EP-985.E= .164 aUnambi. commit.- Unambiguous commitment. bAmbi. commit.- Ambiguous commitment. [W 66 (4) Hypothesis IV. For the over-all groups (Table 16) and for the three regional categories, we again fail to reject H0 at the 5 per cent level for a one-tailed test. Thus Hypothesis IV is not supported by the data. 67 Table l6.--Distribution of the subjects with low self- positiveness/unambiguous or ambiguous commitment on the Guttman score of participation. Low Selfspositiveness Score Unambiguous Commitment Ambiguous Commitment 0 4 l l 7 8 2 l7 8 3 10 9 4 16 12 5 10 13 6 10 10 7 3 5 Total 77 66 Median 3.5 4.1 H test _z_ I 1.62; E I .053 68 Table l7.--Distributions of the subjects with low self- positiveness/unambiguous or ambiguous commitment on the Guttman score of participation for the three regional categories. Latin America N.Africa/M.East Asia Score Low self- Low self- Low self- positiveness positiveness positiveness Unambi. Ambi. Unambi. Ambi. Unambi. Ambi. commit. commit. commit. commit. commit. commit. 0 0 1 3 0 1 0 1 2 4 4 4 l 0 2 5 4 9 l 3 3 3 2 4 6 5 2 0 4 4 4 7 5 5 3 5 5 5 2 3 3 5 6 4 5 4 1 2 4 7 0 4 l 0 2 1 Total 22 31 36 l9 l9 16 Median 4.0 4.1 2.8 3.4 4.0 4.9 _O test 5:.41; p:.34l _z_:.62; 2:.268 _U:llS.5; p>.05 CHAPTER IV INTERPRETATION AND CONCLUSIONS The entire pattern of the data supports the first two hypotheses but does not support the others. The data point to the conclusion that in a dissonant situation, positiveness of self-evaluation is negatively related to selectivity of participation in potentially dissonance-increasing communication activities (Hypotheses I and II). That is, the more positive Self-evaluation, the less selective participation.- It is clear that the high positiveness of self-evaluation acts to obscure the tendency to avoid participation in potentially dissonance-increasing communication activities. In this sense, the data help explain the weak results previously obtained concerning avoidance of dissonance-increasing information. The variable, positiveness of self-evaluation, is of major importance in determining the extent to which one may actually Observe avoidance of dissonant information. In the light of knowledge acquired from this study, therefore, we must come to the conclusion that it is obviously not correct to assert a simple avoidance 69 70 of participation in potentially dissonance-increasing' activities without regard to positiveness of self- evaluation. Avoidance of participation in such dissonance- increasing activities helps in the dissonance-reduction process only for the person with less positive self- evaluation. 0n the other hand, the person with more positive self-evaluation is likely to participate actively in such activities so as to help in the process of dissonance reduction. The specification of self- positiveness as a personality variable in dissonance theory helps to identify more clearly the manner in which dissonance may be reduced and, in turn, may provide the theory with predictive power in many nonobvious realms. If we look at the data a bit more closely, we can find that the evidence for the negative relation- ship between self-positiveness and participation selectivity within the ambiguous commitment group (Hypothesis II) is weak as compared with the evidence for that relationship within the unambiguous commitment group (Hypothesis I). This may indicate that the effect of commitment on the resistance to change of the corresponding cognitive element is lessened if the reality to which the commitment corresponds is ambiguous. However, the evidence for the positive relationship between ambiguity of commitment and selectivity of 71 participation in communication activities within the high self-positiveness group (Hypothesis III) and for the negative relationship between them within the low self-positiveness (Hypothesis IV) is quite weak. Although the evidence is weak, we may be justified in placing some reliance on it because the data show generally the predicted directions of relationships. Thus lack of control over ambiguity of commitment as well as lack of control over positiveness of self-evaluation would produce large variability in the pattern of selective behavior. Although the data suggest that the factor of self-positiveness should be taken into account in making predictions from the theory of dissonance, there may be other pressures which could act to weaken dissonance avoidance tendencies: for example, general curiosity, desire to be well informed, etc. Thus it seems correct to say that dissonance avoidance would be observed with low self-positiveness in situations where other reasons for participationin dissonance-increasing communication activities were absent. Only additional studies will‘ isolate the independent or interactive effects of such pressures. Further research is strongly required to clarify the relations of such pressures to dissonance avoidance or to positiveness of self-evaluation. 72 In many non-English speaking countries, there is a belief that the panacea for English language learning is to throw a person into an English-speaking country and that by being there he will be easily transformed into a good English speaker. However, the data indicate that this is not the case with foreign students whose self-evaluations are less positive. It is evident that only the student with high self-positiveness and unambiguous academic commitment can benefit from participation in a variety of communication activities. Most of the foreign students with low self-positiveness as well as the usual inadequate language preparation would encounter considerable frustration and finally become isolated from American life. The great need for more positive self-evaluation and unambiguous academic commitment should be emphasized on the part of foreign students with objective English language deficiency. A simple recommendation of participation in communication activities for English language learning would not make sense as a part of effective English language program for foreign students. In terms of foreign students with objective English language deficiency, however, generalizations are of course limited because the subjects used are only those foreign students who are male, single, and 18-27 years of age and who arrived in 73 the United States in or after September, 1968 and one month before the survey was made. Further, they are not a random sample from the total papulation of foreign students with objective language deficiency in the United States. Selectivity of participation was also determined by participation in the limited number of communication activities and by participation during a week. Replication with additional demographic, social, and situational data will be required to establish tenable propositions. In terms of participation behavior, the different patterns of participation in communication activities emerged for the three regional categories of Latin America, North Africa/Middle East, and Asia. Although generalizations are limited, the results of Guttman analysis on participation in communication activities may provide additional insights into this area of cross-cultural communication research. It is interesting to note that the item concerning direct contact with American friends scales reliably only for the Latin America category. Another person-to-person type of contact (visiting American homes) characterizes the maximum levels of the Guttman scales for Latin America and North Africa/Middle East categories. No item pertaining to person-to-person 74 Contact scales reliably for the Asia category. This result may make sense in terms of self-positiveness. For the purpose of comparison, let us consider the percentages of subjects in each category with self- positiveness scores of 5 or lower (Table 8). Only 20.6 per cent have scores of 5 or lower in the Latin America category, 28.4 per cent in the North Africa/ Middle East category, and 40.9 per cent in the Asia category. Some of those subjects with low self- positiveness would find encouraging Americans who reinforce their attempts to speak the English language but some would find those Americans who are relatively intolerant of misuse of the English language. In the former case, low self-positiveness would be bolstered and thus frequent contacts with those Americans would be found. In the latter, however, direct contact with Americans would be avoided. If this interpretation is correct, such variability in contact behavior may lead to failure in scaling for the items pertaining to direct contact with Americans in the Asia category. However, this is not the case at least for the Latin America category. Thus the differential patternings of items in the three Guttman scales may suggest that there are cultural patternings in contact behavior. A comparative study of such cultural patternings would be valuable 75 research endeavor. While the patterning of items in the three Guttman scales is unique for each regional category, some trends across categories may be pointed out. For all regional categories, the first three items, though different in order, are the main mass media Of television, radio, and newspapers. That is, in terms of the scaling for the given set of items, the minimum participation experience involves exposure to at least one of these three mass media. The items concerning other media such as movies, magazines, and books occur after the items pertaining to the three mass media of television, radio, and newspapers. This means that subjects who attend movies or read magazines and/or books also tend to expose themselves to television, radio, and newspapers. Of course, replication with additional demographic and situational data will be required to clarify their patterns of exposure to mass media. It is possible to interpret the results of Guttman analysis on participation in communication activities as indicating their perceived usefulness in order to develop language skills. Some activities may be perceived as more useful to develop language skills than others. It seems that subjects are likely to participate in such activities for reasons of usefulness. The perceived usefulness of activities may also act to 76 obscure the tendency to avoid participation in dissonance- increasing activities. There may be possible confounding effects of such perceived usefulness on selective participation. Thus it will be required to incorporate the variable of perceived usefulness in the design of research on selective participation in dissonance- increasing activities. If activities could be arranged so that they were equally useful to the subject, the data would reveal more clearly selectivity of participation in dissonance-increasing activities. It is equally possible, and equally plausible, to interpret the results of Guttman analysis as simply indicating a preference for participation in certain activities. For example, there may be a general preference for exposure to such mass media as television, radio, and newspapers, and conceivably this preference may have existed before the arrival in the United States. The evidence in this study may be obscured by the absence of any comparison between the patterning of participation in the home country and that in the United States. It seems worthwhile to do a study in which one could obtain this comparison. It should be also noted that there may be a positive relationship between foreign students' subjective evaluations of their language skill and their general self-evaluations. 77 The relation of such a specific self-evaluation to general self-evaluation should be explored to clarify positiveness of general self-evaluation employed in this study. Finally, it may be noted that our use of dissonance theory in analyzing a real-life situation provides some confidence in its applicability to the problems of the "real world." Our dissonance interpretation in terms of Self-positiveness and commitment-ambiguity cannot,of course, be ascertained without further study. Nevertheless, this study seems to illustrate a kind of practical implication that dissonance theory should have. BIBLICD RAPHY BIBLIOGRAPHY Adams, J. S. "Reduction of Cognitive Dissonance by Seeking Consonant Information." Journal of Abnormal and Social Psychology, 62(I96I), 74-78. Aronson, E., and Carlsmith, J. M. "Performance Expectancy as a Determinant of Actual Performance." Journal of Abnormal and Social Psychology, 65(1962), 178-82. Bachman, J. G. "Motivation in a Task Situation as a Function of Ability and Control over the Task." Journal of Abnormal and Social Psychology, 6911964) 7271-81 . Bramel, D. "A Dissonance Theory Approach to Defensive Projection." Journal of Abnormal and Social Psychology, 64(1962), 121-29. Brehm, J. W., and Cohen, A. R. Explorations in Cognitive Dissonance. New York: John Wiley & Sons, 1962. Brock, T. C., and Buss, A. H. "Dissonance, Aggression and Evaluation of Pain." Journal of Abnormal and Social Psychology, 65( , - . Brown, E. J. "The Self as Related to FOrmal Participation in Three Pennsylvania Rural Communities."‘ Rural Sociology, 18(1953), 313-20. Brown, R. "Medels of Attitude Change." New Directions in Psychology. Edited by R. Brown, E. Galanter, E. H. Hess, and G. Mandler. New York: Holt, Rinehart and Winston, 1962. Brownfain, J. J. "Stability of the Self-Concept as a Dimension of Personality." Journal of Abnormal and Social Psychology, 47(1952), 597-606. 78 ‘1 I) 79 Burke, R. L., and Bennis, W. G. "Changes in Perception of Self and Others during Human Relations Training." Human Relations, 14(1961), 165-82. Calvin, A. D., and Holtzman, W. H. "Adjustment and the Discrepancy between Self Concept and Inferred Self.’ Journal of Consulting Psychology, 17(1953), 39-44. Cofer, C. N., and Appley, M. H. Motivation: Theory and Research. New York: JOthWiley & Sons, 1967. Cohen, A. R. ”A Dissonance Analysis of the Boomerang Effect." Journal of Personality, 30(1962), 75-88. .. Combs, A. W., and Snygg, D. Individual Behavior. Rev. ed. New York: Harper & Row,l959. Davis, K., and Jones, E. E. ”Changes in Interpersonal Perception as a Means of Reducing Cognitive Dissonance." Journal of Abnormal and Social Psychology, 61(1960), 402-10. Deutsch, M.; Krauss, R. M.; and Rosenau, N. "Dissonance or Defensiveness?" Journal of Personality, 30(1962), l6-28.-- \\~Deutsch, S. E., and Wen, G. ”Some Factors in the Adjustment of Foreign Nationals in the United States." Journal of Social Issues, 19(1963), 115-220 Ehrlich, D.; Guttman, I.; Schonbach, P.; and Mills, J. "Post-decision Exposure to Relevant Information." Journal of Abnormal and Social Psychology, 54(1957), 98-102. Engel, J. F. “Are Automobile Purchasers Dissonant Consumers?" Journal of Marketing, 27(1963), 55.58 c Farace, R. V. “Identifying Regional 'Systems' in National Development Research." Journalism Quarterly, 43(1966), 753-60. {I f.) . - C C ( \ {A 1': fr 80 Feather, N. T. "Cognitive Dissonance, Sensitivity, and Evaluation." Journal of Abnormal and Social Psyphology, 66(1963), 157-63. Feldman, 8., ed. Cognitive Consistenoy. New York: Academic Press, 1966. Festinger, L. "Motivations Leading to Social Behavior." Nebraska Symposium on Motivation, 1954. Edited by M. R. Jones. Lincoln: University of Nebraska Press, 1954. . A Theory Of Cognitive Dissonance. Stanford, Calif.: Stanford University Press, 1957. . "The Motivating Effect of Cognitive Dissonance." Assessment of Human Motives. Edited by G. Lindzey. New York: Rinehart, 1958. . Conflict, Decision and Dissonance. Stanford, Calif.: Stanford University Press, 1964. , and Bramel, D. "The Reactions of Humans to Cognitive Dissonance." The Experimental Foundations of Clinical Psychology. Edited by A. Bachrach. New York: Basic Books, 1962. Gerard, H. B.; Blevans, S. A.; and Malcolm, T. "Self- Evaluation and the Evaluation of Choice Alternatives." Journal of Personality, 32(1964), 395-410. Goode, W. J. ., and Hatt, P. K. Methods in Social Research. New York: McGraw-Hill, 1952. \\\Greenwald, A. G. "Behavior Change Following a Persuasive Communication." Journal of Personality, 33(1965), 370- 91. \\Gullahorn, J. E. "A Factorial Study of International Communication and Professional Consequences Reported by Fulbright and Smith-Mundt Grantees, l947-1957.’ Unpublished Ph.D. dissertation, Michigan State University, 1964. 81 , and Gullahorn, J. T. "American Students Abroad: Professional versus Personal Deve10p- ment.” The Annals of the American Academy of Political and Social science, 368(1966), 43159. \\Jacobson, E.; Kumata, H.; and Gullahorn, J. E. ~"Cross- Cultural Contributions to Attitude Research." Public Qpinion Quarterly, 24(1960), 204-23. Kaufmann, H. "Task Performance, Expected Performance, and Responses to Failure as Functions of Imbalance in the Self-Concept." Unpublished doctoral dissertation, University of Pennsylvania, 1962. Kiesler, C. A., and Sakumura, J. "A Test of a Model for Commitment." Journal of Personality and Social Psychology, 3(1966), 349-53. Kuhn, M. H., and McPartland, T. s.‘ "An Empirical Investigation of Self-attitudes.” American Sociological Review, 19(1954), 68-76. Kumata, H. "A Factor Analytic Study of Semantic Structures across Three Selected Cultures." Unpubliched Ph.D. dissertation, University of Illinois, 1958. , and Schramm, W. ”A Pilot Study of Cross- Cultural Methodology." Public Opinion Quarterly, 20(1956), 229-38. Lazarsfeld, P. F., and Thelens, W., Jr. The Academic Mind. Glencoe, Ill.: Free Press, 1958. Lecky, P. Self-consistency: A Theory of Personaligy. New York: Island Press, 1945. Maclay, and were, E. E. "Cross-Cultural Use of H. the Semantic Differential." Behavioral Science, 6(1961), 185-90. ' McPartland, T. s., and Cuming, J. H. "Self-Conception, Social Class, and Mental Health." Human Organization. 17(1958), 24-29. I‘J f; 82 Mills, J.; Aronson, E.; and Robinson, H. "Selectivity in Exposure to Information.’ Journal of Abnormal and Social Psychology, 59(1959), 250- 53. Mishler, A. L. "Personal Contact in International Exchanges." International Behavior. Edited by H. C. Kelman. New York: Holt, Rinehart and Winston, 1965. . \§cMorris, R. T. The Two-way Mirror: National Status in Foreign Students' Adjustment. Minneapolis: University of Minnesota Press, 1960. Murphy, G., and Likert, R. Public Qpinion and the Individual. New York: Harper, 1938. CBgood, C. E. "Cognitive Dynamics in the Conduct of Human Affairs." Public Qpinion Quarterly, 24(1960), 341- 65. ; Suci, G. J.; and Tannenbaum, P. H. The Measurement of Meaning. Urbana, Ill.: University of Illinois Press, 1957. Pool, I. de S. "Effects of Cross-National Contact on National and International Images." International Behavior. Edited by H.. C. Kelman. New York: Holt, Rinehart and Winston, 1965. Rosen, S. "Pest-decision Affinity for Incompatible Information." Journal of Abnormal and Social Psychology, 63(1961), 188-90. Rogers, C. R. Client-centered Therapy; Its Current Practice, Implications,and Theory. Boston: Houghton-Mufflin, 1951. \\Se11tiz, C.; Christ, J. B.; Havel, J.; and Cook, S. W. Attitudes and Social Relations of Foreign Students in the United States. Minneapolis: University of Minnesota Press, 1963. Sewell, W. H., and Davidsen, O. M. Scandinavian Students on an American Campus. Minneapolis: University of Minnesota Press, 1961. 83 Siegel, S. Nonparametric Statistics for the Behavioral Sciences. New York: McGraw-Hill, 1956. Smith, M. 3., ed. "Attitudes and Adjustment in Cross- cultural Contact: Recent Studies of Foreign Students." Journal of Social Issues,, 12(1956), no.1. Stouffer, S. A.; Guttman, L.; Suchman, E. A.; Lazarsfeld, P. F.; Star, S. A.; and Clausen, J. R. Measurement and Prediction. Princeton: Princeton University Press, 1950. Suci, G. J. "A Comparison of Semantic Structures in American Southwest Culture Groups." Journal of Abnormal and Social Psychology, 61(1960), 25-30. Tanaka, Y.; anma, T.; and CBgood, C. E. "A Cross- , Cultural and Cross-Concept Study of the Generality of Semantic Spaces." Journal of Verbal Learning_and Verbal Behavior, 2(1963), 392-4050 ‘ ‘ Triandis, H. C., and Osgood, C. E. "A Comparative Factorial Analysis of Semantic Structures in Monolingual Greek and American College Students." Journal of Abnormal and Social Psychology, 57(1958), 187-96. Vroom, V. H. Werk and Motivation. New York: John Wiley & Sons, 1964. Zajonc, R. B. "The Concepts of Balance, Congruity, and Dissonance." Public Qpinion_guarter1y, 24(1960), 280-96. Zimbardo, P. G. "Involvement and Communication Discrepancy as Determinants of Opinion Change." Journal of Abnormal and Social Psychology, 60(1960), 86491. APPENDIX APPENDIX QUESTIONNAIRE Please complete all of the questions. 1. 2. 5. 6. 7. 10. Are you enrolled in full time work in English? Yes. No. When did you enter this English language training program? Month: Year: When did you arrive in the United States? Month: Year: I Are you: A Male Female Single Married What is your age? What is your home country? Are you planning to study at an American university or college after completing study of English? Yes. No. Have you submitted an application for admission to that university or college? Yes. No. Have you been approved for admission to that university or college? Yes. No. About your major work, have you consulted with your academic adviser in your major department at that university or college? Yes. No. 84 85 11. Please think back over the last week. Did you participate in the following activities last week? Please place an "X" in the space representing the. statement which indicates your participation in each activity the best. Also, please place only one ”X" for each activity. No, not Yes, but Yes, Yes, at all very sometimes frequently seldom Talking with American friends Talking with friends from other countries (not your home »country) Visiting. American homes Attending lectures : Attending plays Attending movies Listening to the radio watching TV Reading American Books (not text-books) Reading American magazines : Reading American neWSpapers 86 12. We are interested in your feelings about yourself. The first thing to do is to look at the a jectives at each end of the scale. Then, decide which one fits. For example, if you feel that "I am very strong," place an ”X' in the space next to the word "Strong." If you feel that "I am quite weak," place an "X" in the second space from the word "Weak.” The other spaces represent degrees of your feelings about yourself. (EXAMPLE) Very Quite Slightly (Neutral) Slightly Quite very Strong—1L5 : :0 P: : :____Weak Strong____5_____3 : : : X :____Weak Please indicate your feelings about yourself by placing one "X" on each scale. ‘We want your first impression, so go through the scales rapidly. V Q S N S Q v e u 1 e 1 u e r I 1 u I 1 r y t g t g t Y e h r h e t a t 1 1 1 Y Y Bad . . . Good Important : : : : : : Unimportant Sad : : : : : : Happy Dishonest : : : : : : Honest Unfriendly Friendly : : : : Unsuccessful : : : : : : Successful ‘111111111111