g-xv' n . .‘\.‘4.‘\-:-,A A vn . .H. . ... .., ‘ ‘ \ .M'V'".’\'I‘T‘-f‘,vl- v'xf‘b- . v"- I'~.'n. \ I'.‘.5'.‘.-K‘ ‘4‘V'V‘y‘.‘.'.'.| '. glib-I‘Ilu:::v’1'-'¢'.'v'1'.’.',‘ I'.'.‘.‘ ‘1".',_' ‘3‘" ‘ ' - 3 -’-'r’ ' ' ‘.' ' ' ‘ .. ‘ > TVA A -v . _ ..»......,.... ., ~.-- ' “raw DIFFERENCES I IN HETEROPHILY AND COMMUNICATION INTEGRATION BETWEEN MODERN AND TRADITIONAL INDIAN VILLAGES IN TWO TYPES OF DYADIC ENCOUNTER Thesis for the Degree of Ph. D. 7 MICHIGAN STATE UNIVERSITY DILIP KU‘MAR BHOWMIK A " 11972 - ' - I I I'" e 1’ LIBRARY Michigan State University This is to certify that the thesis entitled DIFFERENCES IN HETEROPHILY AND COMMUNICATION INTEGRATION BETWEEN MODERN AND TRADITIONAL INDIAN VILLAGES IN TWO TYPES OF DYADIC ENCOUNTER presented by Dilip K. Bhowmik has been accepted towards fulfillment of the requirements for Ph. D. degree in Communication Date Z//5)//702' I 0-7639 ‘2" v "J BIN?! NG BY mmaaswe mm mm me. ‘ LIBRARY BINDERS trimaran-1mm a“ _.._-—p————n—-—‘__-_-,— "Fw'< m— >4. ABSTRACT DIFFERENCES IN HETEROPHILY AND COMMUNICATION INTEGRATION BETWEEN MODERN AND TRADITIONAL INDIAN VILLAGES IN TWO TYPES OF DYADIC ENCOUNTER BY Dilip Kumar Bhowmik The present dissertation deals with two major issues related to interpersonal communication: 1) heterophily among interacting dyads, and 2) communication integration in information-seeking and friendship communication. In formulating the problem, four questions were raised about communication systems in rural Indian villages. They are: (1) Do heterophily relationships between sources and re- ceivers differ from one communication situation to another? (2) How are the heterophily relationships between sources and receivers related to the modernity levels of rural vil- lages? (3) Does the extent of integration through interpersonal contacts differ from one communication situation to modern- ity levels of rural villages? Dilip Kumar Bhowmik (4) How is the degree of integration through inter- personal contacts related to modernity levels of rural villages? Thus, the major objectives of the present study were: 1) to compare the degree of heterOphily in information-seeking and friendship communication; 2) to compare the degree of communication integration in information-seeking and in friendship communication; and 3) to find out the relationships between modernity levels of rural Indian villages and (a) heterophily, and (b) com- munication integration. Data used in the present study come from the Indian part of a larger study, "Diffusion of Innovations in Rural Societies, conducted in Brazil, India, and Nigeria, by the Department of Communication at Michigan State University. The present study is based on the analysis of the data from the second phase of the Indian research project, in which the purpose was to determine the factors affecting the in- novative behavior of Indian farmers in rural settings. Data were obtained from 680 farmers in eight villages lo- cated in three states (Andhra Pradesh, Maharashtra, and West Bengal), through personal interviews. Dilip Kumar Bhowmik Six general hypotheses were presented in the present study. Of the six hypotheses, three hypotheses deal with heterophily relationships of source-receiver dyads, and the other three deal with communication integration in information-seeking and friendship communication. In the present study heterophily is measured as the absolute difference between interacting individuals on se- lected variables. Attempt was made to select those vari- ables on which low heterophily is relevant in friendship communication and high heterOphily is relevant in information-seeking communication. After obtaining each dyad's heterophily scores on the selected variables, factor analysis was used to determine heterophily dimensions. Of the six hypotheses, only two were supported by our data. They are: 1) There is a greater degree of het- erophily with respect to certain relevant attributes, among dyads engaged in information-seeking communication than in friendship communication; 2) communication integration in friendship communication is higher in more modern villages than in more traditional villages. Although no hypotheses were formulated comparing dyads engaged in both information- seeking and friendship communication with dyads engaged only Dilip Kumar Bhowmik in friendship communication, our findings suggest that low heterophily in status dimension is relevant and perhaps necessary for friendship communication to occur, and high heterophily in change contact dimension is relevant and might be necessary for information-seeking communication to occur to rural villagers in India. DIFFERENCES IN HETEROPHILY AND COMMUNICATION INTEGRATION BETWEEN MODERN AND TRADITIONAL INDIAN VILLAGES IN TWO TYPES OF DYADIC ENCOUNTER BY Dilip Kumar Bhowmik A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Communication 1972 ACKNOWLEDGEMENTS The accomplishment of this dissertation in the present form is indeed due to invaluable assistance and helpful suggestions received from several sources which I can only acknowledge. I am greatly indebted to my academic advisor and dissertation director, Dr. Everett M. Rogers, who provided intellectual stimulation, encouragement and friendly guid- ance throughout the present study. I express my gratitude to Professor Edmond W. Alchin, Professor R. Vincent Farace, Professor Gerald R. Miller, and Professor Lawrence E. Sar- baugh, who served as members of my guidance committee and offered valuable suggestions and comments in regard to the present study. I am grateful to Professor R. Vincent Farace for the significant methodological suggestions especially deal- ing with measures of heterophily among dyads. I wish to express sincere appreciation to my fellow graduate students in the Department of Communication, espe- cially Mr. K. S. Sripada Raju, and Lytton Guimaraes, with ii whom I used to have stimulating discussions on subjects re- lated to the present dissertation. Special thanks are due to Ms. Anita Immele, and to Joanne Helfrich for their indefinable help with computer programings. I wish to thank Professor Carroll J. Bourg of Fisk University, for sparing some of my faculty time for the preparation of the manuscript of this dissertation and for his editorial help. To "Shonie" and "Ruchi" my deep gratitude for their sacrifice of "play with daddy." My deep sense of gratitude to Shubha, my life partner, for her encouragement, help and cooperation all through my graduate work. Finally, I wish to express my warmest regards to 2393 and Ma, my parents, without whose motivation, aspira- tion and support for higher education, I would have been a drop-out before completing high school. iii Chapter I. II. TABLE OF CONTENTS INTRODUCTION AND THEORETICAL BACKGROUND. . . . Objectives. . . . . . . . . . . . . . . . . Information-Seeking and Friendship' Communication . . . . . . . . . . . . . . Source-Receiver Relationships: Heterophily in Information-Seeking and Friendship Communication . . . . . . . . . . . . . . A Conceptual Framework. . . . . . . . . . . Modernization, Level of Modernity, and Heterophily . . . . . . . . . . . . . . . Modernization and Communication Integration HYPOTHESES AND RATIONALE . . . . . . . . . . . Heterophily in Information-Seeking and Friendship Communication. . . . . . . . . Selection of Relevant Variables for Heterophily Analysis in Indian Villages . Information-Seeking and Friendship Communication and Village Modernity . . . Communication Integration and Village Medernity O O I O O O O O O O O O O O O 0 iv Page 11 14 20 25 29 29 32 41 44 TABLE OF CONTENTS (cont.) Chapter III. RESEARCH METHODOLOGY . . . . . . . . . . . . . Data and the Sample . . . . . . . . . . . . Description of the Villages Studied . . . . Instrument Construction and Data Collection Operationalization of Concepts. . . . . . . Data Analyses and Tests of Hypotheses . . . Iv. FINDINGS O O O O O O 0 O O O O O O O O O O O O Heterophily in Information-Seeking and Friendship Communication. . . . . . . . . HeterOphily in Information-Seeking and Friendship Communication and Village Modernity . . . . . . . . . . . . . . . . Communication Integration in Information- Seeking and Friendship Communication. . . Communication Integration and Village MOdernity O O I O O O O O I O O O O O O O V. SUMMARY AND CONCLUSIONS. . . . . . . . . . . . sumry O O O O O O O O O O O O O O O O O 0 Conclusions . . . . . . . . . . . . . . . . Implications for Change Agencies. . . . . . Suggestions for Future ReSearch . . . .., . BIBLIOGMPHY O I O O O O O O O O O C O O I O O I O O 0 .APPENDIX. Page 47 47 50 54 75 85 85 96 99 100 103 103 112 113 115 119 131 Table LIST OF TABLES Page Location and size of sample villages for the Phase II Study in India . . . . . . . . . . 51 Zero-order product-moment correlation and average absolute differences as measures of heterophily among some hypothetical dyads in seven hypothetical villages. . . . 63 Percent of total variance explained by the two rotated solutions meeting a Kiel- Wrigley criterion for terminating the factor rotation. . . . . . . . . . . . . . . 67 Heterophily scores on nine selected variables and factor loadings in the three-factor SOlution O O O O O O O O O O O O O O O O O O 68 Rank-order of eight villages according to level of modernity . . . . . . . . . . . . . 73 Number of total, mutual, and common dyads in information-seeking and friendship communication. . . . . . . . . . . . . . . . 76 Data-analyses and hypotheses-testing . . . . . 80 Average heterophily scores and standard deviations of heterophily scores, and the . number of dyads involved in different types of communication situations. . . . . . . . . 86 Computed "t" values for differences between mean heterophily scores on each heterophily dimension for dyads involved in different communication situations . . . . . . . . . . 89 vi LIST OF TABLES (cont.) Table Page 10. Calculated "t" values for differences in mean heterophily on the selected variables among dyads in different communication situations . . . . . . . . . . . . . . . . . 93 ll. Calculated "t" values for differences in mean heterophily on the selected variables among dyads in different communication situations. 94 12. Spearman rank—order correlation between the degree of heterophily in information-seeking communication, and in friendship communica- tion, and village modernity. . . . . . . . . 97 13. Communication integration scores in information-seeking and friendship communi- cation in eight Indian villages. . . . . . . 101 14. Summary of results of the hypotheses tested in the present study . . . . . . . . . . . . 107 vii LIST OF FIGURES Figure Page 1. Venn Diagrams showing the relationship between information-seeking and friendship communication. . . . . . . . . . . . . . . . 10 2. Hypothesized degree of heterophily on dimen- sions composed of selected variables in information-seeking and friendship communi- cation in modern and traditional Indian villages . . . . . . . . . . . . . . . . . . 3O 3. Mean heterophily on status, change contact, and movie exposure for dyads engaged in different communication situations . . . . . 88 viii CHAPTER I INTRODUCTION AND THEORETICAL BACKGROUND The present dissertation deals with two major issues related to interpersonal communication: 1) heterophily among interacting dyads, and 2) communication integration in information-seeking and friendship communication in Indian villages. In recent years in the developing world, a common phenomenon is the process of modernization1 through the diffusion of innovations. Interpersonal communication is the major channel through which messages about innovations flow to the majority of rural people.2 Pye (1963, p. 27) suggests that "the process of development is less dependent upon increased investment in the modernized, urbanized, or mass media system than it is upon the adjusting of the in- formal, rural systems to each other and to the mass media system." lModernization is the process through which systems change from a traditional way of life to a more complex, technologically-oriented way of life. 2And, in fact, to most other audiences, like medical doctors in the U.S. (Coleman and others, 1966). 1 Before any adjustment of the "informal" rural sys- tems to each other and to the mass media system is possible (as suggested by Pye), an understanding of the informal com- munication systems1 in rural societies is essential. For an understanding of these rural communication systems, the knowledge of the existing source-receiver relationships (which account for many aspects of communication, such as credibility, persuasibility, etc.), and the links of inter- personal contacts through which messages pass through in these rural societies are essential. Guimaraes (1970a, p. 1) points out that in actuality many communication studies are studies of individual behavr igr rather than communication behavior. Coleman (1958) rea- soned that "survey research approaches were in part respon- sible for the heavy foCus on individuals, rather than their relationships." Communication is a process of coupling or‘ linking at least two individuals in a situation for"inter- change of ideas, opinions, expressions, etc. The "coupling" or "linking” between two individuals is better understood in terms of their relationships: 1 O l ’0 I A communication system conSists of a set of commun- icating (or potentially communicating) units (Guimaraes, 1970b, p. 8). Pointing to the importance of source-receiver rela- tionships, Berlo (1960, pp. 53-54) suggests, "A great por- tion of communication theory must be dyadic in nature. Our discussion and analysis must be phrased in terms of the relationships between communication ingredients, rather than in terms of the values of a particular ingredient for a given person." Typically, in most experimental communi- cation research: Studies looking at the influence of source attributes have controlled the influence of re- ceiver differences either by using a variety of receivers assigned randomly to sources or by us- ing only one kind of receiver. Studies of the influence of receiver attributes have controlled the influence of source differences in similar ways. Few studies have looked systematically at source-receiver interaction effects on communi- cation success (Joyce; 1970, p. 4). In any social system, an individual has a choice of the individuals with whom he may interact. An individual's choice of interacting with "someone" rather than others may also differ from one communication situation to another. Simons and others (1970) suggest that source-receiver rela- tionships on certain attributes are psychologically relevant in persuasive communication. We raise the question of whe- ther individuals, in deciding with whom to interact (in a particular communication situation), make their choices on the basis of their relationship on relevant (relevant to the particular communication situation) attributes or var- iables. In other words, do high or low degree of heter- ophilyl between source and receiver on certain attributes become more crucial in certain communication situations than in others? Along with the process of modernization, all ,aspects of human activity undergo transformation at the same time (Black, 1966, p. 9). Rural, informal communica- tion systems in developing societies had been stabilized for generations. The pattern of who interacts with whom had been "informally formalized" over time. "Communication flows almost entirely horizontally" in traditional villages (Rao, 1966, p. 43). What is the extent of difference be- tween modern and traditional villages regarding source- receiver relationships? As mentioned earlier, an understanding of the in- formal rural communication systems is not complete, without the knowledge of the degree to which the individuals in these rural societies are interconnected through interper- sonal communication links. Are the rural societies so 1HeterOphilyis the degree to which pairs of inter- acting individuals are dissimilar in attributes. integrated that once a message reaches a few members, it diffuses through a maze of interpersonal relationships or is it restricted within certain distinct groups? Studies dealing with integration or cohesiveness have been limited mostly in small groups (Blau, 1960; Lott and Lott, 1965; Cartwright, 1968; and Schachter, 1968). Only in recent years, some communication researchers (Rao, 1966; Yadav, 1967; and Guimaraes, 1970b) have given some emphasis on studying integration through interpersonal communication in larger systems. The questions—-as the process of mod- ernization progresses in rural societies, how the members adapt to the new environment in terms of their communica- tion behavior and in their interpersonal relationships-- have not yet been explored fully. Thus, the four basic questions regarding rural communication systems presented in the present section are: 1. Do the homOphily-heterophily relationships between sources and receivers differ from one communication situation to another? 2. How are the homophily—heterophily relationships between sources and receivers related to the levels of modernity1 of the rural systems? 3. Does the extent of integration through interpersonal contacts differ from one communication situation to another? . . . . . 2 4. How is the degree of communication integration re- lated to the levels of modernity of rural systems? Objectives The main objectives of the present study are: 1. To compare the degree of heterophily in information- seeking and in friendship communication. 2. To compare the degree of communication integration in information-seeking and in friendship communica- tion. 1Level of modernity of a system is its relative condition or state in the modernization process compared to other systems at a particular point of time. 2 . . . . . . Communication integration is the extent to which units of a social system are interconnected through inter- personal communication links. 3. To compare the degree of (a) heterophily and (b) communication integration in modern and in tradi- tional villages. Information-Seeking and Friendship Communication . l . . . Information -seeking communication occurs when an individual interacts with another to obtain information, advice or evaluation for making certain decisions. Friend- ship communication occurs when an individual interacts with another to have an informal and intimate affective associa- tion. On the basis of the kind of motivation which ini- tiates the interaction, Festinger (1950) differentiates be- tween two types of communication: "instrumental" and "con- summatory." In instrumental communication, the need is to "reduce the discrepancy that exists between source and re- ceiver, whereas in consummatory communication the reduction 1Information is transferred patterned matter/energy between any system and its environment or between elements in a system, leading to either increasing or decreasing the number of perceived alternatives in any decision-making situ- ation and/or providing logic for the alternatives (motiva- tional) and/or how to attain the alternatives (instructional) (Berlo, 1969). of the force to communicate occurs as a result of the ex- pression and does not depend upon the effect it has on the receiver." On the same line of thought, Bordenave (1966, p. 9) differentiates instrumental and consummatory content on the basis of their application: 1) instrumental content is used to modify an individual's behavior; and 2) consum- matory content is used to produce a sensation of well-being in oneself. Blau (1962) classified interpersonal relationships on the basis of: l) purpose of an interactional choice and 2) whether the choice represents an actual interaction or a preference. On the basis of purpose, Blau (1962) differen- tiated between social and instrumental interaction. Social interaction occurs when an individual interacts with another (or prefers to interact with another) person to derive sat- isfaction from the association and not for the promotion or achievement of a specific purpose. Instrumental interac- tions are basically goal oriented; that is, there is at least a specific purpose of promotion or achievement in- volved in the initiation of such interaction. In a recent study, Lionberger and Campbell (1971) followed Blau's (1962) classification and studied social and informational rela- tionships among farm Operators in a Missouri community. Although, conceptually instrumental and social interactions are considered mutually exclusive, empiric- ally it is hard to consider them as such (Figure 1). Berlo (1960, p. 17), pointing toward the difficulty of the assumption of mutual exclusiveness of instrumental and consummatory purposes of communication, suggests that there is a need to place the purpose of communication on a consummatory-instrumental continuum rather than to re- gard it as a dichotomy.2 Thus, the assumption that friend- ship communication is always consummatory is questionable. We assume that in friendship communication, the content tends to be more consummatory than instrumental. 1Most interactions between a student-teacher dyad, say A and B, are instrumental; suppose both A and B live in the same neighborhood, they may have also some social interactions between them. 2Although we agree to the notion of a consummatory- instrumental continuum, our measures for information- seeking and friendship communication are dichotomous, rather than continuous. 10 Information-Seeking Friendship Communication Communication Fig. l.--Venn Diagrams showing the relationship between information-seeking and friend- ship communication. ll Source-Receiver Relationships: Heterqphily in Information-Seeking and Friendship Communication1 Source-receiver relationships are conceptualized in terms of who interacts with whom. One way of organizing the analysis is to determine the relationships between the "who" and "whom," that is, to determine the degree of heter- ophily between source and receiver. Heterophily is the degree to which pairs of inter- acting individuals are dissimilar in attributes. The terms "homophily" and "heterophily" were orig- inally used by Lazarsfeld and Merton (1954), and have since then been used by several communication researchers (Chou, 1966; Yadav, 1967; Ho, 1969; and Sen, 1969); some social scientists, however, have mainly used "similarity- dissimilarity" in their studies (Blau, 1962; Byrne, 1961; Jones and Daughtery, 1959; Runkel, 1956; Triandis, 1959; and many others). Homophily-heterophily is conceptualized at the individual level (Feldman, 1966), dyadic level (Chou, 1966; and Yadav, 1967), and system level (Ho, 1969), with individuals, dyads, or social systems as the units of analy- sis, respectively. In studying friendship relationship, 1Some aspects of heterophily discussed in this section are from Rogers and Bhomik (1971). 12 Lazarsfeld and Merton (1954) used "homophily" and "heter- ophily" as dichotomy of relationship. Homophily is the de- gree to which pairs of interacting individuals are similar in attributes. In the present study, homophily-heterophily are conceptualized as polar concepts. If there is a greater degree of heterophily between individuals A and B, then there is a lesser degree of homophily abetween A and B. Similarly, if there is a lesser degree of heterophily be- tween A and B, then there is a higher degree of homophily between A and B. Studiesl dealing with similarity-dissimilarity of dyads, interpersonal attraction, and interaction suggest that (1) similarity between individuals (on values, atti- tudes, cognitions, status, prestige, etc.) is related to interpersonal attraction, and (2) interpersonal attraction is related to frequency and/or effectiveness of communica- tion. In persuasion research, it is widely assumed that communicators who are similar (that is, homophilous) to their audiences are more likely to effect persuasion than 1For example, Barlund and Harland (1963); Blau (1962); Broxton (1963); Byrne (1961); Festinger and others (1950); Larsen and Hill (1958); Lazarsfeld and Merton (1954); Newcomb (1956); Runkel (1956); Smith (1957); and Triandis (1959). For a general review of studies on similarity-dissimilarity, see Simons and others (1970). 13 those sources who are dissimilar; that is, sources having heterophilous relationships with receivers are less effec- tive in persuasion (Berscheid, 1966; Bettinghaus, 1968; Brock, 1965). More effective communication of meaning is found among individuals who belong to the same generation or have similar past experiences (Wood and others, 1971; Vick and Wood, 1969). In contrast to the proposition about homophily-persusasion, Simons and others (1970) sug- gest that some degree of heterophily between sources and receivers makes the sources more credible and competent, thereby leading to greater persuasion. Klapper (1960, pp. 34-35) describes the opinion leader1 as "a kind of super representative of his own group"; he is more competent within his specialty and has access to wider sources of pertinent information. Lazars- feld and Menzel (1963) found that opinion leaders from whom information was sought were generally more competent, more interested, better informed and more gregarious than the seekers, and had access to wider sources of information. 1An opinion leader is an individual who is sought by more members of a system than others, for information, advice, or opinion. Studies dealing with opinion leaders have often used monadic analysis. An analysis is monadic when a concept is defined in terms of one person without reference to his relationships with others and individual is used as the unit of analysis. l4 Lionberger (1953, 1955) noted that opinion leaders function as such because they are more competent than those who seek information.1 In a study of medical opinion leadership among the middle-aged and elderly, Booth and Babchuk (1972) found that the active opinion leaders were different in a number of ways from opinion-seekers; they were well-read and had first-hand knowledge of the health facilities or personnel about which they offered advice. Greater simi- larity among friends than among non-friends with respect to a variety of issues have been reported by several re- searchers (Bonney, 1946; Loomis, 1946; Newcomb, 1956; Precker, 1952; Richardson, 1940; Winslow, 1937). A Conceptual Framework The studies of similarity-dissimilarity between sources and receivers, persuasion, opinion leadership, and friendship, provide somewhat conflicting evidence about heterophily relationships between sources and receivers. For a better understanding of such conflicting evidence 1Rogers with Svenning (1969, p. 237) observes a general tendency for information—seekers to obtain informa- tion from opinion leaders who are somewhat more competent in technical knowledge and more innovative. 15 of source-receiver relationships, some basic questions need to be raised: Heterophily on what, and why? Pointing to the prevalent notion that similarity between individuals leads to greater interpersonal attrac- tion, Newcomb (1956) suggests: It [the notion of similarity leading to greater interpersonal attraction] is not a useful notion, however, because it is indis- criminate: We have neither good reason nor evidence for believing that persons of simi- lar blood type for example . . . are specially attracted to each other. The answer to the question "Similar to what?"--is enormously complex. Though the answer to the question--homophily- heterophily on what?--i§ complex, attempts need to be made to find out the basic relevance of attributes in particu- lar communication situations. Simons and others (1970) differentiate between relevant and irrelevant similarity- dissimilarity (on attributes) on the basis of how similarity-dissimilarity on certain attributes are "psycho- logically relevant" in a particular communication situation. Joyce (1970, pp. 1-2) posits that in a communication situa- tion, where source and receiver are completely alike--a rare event--the source will be able to transmit a message 16 with perfect fidelity, but the message will be totally re- dundant and it will have no information for the receiver.1 Berlo and others (1969) in their factor analytic study, found three independent dimensions for the construct of "dimensions for evaluating message sources": Safety, qualification, and dynamism. In using the three dimensions as an index of source credibility, they suggest the follow- ing scales as most representative: 1. Safety: Safe-unsafe; just-unjust; kind-cruel; friendly-unfriendly; honest-dishonest; 2. Qualification: Trained-untrained; experienced- inexperienced; skilled-unskilled; qualified- unqualified; informed-uninformed; 3. Dynamism: Aggressive-meek; emphatic-hesitant; bold-timid; active-passive; energetic-tired. In any communication situation, in making a choice with whom to interact, individuals evaluate others on the safety, qualification, and dynamism dimensions. 'The 1The amount of information one person can give another about a phenomenon is defined as equivalent to the difference between their perceptions of that phenomenon (Rapoport, 1953, pp. 54-55, as quoted in Joyce, 1970). 17 question of "heterOphily on what, and why" can be explained by considering the importance of some dimensions for eval- uating sources in a particular communication situation. Although we know that individuals evaluate others as mes- sage sources, we do not know what characteristics of others become relevant, and what kind of relationship (degree of heterophily regarding the relevant characteristics) make others safe, qualified, and dynamic. Are the three dimen- sions of evaluating sources equally valued in different communication situations, such as information-seeking and friendship communication? Individuals who are sought for information, advice, or opinion, are trained, experienced, skilled, qualified, and informed, or at least that is how they are perceived by the information seekers. Safety dimensions include a general evaluation of the affiliative relationships between interacting individuals. When the purpose of communication is to seek information, advice, or opinion, more importance . . . . . . . . 1 or weight is given to qualification dimenSion than others, 1Although in evaluating a source all three dimen- sions are considered, some are more emphasized in one com- munication situation than in another. The dynamism dimen- sion of the source is probably equally emphasized in both information-seeking and friendship communication. bl. va as 18 in evaluating sources. In friendship—communication, it is the safety dimension which is given more weight than others. An individual should be more qualified, more trained, more experienced, more skilled, and more informed, to be perceived as a qualified source in information-seeking communication. Thus, there has to be some high degree of heterophily between information-seekers and soughts accord- ing to certain attributes which are considered to be the differentiating attributes between qualified and unquali- fied individuals. The attributes which differentiate be- tween the qualified and unqualified may vary from one so- cial system to another, and from one point of time to an- other. To be perceived safe, friendly, kind, etc., a lower degree of heterOphily is desired. The relevant attributes on which low heterophily is needed depend on social norms and values. It is evident from the previous discussion that for a source to be qualified, there should be some high degree of heterophily between the source and his receiver. But in certain types of information-seeking, the safety dimension is much more emphasized, such as, seeking information about abortion, venereal diseases, and other taboo topics. In a 19 study of seeking an abortionist, Lee (1969, p. 144) found that "both numerically and socially, the most important channels" used in seeking the "invisible" service of an abortionist, were "intimate 'friends'--equal age and equal status contact between people who voluntarily associate with each other, sharing leisure time activities." Lee (1969, p. 141) also observed that the sources who were of different generations (parents, children), different author- itative levels (employers, teachers, school authorities, etc.) were intentionally avoided by the abortion-seekers. Thus, when the information sought is highly confidential, the safety dimension becomes more important in evaluating message sources. Duff (1971) observed that in mixed residential areas in a Philippino city, the lower-middle class residents function as sources of information for their lower-class neighbors, increasing the quality and quantity of informa- tion (about family planning) circulating within the lower- class neighbor groups, whereas in a segregated (all lower- class residents) area, the residents were relatively de- prived with respect to information on family planning. Duff's study (1971) suggests that some high degree of heterophily in socio-economic status between source-receiver 20 in the Philippines facilitates rapid diffusion of certain confidential type of information or ideas. In another study of strawberry growers in Canada, Alleyne and Vernor (1969, pp. 54-56) found (1) a high degree of heterophily among dyads on the basis of their innovativeness, and (2) a low heterophily on the basis of ethnic origin of the dyads engaged in information-seeking communication. Alleyne and Verner's (1969) findings suggest that the qualification dimension was attributed to innovativeness, whereas the safety dimension was attributed to ethnic origin. Considering the three dimensions of evaluating sources, one may assume that in any communication situation, both high heterophily and low heterophily on relevant attri- butes are necessary conditions for any communication to occur. The empirical questions needing answers are: How much, on what attributes, in which communication situations (if information-seeking, what type of information)? Modernization, Level of Modernipy, and Heterophily Modernization has been conceptualized in many dif- ferent ways by social scientists. A precise common meaning 21 for modernization has yet to be evolved (Weiner, 1966, p. 2; and Sen, 1968). Weiner (1966, p. 3) summarizes how dif- ferent social scientists view modernization: Economists view modernization primarily in terms of man's application of technologies to control natural resources to bring growth of per capita income. Sociologists and social anthro- pologists have been primarily concerned with the process of differentiation that characterizes modern societies. Political scientists have emphasized the problem of nation and government building as modernization occurs. Black (1966, p. 9) suggests that "all aspects of human ac- tivity have been undergoing transformation at the same time, and the process of modernization is too complex to be re- duced to simple terms without the danger of grave distor- tion." Rao (1966, p. 7) uses the term "development" to refer to "the complicated pattern of economic, social, and political changes that take place in a community as it progresses from traditional to a modern status." Rogers with Svenning (1969, p. 14) define moderniz- ation as "the process by which individuals change from a traditional way of life to a more complex, technologically- oriented and rapidly changing way of life." We conceptu- alize modernization as a multi-dimensional process, and define modernization as the process through which systems (individuals, villages, nations, etc.) change from a 22 traditional way of life, to a more complex, technologically- oriented way of life. Level of modernity of a system is its relative condition or state in the modernization process as compared to others at a particular point in time. It is realized that as the process of modernization proceeds, various aspects of human activity continually in- teract and undergo transformation (Black, 1966, p. 9). Sel- dom have diffusion researchers inquired about what happens to interpersonal relationships between sources and receivers as the process of modernization progresses in a social sys- tem. Yadav (1967) found lower heterophily (a higher degree of homophily) with respect to innovativeness and agricul- tural knowledgeability in information-seeking communication in a traditional social system than in a modern system. Rogers with Svenning (1969, pp. 230-231) found that in information-seeking communication, there was a higher degree of heterophily (with respect to innovativenss and social status) between sources and receivers in modern systems than in traditional systems. Van den Ban (1963) reports a higher degree of heterophily with respect to innovativeness in modern than in traditional systems in the Netherlands. The communication process in traditional societies is intimately related to the basic structure of these 23 societies. In contrast to modern systems, in traditional systems information flow from "who" to "whom" is largely dependent on ascribed status relationships and personal ties. Reliability of messages is accounted for by personal relationships between interacting individuals. As Pool (1963, p. 242) put it, "Distrust of those who are not in one's own family, tribe, or caste dominate any objective test of truth [of messages] in most economically non- expansive societies." In traditional societies, communi- cation flows almost entirely horizontally and is limited within distinct peer groups (Rao, 1966, p. 43). Although Rao (1966, p. 100) suggests that "the in- formed person has always commanded respect in traditional societies," the criteria (ascribed vs. achieved) used to evaluate the "informed persons" in traditional societies are different than in modern ones. Most communication flows (both information-seeking and friendship communica- tion) in traditional villages are restricted within closely- knit family, peer, and caste groups. In traditional sys- tems, where almost all human activities are somewhat repeti- tious, generation after generation, persons with higher ascribed statuses (such as age) are judged more experienced, qualified, skilled, and more "informed." 24 Modernization, along with its various components (economic, political and social development), brings a new environment and a whole "new psychology." Education be- comes more popular, individuals are more exposed to mass media and more people desire to be informed. The new en- vironment brings a marked change in interpersonal communi- cation patterns. "Communication helps shift influence from age and traditional status to knowledge and ability" (Rao, 1966, p. 100). Individuals seek information from those who are informed; the differentiating attributes of informed and uninformed persons being evaluated by their level of education, mass media exposure, knowledge, etc. Thus, a new set of characteristics becomes relevant in evaluating the qualification dimension of informed sources. Although, in general, friendship communication tends to be directed to those who live in similar situa- tions, enjoy similar status and are similar, the boundary of one's peer group is broadened from the closely-knit family and caste groups, etc., to the larger community, as the systems modernize. Any communication between two SEE? pletely dissimilar individuals is dissonance-producing; this is more so when the purpose of interaction is to have affective, informal association. But with the new inputs 25 of modernization (education, mass media exposures, and a general awareness of community, etc.), individuals are able to have friendship interaction with others who are somewhat dissimilar. In other words, a change in the range of dif- ferences in the criteria for evaluating safety dimensions of sources takes place as systems modernize. Modernization and Communication Integrationl Integration has long been a subject of interest for social scientists. Studies dealing with integration can be broadly grouped under two headings: 1) integration of so- ciety, nation or state, and 2) interpersonal and group in- tegration. Durkheim (1960), Sorokin (1937, 1951), and Smend (1928)2 are the pioneers in the study of integration in so- ciety. Social integration has been conceptualized as: 1) consensus of beliefs and values among members of society and 2) interdependence of members in terms of services (Durkheim, 1960). 1The present review of studies dealing with integra- tion is largely based on Guimaraes (1970b). 2Based on an article about some of Smend's works on integration theory by Landecker (1950). star .0 HI) EEO ter hes grc tra ‘ . .Ic tn: $11 26 In small group research, group integration has been studied in terms of group cohesiveness. Blau (1960) defines group cohesiveness "as the prevalence of integrative bonds among group members." When the members of a group are in- terlinked with ties of social attractions, the group is co- hesive. Social integration of a group is a function of group cohesiveness, which in turn, depends on members' at- traction to each other. Several social scientists (Lott and Lott, 1965; Cartwright, 1968; Schachter, 1968) have studied group cohesiveness in terms of interpersonal at- traction. Landecker (1951) identified three types of social integration: 1) communicative; 2) normative; and 3) functional. Communication integration is the extent to which the units of a social system are interconnected through interpersonal communication links. Smend (1928) suggests that integration through persons can be either "direct" or "indirect." Direct integration through persons occurs when individuals interact face-to-face. Indirect integration occurs when the integrative relationship be- tween individuals is transmitted through one or more inter- mediaries. Normative integration is the extent to which the conduct of members of a social system conforms to system 28 says Rao (1966, p. 30). In modern systems, communication literally takes the form of the society itself, whereas in traditional villages it is limited to specific groups. Al— though there is less difference between the amount of inter— personal communication in modern and in traditional socie- ties, in modern systems it is faster, more complex, and more extensive (Rao, 1966, pp. 57-59; Schramm, 1963, p. 34). Thus, the present empirical study is designed to find out the degree of (l) heterophily on certain relevant attributes, and (2) communication integration in information-seeking and friendship communication; and (3) relationship between level of modernity of rural Indian villages, degree of heterophily among source-receiver dyads, and communication integration. SiXI Of t Ophi and - grat. Ophil both Vant (1951 CHAPTER II HYPOTHESES AND RATIONALE In line with the objectives outlined in Chapter I, six general hypotheses are presented in the present chapter. Of these six hypotheses, three hypotheses deal with heter- Ophily relationships of source-receiver dyads (Figure 2); and the other three hypotheses deal with communication inte- gration. Heterophily in Information-Seeking and Friendship Communication General Hypothesis 1: There is a greater degree of heterophily with respect to certain relevant attributes, among dyads engaged in information- seeking communication than in friendship commun- ication. In the previous chapter, while discussing heter- ophily, we concluded that in any communication situation both high heterophily as well as low heterophily on rele- vant attributes are necessary. Festinger (1950), Back (1951), Festinger and Thibaut (1951), and Schachter (1951) 29 H1; DH ere”. fJ Db «"0 nalv 71v ”9 High Low Fig. Information- Seeking Communication GH-l Friendship Communication 30 Modern Village GH-2 Traditional Village Modern Village GH-3 Traditional Village 2.--Hypothesized degree of heterophily on dimensions composed of selected variables in information-seeking and friendship communication in modern and traditional Indian villages. pre sen ind MCG Stu pos sys otIn b“ shi Sys dis sir the ien. In ; amp} I815 Act hYpo ; ~Ouhl 31 present evidence that communication behavior is initiated by'perceived discrepancies of opinion. Others have pre- sented evidence that communicative exchanges occur where individuals perceive their views to be similar (Altman and McGinnies, 1960). No attempt had been made in these earlier studies to attribute these conflicting findings to the "pur- pose" involved in the communication process. In a rural system, individuals are continually interacting with each other in several relationships. Through these relation- ships, individuals gain new information about others in the system and thus their perception of others being similar or dissimilar is dependent on the degree of differences or similarities on some characteristics. When an individual seeks information from another, the other person should be more knowledgeable, more exper- ienced, more qualified, and should also be safe and dynamic. In friendship communication the safety aspect is more emphasized. There has been only one study (Chou, 1966) which .relates directly to our General Hypothesis 1. Chou did ncnt find support1 for her general hypothesis of greater 1Although Chou (1966) did not find support for the hYpothesis for all three Colombian villages together, she found support for greater homophily in friendship 32 greater homophily (lower heterophily) in friendship commun- ication than in information-seeking communication, except for one variable: innovativeness. The reason that Chou did not find support for other variables such as age, so- cial status, literacy, mass media exposure, social partic- ipation, etc., might be that in information-seeking commun- ication in Colombian villages, innovativeness is the rele- vant attribute on which informed sources are evaluated. Selection of Relevant Variables for Heterophily Analysis in Indian Villages In selecting the variables in the present study, we attempted to select variables on which low heterophily is relevant in friendship communication and high heteroph- ily is relevant in information-seeking, in order to enable us to distinguish the characteristics which differentiate the intracting dyads' relationships. Traditionally in rural India, higher status had been accorded to the scholar priest and the rulers (or laJndlords) because they were the informed persons. conununication in the most traditional village with respect to Iliteracy, mass media exposure, cosompoliteness, social Par1:icipation, age, and social status. 33 Education was a "luxury" which only the rich could afford. Thus information and knowledge were a monopoly of the ruler-priest class. The social structure of rural socie- ties was highly stratified through the caste system. In seeking information, it was difficult for an individual of a lower caste to cross one, two, three, or more steps of caste barriers. Even in their housing arrangements, different caste groups had, and even have now, their houses located in distinctly different sections of villages (Rao, 1966, p. 10). Thus all communication activities (both in- formation-seeking and friendship communication) were limited to occur within specific groups. In information- seeking, the crucial attribute which was considered impor- tant in evaluating the qualified information-giver was age. Human activities in such traditional societies were repeti- tious and information passed from older, experienced information-givers to younger information-seekers. The introduction of education, mass media, and technological innovations brought new information and new knowledge about alternate human activities. The kind of information individuals in rural societies once sought was different than the kind of information sought now. Thus the shift in the criteria of influence from age and 34 traditional status, to knowledge and ability. An informa- tion source, to be qualified, should be more knowledgeable, and more competent. To be more knowledgeable, the informa- tion source should have more education, more mass media ex- posure, more outside contact, and have more access to other information sources; and to be more competent, he should be more innovative than the seekers. Historically, the higher caste members have enjoyed economic advantages, and it was they who could afford to educate their children and they were the ones who could read newspapers, buy and listen to radios and also understand the language used in radio pro- grams.1 Thus, differences in caste and level of living also become indirectly relevant in information-seeking com- munication.2 Although the information sources are more knowledgeable and competent than the seekers, they are safe as they are part of the same community and share the com- mon values and norms of the community. Interpersonal attraction is of more importance in a communication situation where the purpose is to have 1Although the language in a state is the same, the dialects differ from one part to another and the radio sta- tions' broadcasts are mostly in the language of city peOple. 2As the data for the present study come from a larger study, which focused on the diffusion of agricultural 35 informal affective association. Interpersonal attraction stems from attitude similarities. Individuals in rural societies who experience similar living conditions gener- ally share similar attitudes. Although we agree that in- dividuals connected with friendship bondage are not com- pletely alike, their differences in educational level, mass media exposure, outside contact, etc., are lesser than among the source—receiver dyads engaged in information- seeking communication. Thus a greater degree of heteroph- ily on the following selected variables is more relevant in information-seeking than in friendship communication. 1. Caste Caste is a category of individuals' ascribed posi- tions is a ritual hierarchy. An individual's caste rank is determined on the basis of his acceptance of drinking innovations, our selection of variables is restricted. Attitudinal variables (affecting credibility) were not measured in the larger study and could not be included in the present investigation. 1Evidence of heterophily on the selected variables in information-seeking communication also comes from past research on Opinion leadership: Carlson (1965); Patel (1964, 1966); Sen (1969); Troldahl and Van Dam (1965); Lionberger (1953, 1955); Rogers and van Es (1964); Rogers and Leuthold (1962); Rahim (1961, 1965); Emery and Oeser (1958); Rogers and Burdge (1962); Savale (1966); Kahlon and Kaushal (1967); and Radudkar (1960). in vi SE i: "1 36 water and cooked food from others. If a person of caste B can accept food and drinking water from a person of caste A, but not vice versa, then caste A is ritually higher than caste B. Sen (1969, p. 24) found caste to be one of the best predictors of opinion leadership in India, suggesting that information-seeking communication is directed towards per- sons of higher castes rather than lower castes. Bose (1967) found a marked degree of low heterophily (homophily) on the basis of caste in friendship communication in an Indian village. Individuals belonging to the same caste share simi- lar values and beliefs, etc., and are more likely to com- municate with others of the same caste. In information- seeking communication (when the issue on which information is sought is not directly related to ritual beliefs) indi- viduals seek out knowledgeable and competent others who are of different castes. 2. Level of Living Level of livipg_is the extent of an individual's possession of indicators of wealth. Caste provides ascribed status to individuals, and level of living provides 37 achieved status. As discussed earlier, since both caste and level of living are associated with other variables which determine who is more knowledgeable and competent, they become important in structuring interpersonal commun- ication in Indian villages. 3. Education In a changing society like India, formal education affects communication flow in social systems. Bose (1967) found a high degree of homophily on education in friendship communication in an Indian village. Sen (1969, p. 24) found a positive relationship between education and opinion leadership, which indirectly implies that there is heteroph- ily with respect to education between information-givers and information-seekers. 4. Mass Media Exposure Media exposure such as the movies, radio, news- papers, etc., is an important attribute which makes one more knowledgeable than others. In information-seeking communication, individuals seek information from those who have more exposure to mass media. In friendship 38 communication, if the individuals' knowledge level is too different (probably due to differences in the degree of exposure to mass media), they may not have common issues to discuss in their affective relationship. 5. Cosmopoliteness Cosmppoliteness is the degree to which an individ- ual is oriented outside of his immediate social system (Rogers with Svenning, 1969, p. 147). An individual's out- side orientation makes him more knowledgeable, and thus he becomes a source of information in traditional societies. Cosmopolite individuals link the village system with ex- ternal systems from which most new ideas come. The role of cosmopolites as linkers is of great importance in so- cieties where use of the mass media is limited. Merton (1957, pp. 387-420) found that cosmopolites belonged to more organizations than the localites, they were more willing to live elsewhere, and they made friend- ships with other cosmopolites. Menzel (1960) found that individuals who were in touch with the outside world re- ceived more information about innovations than those who were not. By virtue of their greater contact with external 39 systems, cosmopolites have more information and they are sought by the localites for information. In friendship communication, as suggested by Merton (1957, pp. 387-420), cosmopolites like to interact with other cosmopolites, and localites interact with localites. 6. Change Agent Contact Beyond media exposure and outside orientation, the other source from which individuals in rural India gain in- formation are change-agents from development agencies. Change agent contact is the degree of an individual's in- teraction with professional representatives of change agencies. Emery and Oeser (1958, p. 50) and Rogers and Burdge (1962) in their studies in Australia and the U.S., respec- tively, concluded that leaders from whom information was sought, were more likely to have a higher degree of change agent contact than others. Information-seeking communica- tion is expected to occur between an individual with a higher degree of change agent contact and another individ- ual with a lesser degree of change-agent contact. The assumption is that if individuals have more or less the same degree of change agent contact, there is less of an 40 information gap between the two, and hence less information- seeking. Rather, individuals with more or less the same degree of change-agent contact engage in friendship commun- ication. 7. Innovativeness Innovativeness is the degree to which an individual is relatively earlier in adopting new ideas than the other members of his social system (Rogers, 1962, p. 19). Studies in the U.S., the Netherlands, and Colombia (Lionberger, 1955 and 1959; van den Ban, 1963; Rogers and van Es, 1964) indicate that information is sought from in- dividuals who are more innovative than the seekers. Emery and Oeser (1958, p. 49) found the same result in Australian farm communities. Information-seeking communication in villages is concerned with innovations; a low innovative individual will seek information from another individual who is more innovative. Our argument of lesser heterophily with respect to innovativeness in friendship communication is similar to that discussed in the section dealing with change agent contact. If individuals are too different with respect 41 to innovativeness, an affective friendship relationship is less likely to occur. Empirical Hypothesis 1: There is a greater degree of heterOphily with respect to the dimensions of caste, education, level of living, radio listening, newspaper exposure, movie exposure, cosmopoliteness, Change agent contact, and innovativeness among dyads engaged in information-seeking than in friendship communication. Information-Seeking and Friendship Communication and Village Modernity General Hypothesis 2; There is a greater degree of heterophily among dyads engaged in information- seeking communication in more modern villages than in more traditional villages. General Hypothesis 3: There is a greater degree of heterophily among dyads engaged in friendship communication in more modern villages than in more traditional villages. Few researchers have inquired about what happens to interpersonal relationships between sources and re- ceivers in rural societies as they modernize. Yadav (1967) studied source-receiver relationships in both information- seeking and friendship communication situations in two Indian villages, one more modern than the other. He 42 hypothesized a greater degree of homophily in both communi- cation situations in traditional social systems than in modern systems. Yadav found support for his hypothesis only with respect to innovativeness and agricultural knowl- edgeability in information-seeking communication. Yadav (1967, pp. 183-188) presented some methodological and the- oretical arguments for his unsupported hypothesis such as that the study was limited to only two social systems; the number of information-seeking dyads was very few, whereas there were a large number of friendship dyads (individuals were allowed to make a choice of six friends as compared to three choices in information-seeking). As systems become more modern, individuals make more "objective" evaluations of other individuals from whom they seek information. In less modern villages, age and other social status are the criteria on which the information-sources are evaluated as to whether they are qualified or unqualified. In modern villages, individuals seek information from persons who are more innovative, more educated, have more media exposure, etc., and so are more knowledgeable and competent. In traditional systems, individuals' friendship interactions are limited within close-knit family and caste 43 groups; as the systems modernize, the boundary of one's peer group is broadened to include members of the larger community. The hierarchy of social relationships1 changes because of the new awareness. With education and media exposure the traditional norms of social relationships are no longer followed and individuals no longer feel uncom- fortable in having friendship communication with others who are somewhat dissimilar. Empirical Hypothesis 2: There is a greater degree of heterophily with respect to the dimensions of caste, education, level of living, radio listening, newgpaper exposure! movie exposure, cosompoliteness, change agent contact, and innovativeness among dyads engaged in information-seeking communication in more modern villages than in more traditional villages. Empirical Hypothesis 3: There is a greater degree of heterophily with respect to the dimensions of caste, education, level of living, radio listening, newspaper exposure, movie exposure, cosmopoliteness, change agent contact, and innovativeness among dyads engaged in friendship communication in more modern villages than in more traditional villages. 1In traditional Indian villages, when individuals of different castes, different socio-economic levels, etc. are invited to a marriage dinner, they are seated separr ately. This provides an example of how affective friend— ship between high and low (on any attribute) individuals is impossible in traditional villages. 44 Communication Integration and Village Modernity General Hypothesis 4: Communication integration is higher in information-seeking communication than in friendshipycommunication. Communication integration (the degree of intercon- nectedness among members of a social system) is a necessary condition to achieve normative integration in any society. Although normative integration is acheived through both information-seeking and friendship communicative integraf tion, in a society where transformation of normative inte- gration is in progress, information about new ideas and norms diffuses more through the information-seeking communi- cation structure. Once the members of a society accept the objectives of this transformation, individuals cross differ- ent traditional barriers of caste, status, education, etc., to obtain information. Thus, in changing societies like India, we expect individuals to be more inter-connected through direct or indirect interpersonal communication links in information-seeking communication situations than in friendship communication. it. I II-I.IY.II’.|..¥I rm! II 45 Empirical Hypothesis 4: Individuals are more interconnected through interpersonal communi- cation links in information-seeking than in friendship,communication. General Hypothesis 5: Communication integration in information-seeking communication is higher in more modern villages than in more traditional villages. General Hypothesis 6: Communication integration in friendship communication is higher in more modern villages than in more traditional villages. In a comparative study of two Indian villages, Rao (1966, pp. 57-59) found that in a more developed village, "contacts are spread over a wider area," whereas in a tra- ditional village, "Communication is limited to specific groups." Yadav (1967) found that in a modern village, individual members were more integrated in both information- seeking and friendship communication than in a traditional village. Guimaraes (1968, 1970b) reported a similar find- ing for friendship communication in Brazilian communities. Modernization is viewed essentially as a process of transfermation of systems integrated with traditional norms to systems integrated with modern norms, and as this trans- formation takes place, individuals' interaction patterns change; individuals attempt to identify more objectively the informed people and more people get, directly or 46 indirectly, interpersonally connected with them. Even in friendship relationships, individuals in modern systems, come out of the boundary of traditional peer groups limited within the same caste, or the same occupation, and interact with other members of the community. Thus in modern vil- lages, more people are integrated directly or indirectly in both information-seeking and friendship communication in modern villages than in traditional villages. Empirical Hypothesis 5: Individuals are more interconnected through interpersonal communica- tion links in information-seeking communication in more modern villages than in more traditional villages. Empirical Hypothesis 6: Individuals are more interconnected through interpersonal communica- tion links in friendship_communication in more modern villages than in more traditional villages. CHAPTER III RESEARCH METHODOLOGY In the present chapter, the following are presented: a) source of data and sample selection, b) a short descrip- tion of the villages under study, c) instrument construc- tion and data collection, d) Operationalization of the concepts used in the present study, and e) data-analyses and tests of hypotheses. Data and the Sample Data used in the present study come from the Indian part of a larger study, "Diffusion of Innovations in Rural Societies,” conducted in Brazil, India, and Nigeria, by the Department of Communication at Michigan State University. The Indian part of the research project was conducted in three phases. In the first phase, an attempt was made to .analyze the community setting in which rural people live, in order to determine to what extent the nature of the 47 48 community itself affects the adoption of innovations. The second phase of the study was designed to determine the factors which affect the innovative behavior of the indi- vidual farmer within a community setting. The third part of the Indian study involved field experiments to determine the effects of certain communication strategies, such as radio farm forums, adult literacy classes, etc. in inducing greater acceptance of innovations in a limited number of villages (Roy and others, 1968, pp. 1-2). The present study is based on analysis of the data from the second phase of the Indian research project. Data were obtained through personal interviews with farmers in rural settings. The three states, Andhra Pradesh, Maharash- tra, and West Bengal, were purposely selected to represent different modes of involvement of local self-government in development administration.1 Andhra Pradesh represents 1A state in India generally corresponds to a lin- guistic region. There were 17 states at the time the data twere collected. For administrative purposes each state is (Sivided into several districts. Administrative districts anre divided into development blocks, consisting of roughly (one hundred villages. The block development staff is ad- rninistered by a Block Development Officer who heads a team «of specialists in agriculture, co-operatives, animal hus- bandry and so on. The villages in a block are organized into circles of seven to ten villages, and at least one nuilti-purpose worker, called a Village Level WOrker, is assigned to each circle. 49 locally-elected people at the block level; Maharashtra represents locally-elected people at the district level, and West Bengal represents areas in which the emphasis on local self-government has only recently been instituted and popular control over development administration was still forthcoming from the state level. Within each state, three districts with differing intensities of development inputs1 such as (1) an Intensive Agricultural Development Program district, (2) a district in which at least one of the blocks was under the tribal development program, and (3) a district with more or less usual development inputs, were purposively selected. Beyond the purposive selection of the three states and nine districts, a three-stage ran- dom sample was used to select nine blocks, 18 Village Level worker circles, and 36 villages in each state. Thus a sam- ple of 108 villages were obtained in Phase I of the project}2 1Different strategies of development have been in- corporated in different districts of a state. At least one district in each state has been singled out for more inten- sive development efforts under a national programme (often referred to as IADP or the Package Program, the latter in view of the emphasis on supplying agricultural inputs in combination, or in a "package"). Further, areas having sub- stantial proportions of tribal people are often provided dif- ferent and more intensive development inputs under the na- tional tribal development program. 2For further details, see Fliegel and others (1968, pp. 4-7). 50 In the second phase of the study, eight villages were selected which were approximately typical of the range of variability on the success or failure of agricultural development programs in the 108 villages (Roy and others, 1968, p. 9). The sample of farmers in the eight selected villages were chosen purposively for the larger study and was thus restricted to farmers who operated a farm of at least 2.5 acres and who were not older than 50 years. Per- sonal interviews were conducted with a total sample of 680 farmers in the eight villages, three in Andhra Pradesh, two in Maharashtra, and three in West Bengal. Description of the Villages Studiedl Manchili (N=78), Kanchumarru (N=33), and Polamuru (N=99) are the three villages selected in West Godavari district of Andhra Pradesh. All three villages are well served by a network of irrigation and transport facilities. The three villages are all situated three to six miles from each other in the same administrative block. 1For further details, see Roy and others (1968) and Raju (1969). 51 The district in which the three villages are situ- ated is a "Package District" meaning that an intensive agri- cultural development program (IADP) was taken up in the Dis- trict (in 1961). The program was designed to ensure that needed resources-~seeds, fertilizer, irrigation, implements, and technical assistance, would be available so that agri— cultural develOpment could be intensified. The villages had populations of 1500, 2600, and 3400, respectively (Table 1). TABLE l.--Location and size of sample villages for the Phase II Study in India. . . . Rmmr- Nmflxx'of State District Village lation ndents Andhra Pradesh west Godavari Manchili 1500 78 Andhra Pradesh West Godavari Kanchumarru 2600 33 Andhra Pradesh west Godavari Polamuru 3400 99 Maharashtra Yeotmal Pophali 1149 100 Maharashtra Yeotmal Mulawa 3348 146 wast Bengal Birbhum. Amdole 2460 103 West Bengal Birbhum Harishpur 1709 59 West Bengal Birbhum Laxmidanga 1573 62 Scmmceu Roy and others (1969), and Raju (1969, p. 22). Pophali (N=100) and Mulawa (N=l46) are the two sample ‘villages in Maharashtra state. Both the villages are located tmithin 10 miles of a town and are connected by an all-weather 52 road and state-owned bus transportation system. The popula- tion of these villages in 1961 was 1149 an 3348, respec- tively. The three villages in West Bengal are Amdole (N=103), Harishpur (N=59), and Laxmidanga (N=62). The population of these three villages was 2460, 1709, and 1573, respectively. Only Laxmidanga is situated on an all-weather road. Amdole and Harishpur are only four miles away from each other, and both are difficult to reach especially in rainy season. Most people in Amdole are Hindus, whereas all in Harishpur and Laxmidanga are Muslims. Instrument Construction and Data Collection The interview schedule constructed to measure the desired variables was first translated from English to Te- lugu, the language of Andhra Pradesh state, and was pre- tested in that state. After making appropriate changes ‘the instrument was translated to the languages of the two crther states (Marathi in Maharashtra, and Bengali in West ‘Bengal), and a second pretesting was done in all three states. The second pretest responses were then tabulated 53 and screened to identify the meaning and use of expressions familiar to farmers in the three language areas. Several interviewer training sessions were held both before and after the pre-testing of the questionnaire. Field work was conducted during March and April, 1967. Interviewing was done by teams of four interviewers, led by a supervisor, in each of the three states. Because of language differences, the team members, who spoke the appropriate regional language as their mother-tongue, worked only within their home state. All teams had prior field in- terviewing experience and had participated in Phase I inter- views, training sessions, and Phase II pre-testing. All the interviewers had Masters' degrees in sociology, economics, social anthropology, or agricultural sciences, plus exper- ience in rural studies. The present author was one of the interviewers in West Bengal state. The interviewing teams established residence in a sample village, usually in a private residence. They made lists of eligible respondents by consulting voter registra- tion lists, village official's cultivators' lists, and knowledgeable people in the village. They then divided the list among themselves and proceeded to interview elig- ible respondents. As far as possible, each interview was 54 conducted in private and typically lasted about one hour and fifteen minutes. Since the interviewing teams had visited the villages during Phase I of the study, there generally was good rapport. Interview schedules were checked by the supervisor in the field, making it possible to return to the respondent if a question was omitted or misunderstood. After completing the field work, the team members coded all the interview data for computer processing. Code categories were established on the basis of a sub-sample and then the data were systematically converted to numerical codes. All coding was checked for random, as well as syste- matic, errors (Roy and others, 1968, pp. 11-12). Qperationalization of Concepts Based on the hypotheses listed in the previous chap- ter, the concepts of primary interest in the present study are: l. Information-seeking communication 2. Friendship communication 3. Caste 4. Level of living 55 5. Education 6. Radio listening 7. Movie exposure 8. Newspaper exposure 9. Cosmopoliteness 10. Change agent contact 11. Innovativeness 12. Heterophily on dimensions emerging out of nine previously-listed variables 13. Village modernity 14. Communication integration. 1. Information-Seeking Communication Information-seeking communication occurs when an in— dividual interacts with another to obtain information for making certain decisions. Information-seeking communication was measured by sociometric questions: If you needed advice on problems regarding the following matters from which one person in this village would you seek advice first? (a) Technical problems associated with farming (b) Obtaining credit (c) How to get maximum returns for your products. Only one choice was allowed for each issue. 56 2. Friendship Communication Friendship communication occurs when an individual interacts with another to have an informal and intimate affective association. Friendship communication was measured by a sociometric question: Who are the people with whom you visit most fre- quently? Let us limit ourselves to three. The best possible sociometric question to measure friendship communication would be: Who are your most inti- mate friends with whom you communicate most frequently? In Indian Village situation, such type of direct questions are avoided as the respondents may become suspicious about the intentions of the researcher. Although the sociometric ques- tion asked to the respondents refers to visit, in translating it to Indian languages, attention was paid to the conceptual equivalence for intimate friendship rather than simply visit- ing. In Bengali language, the question asked was "Who are the people with whom you stand and sit I2E§E.EE§EE) most?" Thus, in all three states the questions asked were to find out the three most intimate friends of the respondents with whom they interact most. 57 3. .Caste EEEEE is a category of an individual's ascribed posi- tion in a ritual hierarchy. Each respondent was asked to identify his caste. To obtain rank positions of respondents' castes in the hierarchy, a series of photographs showing some individuals (not from the sample villages) with cues to caste in terms of occupation, dress style, etc., were presented to a sample of key informants (one from each caste), who arranged the photographs in descending order according to who could accept cooked food and/or drinking water from whom. Caste rankings given by the majority of the key informants in a village were the final ranks of the castes for a village.1 4. Level of Living Level of living is the extent of an individual's material possessions or indicators of wealth. Operationa ally, data on level of living were obtained on such material possessions as clothes, shoes, jewelry, wrist watch or clock, 1Two of the sample villages are populated with Mus- lims only; as such, no caste ranking could be done. Thus, in our hypotheses dealing with caste, we use information about 559 respondents, i.e., excluding the sample in the two villages. 58 flashlight, wooden/metal furniture, mosquito nets, bicycle/ motorcycle/scooter, brick or stone house, windows with shutters, cement or Stone floor, tiled/tin/asbestos/cement roof, separate sitting room, well/tube well, separate bath- room/latrine, and two-storied house. Individuals received one point for owning each item. 5. Education Education was measured by the number of years of formal schooling completed by the respondent. 6. Radio Listening Radio listening was operationalized as to whether an individual listened to radio or not. The respondents were asked: Do you listen to radio? 7. Movie Exposure Movie exposure is the frequency with which an indi- vidual attends movies. The following question was asked: Did you see any cinema films1 during 1966? How many? 1Reference is to commercial films, not those shown by the Block Development staff. 59 8. Newspaper Exposure Newspaper exposure is the degree to which an indi- vidual reads newspapers or has them read to him. It was measured by asking: Did you read (did anyone read to you) any news- paper in the past week? How many? 9. Cosmopoliteness Cosmopoliteness is the degree to which an individual is oriented outside of his immediate social system. Cosmop- oliteness was measured by the number of visits to nearest cities and towns. Respondents were asked: How many times have you visited the following places last year? (a) Nearest town (less than 100,000 population) (b) Nearest city (population 100,000 or more) 10. Change Agent Contact Change agent contact is the degree of an individual's interaction with professional representatives of change agencies. It was operationalized by noting the frequency *with.which a respondent talked with extension agents, saw .block:films, and attended agricultural demonstrations, as 60 each of these activities entailed direct contact with pro- fessional change agents. An index was constructed by sum- ming the responses to the following questions: Last year (1966) how many times did you: (a) Talk with the Block Development Officer? (b) Talk with the Village Level Worker? (c) See a block film on agriculture? (d) See an agricultural demonstration? ll. Innovativeness Innovativeness is the degree to which an individual is relatively earlier in adOpting new ideas than the other members of his social system. This variable was measured as the number of 10 selected agricultural innovations ever used by an individual. Individuals received a score of one for each innovation ever used. The ten innovations are: l. Ammonium sulphate fertilizer . 2. Superphosphate fertilizer 3. Fertilizer mixtures ' 1L 1... Fa- urn. .— 4. Insecticides for plant protection 5. Green manure 6. Cultivator or weeder 7. Improved breeding of cattle 8. Animal inoculation for disease prevention 61 9. Rat poison 10. High yielding varieties: Taichung Native 1 or IR 8 rice or any improved variety of corn, jowar, or bajra. 12. Heterophily on Different Dimensions Heterophily is the degree to which pairs of inter- acting individuals are dissimilar on certain attributes. In previous studies (Chou, 1966; Yadav, 1967) deal- ing with homophily-heterophily, the concepts have been operationalized as the zero-order product-moment correla- tion between the scores on certain attributes of interact- ing dyads. When product-moment correlation is used as a measure of homophily-heterophily, it indicates the degree of variation of an attribute of individuals who interact, associated with changes in the attribute of individuals with whom they interact dyadically, and vice versa. Our definition of heterophily emphasizes the degree of dissimi- larities on attributes of interacting dyads. In the present study, heterophily is measured as the absolute difference in the selected dimensions (caste, level of living, education, radio listening, movie exposure, newspaper exposure, cosmo- politeness, change agent contact, and innovativeness) 62 between individuals engaged in information-seeking and friendship communication. When the absolute difference in a dimension between interacting individuals is 0, there is least heterophily (perfect homophily); if the difference is not 0, then there is some degree of heterophily. A comparison between absolute difference scores and zero-order product—moment correlation, as measures of homophily-heterOphily can best by illustrated by using some hypothetical dyads engaged in dyadic interaction. For ex- ample, X and X s are nine interacting dyads in Village A, ls 2 X38 and X48 in village B, and XSS and X68 in village C, and the number of newspapers read by them in a year are as pre- sented in Table 2. The zero-order product-moment correlation between attributes of the hypothetical dyads is 1.0 in all three villages. According to Chou's (1966) and Yadav's (1967) Operationalization of homophily-heterophily, a correlation of 1.0 indicates perfect homophily. With zero-order product-moment correlation as a measure of homophily- heterophily, one would conclude that in all the three vil- lages (A, B, and C), perfect homophily or least heterophily exists between interacting dyads. By examining the scores of the interacting pairs of dyads in the three villages, we 63 III! In. DI .. n . .IE .Im 00.0 m.0a mm.0 mm.s 00.0 mN.0N .mN.0 mucwummmso musaomnm mommgfi 0.0: 0m.0u NN.0I 0.0 ‘0.N 0.N 0.N u u a NH 0 0N 0 0N 0 0N 0N ma Nm MN 0N ma N NH N NN m NN NH NN NH NH 00 NH NN NH 0 0 m 0H NH 0N N 0H 0 0 N0 0 0H 0 0 N 0 0H N 0N ma «a N N 0N N 0H N 0 0 A NH 0 NH N NH 0 0 0N 0 NH 0 N 0 0 0 ma 0 0 0 0 0 0H 0 0 0 0 m 0 0 0 0 0 0 m 0 NH 0 0 0 NH N ma 0 N 0 m 0 N N 0 N 0 N ma N NH 0 0 0 0 0 H N 0 N N H «ax max Nax aux 0Nx 0x 0x Nx 0x mx 0x mx Nx ax mumasosuuma oioflfl> Tammie, mi0£d> oimflfl> obmflfl> misfit, aimflgw .mmmsHHw> Hmowuonuomhe co>om cw mommo HMOfluonuomms mEom mCOEm haflnmououoe mo mmusmmoe mm moocoquMHm ousHOmnm ommuo>m can coaumHouuoo ucoeoeluosooum HooHOIouoNII.~ mqmde 64 find that although the scores on an attribute of interacting dyads are very highly related (r=l.0 in each village), there actually is not perfect homophily except in village C. By examining the average absolute differences between the in- teracting dyads, we find that there is perfect homophily only in village C, and there is a higher degree of heteroph- ily in village B than in village A. Thus, the average absolute difference score on an attribute of interacting dyads is more appropriate measure of heterophily than the zero-order product-moment correla- tion (for comparison of the two measures when r varies from 1.0 to -l.O, see Table 2). The degree of heterOphily on different dimensions emerging out of the nine selected variables among the inter- acting dyads was measured in the following way: Step 1: Each respondent's scores on the selected vari- ables were standardized.1 lBefore measuring absolute differences between the two individuals in each dyad, individuals' scores on each of the selected variables were standardized so that varia- tions in absolute differences among dyads were not influ- enced by village differences with respect to the selected variables. The formula used to standardize each respon- dent's score on each variable is: m -m‘l’.‘ 7“? " 65 Step 2: Absolute difference between the two individuals in each dyad on each of the selected variables was calcu- lated. Step 3: These absolute differences or heterophily scores1 on nine variables were then factor-analyzed to identify different heterophily dimensions present among dyads engaged in information-seeking and friendship commun- ication. The principal axis solution using varimax rotation with a Kiel-Wrigley criterion for terminating factor rota- tions, yielded two factor-rotated solutions. The percent of total variance explained by each of the rotated solutions X - 1 X Z: s shere, X1 = each respondent's score on a selected variable R = village mean score on the selected var- iable for the village the respondent was from s = standard deviation of the distribution of scores on the selected variable among the respondents in the village the respondent was from. 1Actually, the absolute differences are heterophily scores. When heterophily score or absolute difference is 0, there exists least heterophily; that is, there exists perfect homophily. Thus, in the later sections of the present chapter, and in the later chapters, absolute dif- ference scores will be referred to as "heterophily scores." 66 is presented in Table 3. Among the two rotated factor solutions, the second one containing three factors was selected. In selecting the three-factor solution, the major considerations were: 1) The conceptual meaningfulness of the factors as evidenced by the variables loaded on each factor; 2) The proportion of variance explained by the fac- tors. Of the total variance present among the nine variables, 54.5 percent is explained by the three factors jointly (Table 3). The dimensions emerging from the three factors are: 1) Status heterophily, 2) Change contact heterophily, and 3) Movie exposure heterophily. In deciding on the variables which constitute the factors the following considerations were set: 1) The square of the factor loading of a variable should approach the communality (hz) of that variable. 67 TABLE 3.--Percent of total variance explained by the two rotated solutions meeting a Kiel-Wrigley cri- terion for terminating the factor rotation. Rotated Solution Percent of Total Variance Explained First Rotation Factor I 27.7% Factor II 16.8% Total 44.5% Second Rotation Factor I 23.8% Factor II 19.5% Factor III 11.2% Total 54.5% .2) A variable is included in the factor on which the loading is higher than .50, if the loadings are less than .50 on other factors. Accordingly, the variables were included in a fac- tor if they met at least one of the above two criteria.1 1The only exception was radio listening, which is discussed later in the section dealing with change contact heterophily. 68 Factor I: Status Heterophiiy The dimension tapped by Factor I is status heteroph- ily among interacting dyads. It contains such items as caste, education, innovativeness, and level of living. Het- erophily scores on all four variables are positively loaded on the status heterophily dimension (Table 4). TABLE 4.--Heterophily scores on nine selected variables and factor loadings in the three-factor solution1 Factor Loadings Variables Factor Factor Factor hz I II III Caste .5163 -.2709 .1876 .3751 Education .6227 -.0733 -.1626 .4196 Cosmopoliteness .1820 -.6445 -.1668 .4764 Newspaper exposure .4213 -.6367 -.0632 .5869 Radio listening .4207 .6847 -.0376 .6472 Movie exposure -.0543 -.1298 -.9506 .9235 Innovativeness .6202 -.l9ll .0409 .4228 Level of living .7518 -.0204 .0769 .5715 Change agent contact .3895 -.5748 -.0200 .4825 1 o 0 Highest loadings of the variables on a factor are underlined. For zero-order correlation between the nine variables, see Appendix. 69 Factor II: Change-Contact Heterophily The second dimension of heterophily is change con- tact. It consists of cosmopoliteness, newspaper exposure, and change agent contact. All three items are loaded to the factor in the same direction, and represent exposure to different channels through which new ideas reach vil- lages. Although radio listening loaded on the factor, it is in the opposite direction to that of the other items. Reasons for this seeming anomaly might be: 1) the level of measurement for radio listening was nominal, and 2) about 77 percent of all the respondents responded that they listened to radio, while 23 percent did not. Thus, radio listening was not included in the change contact heterophily dimension. Factor III: Movie Heterophily The third dimension, Factor III is composed mostly of movie exposure. Heterophily in movie viewing is the only item which loads very high (-.95) on Factor III. Movies are the major media of entertainment for Indian 70 villagers.1 Thus, the heterophily dimension tapped by this factor is mostly entertainment. Step 4: Status heterophily and change-contact heter- ophily were calculated for each dyad by dividing the sum of the heterophily scores on all variables constituting each dimension of heterophily, by the number of variables? Thus these scores are averages-across-the-variables included in each factor. 13. Level of Modernity Level Of modernity Of a village is its relative con- dition or state in the modernization process as compared to others at a particular point in time. In an effort to rank the villages on modernity, the following thirty variables 1Radio is also an everyday medium of entertainment, but movies have a greater amusement appeal tO Indian vil- lagers. 2Most movie theaters are located in towns and cities. Thus, visits to movie theaters bring outside con- tacts. Screening of a government documentary (commonly .known as ”news reels") is compulsory before the feature film is screened. This provides an Opportunity for the :Indian villagers to be exposed to messages containing new ideas also. 3This step was necessary, particularly for status heterophily, as no caste ranking was done in two Of the sample villages where all the residents are Muslims. 71 (considered to be aspects of modernity) were purposively selected and were treated to Guttman scale analyses. \OCDQO\UIJ>U)NI—' O 10. ll. 12. 13. 14. 15. 16. 17. 18. 19. 20. 21. 22. 23. 24. 25. 26. 27. 28. 29. 30. Family type (nuclear or joint) Education of respondent Education of children Farm specialization Tenure status Number of Offices held Social participation Farm commercialization Farm size Level Of living Farm labor efficiency Trial of agricultural innovations Cosmopoliteness Radio listening Movie exposure Literacy Newspaper exposure Change agency contact Credit orientation Planning orientation Non-authoritarianism Educational aspirations for children Self-reliance Deferred gratification Economic ambition Achievement motive Political knowledge Secular orientation Empathy Interpersonal trust Of these 30 variables treated to a Guttman scale analysis,1 15 variables were retained with a coefficient 1A computer program for Guttman scale analysis, developed by Dr. David J. Stanfield Of the Department Of Communication at Michigan State University, was used in the present study. The ranking of the villages on modernity was used in an earlier study by Sen (1969). The present author assisted Dr. Sen in analyzing data for the earlier study. 72 Of reproducabilityl of 0.99. The following variables re- mained as scale items in the following order: 1. Number Of Offices held; newspaper exposure; economic ambition, interpersonal trust 2. Political knowledge; farm commercialization 3. Literacy; level Of living; farm labor efficiency 4. Social participation 5. Radio listening; non-authoritarianism 6. Education; cosmopoliteness 7. Family type. There were in all seven scale positions and except in two cases more than one variable was placed in the same position. For the scale positions, where more than one variable was placed, one variable was chosen randomly for each position. After random selection, the variables in the scale Of modernity were: Number of Offices held, po- litical knowledge, farm labor efficiency, social partici- pation, non-authoritarianism, education, and family type. The villages were then ranked according to the mean village 1Coefficient of reproducibility = l - EEK-where, E = number Of errors, N = number of villages, n = number Of scale items (Schuessler, 1971). 73 scores on the variables in the scale. For example, Kanchu- marru village got the highest rank because it showed high- est scores on most Of the scale items, whereas Laxmidanga village was ranked lowest as it showed lowest scores on most of the scale items (for ranks Of all the eight vil- lages, see Table 5). TABLE 5.--Rank-order of eight villages according to level of modernity. Rank (From Most to Least Modern) Village State 1 Kanchumarru Andhra Pradesh 2 Polamuru Andhra Pradesh 3 Mulawa Maharashtra 4 Pophali Maharashtra 5 Manchili Andhra Pradesh 6 Amdole West Bengal 7 Harishpur West Bengal 8 Laxmidanga West Bengal 74 14. Communication Integration Communication integration is the extent to which the units of a social system are interconnected through interpersonal communication links. For measuring communi- cation integration, a computer program called "Network Routine" developed by Guimaraes (1970a) was used. The first step in this analysis is to form a binary (O-l) matrix. A "0" in the matrix indicates no direct communi- cation link between two individuals, and a "1" indicates the presence Of direct communication link. Then matrix multiplication is done by raising the original binary (0-1) matrix to n-powers in order to determine n-links and n-step connections (Guimaraes, 1970a, p. 33).1 The computer pro- gram provided the following information: 1) Sum Of all the communication links2 a respondent had; 2) a relative com- munication integration index for each respondent, which is 1For details about the Network Routine, see Guimaraes (1970a). 2Communication links represent the number of steps (individuals) through which an individual is communicat- ively connected with another both directly and indirectly. Suppose A directly communicates with B, and B directly communicates with C; then A has two communication links with C. 75 the sum Of the communication links divided by N-l (where N = total number of respondents in a village); 3) a vil- lage communication integration index, which is the sum of each respondent's (in a village) communication integration indices divided by N. Data Analyses and Tests of Hypotheses Before measuring homophily-heterophily in information-seeking and friendship communication, the dyads engaged in information-seeking and in friendship communication were identified. We found 976 and 552 dyads engaged in friendship communication and information-seeking respectively. Of the 976 dyads engaged in friendship com- munication, 6 percent (n=62) make mutual choice (Table 6). In the case Of information-seeking communication, only 1 percent (n=7) made a mutual choice. When individual A re- sponds that he communicates with A and B responds that he communicates with A (in the same communication net), we have two identical dyads (A-—»B and B-—»A. As our major interest is to study the degree of similarities and differ- ences between A and B on selected attributes, in the case TABLE 6.-Number of total, mutual, and common dyads in information-seeking and friendship communication. uorqeorunmmog burxees-uorqemlogux pue drqspusrzg qqog ur pafiebua speKq go Jeqmnn uorqeorunmmoo burxaes-uorqemiogul ur speKa aoroqa Tenqnw go Jaqmnn spexq Burxses —uoraem10;u1 Tenor uoraeorunmmoa drqspuerig ur speKa eoroqo Ienanw go qumnu spefia dIuSpuaIIJ tenor Village 76 19 66 93 Manchili. 14 33 59 122 126 Kanchumarru 23 23 39 31 20 19 62 Polamuru 66 132 104 Pophali 16 236 Mulawa 13 155 Amdole 90 40 Harishpur 49 95 Laxmidanga 188 62 552 ,976 Total 77 Of mutual choice dyads (A-—+B, B-—»A), we considered them as one dyad in the analyses, rather than two.1 Thus, in our final analyses for the hypotheses dealing with homophily-heterophily, we had 914 dyads engaged in friend- ship communication and 545 paired dyads engaged in information-seeking communication. General Hypothesis I Of the 914 dyads engaged in friendship communica- tion, and the 545 dyads engaged in information-seeking communication, 188 dyads were common to both. Thus, we had 726 dyads (79 percent of friendship dyads) engaged exclusively in friendship communication, 357 dyads (65 percent Of information-seeking dyads) engaged exclusively in information-seeking communication, and 188 dyads (21 percent Of the friendship dyads; 35 percent Of the information-seeking dyads) engaged in both. As General Hypothesis 1 is specifically a comparison of the degree Of heterophily among dyads engaged in information-seeking with the dyads engaged in friendship communication, we compared the dyads engaged only in information-seeking, 1If in a village all dyads are mutual choice dyads and say, their number is 100, then in actuality there are 50 specific dyads. 78 with dyads engaged only in friendship communication, and discarded those engaged in both types of communication. For comparing differences in heterophily in information-seeking and in friendship communication, aver- age heterophily scores on (1) status heterOphily, (2) change contact heterophily, and (3) movie exposure heter- Ophily, were calculated for 357 dyads engaged in information-seeking communication, and for 726 dyads en- gaged in friendship communication. Thus, the following three Empirical Hypotheses were tested, rather than the one stated in Chapter II: Empirical Hypothesis l-A: There is a greater degree of status heterophily among dyads en- gaged in information-seeking than in friend- ship communication. Empirical Hypothesis 1-B: There is a greater degree Of change contact heterophily among dyads engaged in information-seeking than in friendship communication. Empirical Hypothesis 1-0: There is a greater degree Of movie exposure heterophily among dyads engaged in information-seeking than in friendship communication. 79 TO test Empirical Hypotheses l-A, 1-B, and l-C, the "t" test for unequal N's was used1 (Table 7). General Hypothesis 2 and General Hypothesis 3 General Hypothesis 2 and General Hypothesis 3 deal with comparisons of the degree Of heterophily among information-seeking dyads and friendship dyads across eight villages with different levels of modernity. As mentioned earlier, three heterOphily dimensions emerged from the selected variables. Thus, the Empirical Hypo- theses tested for General Hypothesis 2 and General Hy- pothesis 3 are: t = -—————— where, M1 = average heterophily 2 among information- seeking dyads, M = average heterophily among friendship dyads, n = number Of information- seeking dyads, n = number of friendship 2 2 dyads, (x-Ml) + (x-MZ) and s = ' = pooled variance of heterOphily among both information-seeking l 2 and friendship dyads. 80 .Oumu Eoum OOGOHOMMHO now DGONO summon cowumaouuoo HOOHOIRGMH cmeummmm .muwnuoooe no xcmu mmmaaw> can “on: Imomxo OH>OE “my one .uomucoo mocmco “NV .msumum AHV so mmuoom hawcmououoc ommaaw> m u z mammNHH> .momma 1Hfl> Honofluflomuu OHOE cw can» mommaaw> GHOOOE OHOE cw GONHMOAGSEEOO unflxomm IsofiumEHOmnfi cw oommmco woman UGOEM maflnmououoc mo condom Houmoum m mH muons ”m mwmmeeommm Nosmrmu .mwamfimm unopcmmmccfl Ozu mo mcooe Ozu coo3uoe oocmuommap Mom unduly .GOAHMONGSEEOO no woman 03D MOM “one Imomxo OH>OE Am one .uomucoo monoco AN .msuopm AH amcowmcos IHO money no mouoom maficmououoc omsnmbm mmh u 2 5mm u z momma .noflumowcnfifioo mflcm locowum ca coca :OwuoOflcsE IEOO mnwxoomICOflumEuOmcfl ca command woman mcoam .mmusn Ifluuum ucm>oaou cwmuumo op pommmou cues MHHcmououoc mo ooumoc.uoumonm m we muons «N mmmmeuommm Noamrme monocuomhm assumes no cocuoz Hmowumflumum mOHQMfiHMNV MO mOHDWMO—L m.z can mammamsd mo muHcD monocuommm .mcflumoulmommcuommn can mOmMHmcoloumanl.h mqm can “mouoom nofiumuomucw coauMOHcsEEOo ommaaw> .wuwcuoooe co xsou ommaaw> can «mouoom coflumummucw coauOONGDEEOO ommaaw> m u z mammaas> m n z mammHNN> .mmmoaafl> Hmcowuwo Imus OHOE cw can» mommaafi> GHOOOE OHOE ca Hocmwn ma cowuooac55800 mwcmocowuw as coavmumoucs :OHMMONGnEEOU “e mwmmeuommm Ncnmrmo .momoaaw> Hocowuwomuu OHOE cw cmcu mommaaw> nuance once as Honda: mw COwuoOflssEEOO mafixoomlcOfluoEHOmcfl ca cowumummucfi cowumowsnfieoo "w mmmmeeommm Nssmrme .OHmEmm mean «0 mecca 03» cmmzuwn oocmuomwflo MOM amoulu .cOaumoaan IEOO mo mommy 03¢ How “mouoom coflumumounfl coaumowcsfifioo mmmum>< ommuz maono IMOGH :OwuoOflcseaoo macmocowum cw coca GONUOONGDEEOO mcwxomm IcoflumEHomcw GH umcmwc ma :Owumummucw cowumowsnEEOu “v mmeoeuommm Ncsmrme .OHON Eouw mucoummwwo wow ucmwo iwumooo cowumamuuoo noououxcou csfiuommm .huacumpoe no xsmu omoaaw> Ono “whence Ixo ow>OE Amy was .uomusoo omcmco Ame .mnumum Adv no monoom hawcmououoc mooaaw> m n z mmmmNHN> .mmmmaaa> choauue soup duos cw coca mommHHfl> cumooe OHOE cw coauOOchE IEOO mwcmpsowuw cw comomco mommo mcoeo hawcmououmc no common Houmoum m wwwoumce um mummeuommm Nonmeme 82 Empirical Hypothesis Z-A: There is a greater degree Of status heterophily among dyads engaged in information-seeking communication in more modern villages than in more traditional villages. Empirical Hypothesis 2-8: There is apgreater degree of chagge contact hetegpphily among dyads engaged in information-seeking communication in more modern villages than in more traditional villages. Empirical Hypothesis 2-0: There is a_greater degree of movie exposure heterophily among dyads engaged in information-seeking communication in more modern villages than in more traditional villages. Empirical Hypothesis B-A: There is a greater degree Of status heterophily among dyads engaged in friendship communication in more modern vil- lages than in more traditional villages. Empirical Hypothesis 3-8: There is a greater degree Of change contact heterophily among dyads engaged in friendship communication in more modern villages than in more traditional villages. Empirical Hypothesis 3-0: There is a greater degree Ofymovie exposure heterophily amopg dyads engaged in friendship communication in more modern viliages than in more traditional villages. Average heterOphily for each village, in each type Of communication was calculated for status heterophily, change contact heterophily and movie exposure heterophily. The eight villages were ranked according to their 83 heterophily scores on each Of the three heterophily dimen- sions (from highest heterOphily to lowest heterophily). For testing Empirical Hypotheses 2-A, 2-B, and 2-C, Spear- man rank-order correlation coefficient (rs)1 between vil- lages' ranks on heterophily (among dyads engaged in information—seeking communication) and their ranks on modernity was computed. To test Empirical Hypotheses 3-A, 3-B, and 3-C, Spearman rank-order correlation coefficient between the villages' ranks on heterophily (among dyads engaged in friendship communication) and their ranks on modernity was computed to test statistically the signifi- cance Of difference of the coefficient from zero. General Hypotheses 4, 5, and 6 Three General Hypotheses (4, 5, and 6) deal with communication integration as the dependent variable. The unit Of analysis, for Empirical Hypothesis 4 is individuals l 2 . r8 = l -» 6§di where, di = difference between N -N the rank Of a vil- lage on modernity and on village heterophily scores N = number of villages 84 (N=680), whereas the village is the unit of analysis for Empirical Hypotheses 5 and 6. Empirical Hypothesis 4 deals with the comparison of communication integration between information-seeking and friendship communication. Mean communication integra- tion socres for 680 respondents were computed for both information-seeking and friendship communication. TO test Empirical Hypothesis 4, a t-test1 for difference between two means Of a dependent sample was used. TO test the Empirical Hypotheses 5 and 6, the eight villages were ranked (from highest to lowest) according to (1) their communication integration scores (in information- seeking and in friendship communication respectively, and (2) village modernity (from most modern to least modern), and the Spearman rank-order correlation coefficient rs was computed tx> statistically test the significance Of dif- ference Of the coefficient from zero. 1 - t = -—-§2-——- where, Xd= mean difference between Sd/ /N:I information-seeking and friendship communication integration S = standard deviation of dif- ferences between informa- tion-seeking and friend- ship communication inte- gration N = total number of respondents CHAPTER IV FINDINGS In the present chapter, findings regarding the six general hypotheses are presented. Heterpphily in Information-Seeking and Friendship Communication General Hypothesis 1: There is a greater degree of heterpphily with respect to certain relevant attributesy among dyads engaged in information- seeking communication than in friendship commun- ication. TO compare the degree of heterophily among dyads engaged in information-seeking communication with the de- gree Of heterophily among dyads engaged in friendship com- munication, average heterophily scores for eaCh of the three heterophily dimensions were computed for all dyads involved (1) only in information-seeking, and (2) only in friendship communication (although, no hypotheses were stated for dyads engaged both in information-seeking and 85 86 in friendship communication, average heterophily scores among such dyads were also computed). In Table 8, the average heterophily scores, the standard deviations of heterophily scores, and the number Of dyads involved, TABLE 8.--Average heterophily scores and standard devia- tions of heterophily scores, and the number Of dyads involved in different types Of communica- tion situations. Type Of Heterophily Dimensions Communication Statistics Situation Change Movie Status Contact Exposure Information- Mean 1.12 1.21 0.67 Seeking Only Standard Deviation 0.67 0.83 1.06 N 357 357 357 Friendship Mean 0.81 0.92 0.54 Communication Only Standard Deviation 0.54 0.79 0.93 N 726 726 726 Both Information Mean 0.88 1.06 0.61 Seeking and Friendship Communication Standard Deviation 0.56 0.80 0.96 N 188 188 188 87 are presented (see also Figure 3). The computed t-values for differences between the mean heterophily scores are presented in Table 9. From Table 8 and Figure 3, it is evident that the differences in the degree Of heterophily asnong information-seeking dyads and among friendship dyads are in the hypothesized direction for all three heterOphily dimensions. The "t" values comparing the mean heterophily scores among information-seeking dyads versus among friendship dyads, with respect to status (t = 8.33), change contact (t = 5.66), and movie exposure (t = 2.02), are all significant at the .05 level. In each of the three heterOphily dimensions, the mean heterophily scores were greater for dyads engaged in information-seeking than for dyads engaged in friendship communication. Thus, the general hypothesis that there is a greater degpee Of heter- pphily with respect to certain relevant attributes among dyads engaged in information-seeking communication than in friendship communication, is supported.1 1The decision rule throughout the present disser- tation is that majority of the empirical hypotheses have to be supported for support Of the general hypotheses. II 4) *3 G) (D III ‘1! FH'D":!"!’ (3 ‘1 (D cfl'40 :33 100 - 0.8 -‘ 0.5 4 00" ~ 002 -( Friendship Communication Fig. Status Heterophily only .30 0" ...O a. a-° -° 0.9. . . 0:. .:'..o. O. .0. .‘O o . . - a... a o... H 0.0 0...... .0 .oo’.o '. O . ... o 0.. O . . ‘. ‘ - .;n~ 3" 'o:oo.o °. . 0.. U :0 o... o 00" O. o O .0 90:. So a I '.".-'. ...... 0.0.... ‘00 ‘ 0......0 .2.. 0’0. ': o . . O. O °'..o a: .10 0.: a ' ‘0' .0... O. .. O . , o o.‘ a o ' ‘0 a... 1.1 :0 z. .0. ". a.-.5 u‘ '0 o .0 0.. 0'. O. .. ..o. 0' 0. 0.... e .' O. .... .. 0...... 0.. .0 9-43 :0:0.0... 0". 0.. :Io'.o '.f0: 0 o . '00. 30.... ..’:00. f. ,' 0.... ’0 a. l;g' J U 88 Movie Erposure HeterOphily Information- Seeklng only Change Contact Heterophily IIIIIII Both Informa- tion-Seeking and Friendship Communication 3.--Mean heterophily on status, change contact, and movie exposure for dyads engaged in dif- ferent communication situations. 89 TABLE 9.--Computed "t" values for differences between mean heterophily scores on each heterophily dimension for dyads involved in different communication situations. t4values for Differences in Mean General : ison Heterophily Scores Ifipoflmsis oflbeds Tbsuxl seams Chasm: ltwie Canmm. memmre General Information- Hypothesis 1 seeking vs. iiiemkflfip cxxnmmnication 8.33* 5.66* 2.02* IfipOUmsisves Influmathr not formulated seeking vs. cannon dyads 4.25* 2.09* 0.63 InpOUEeflswas Ommoncheds not formilated vs. friendship dyads 1.58 2.15* 0.89 *Significant at the .05 level, one-tailed test. The dyads engaged in both information-seeking and friendship communication were not included in the analyses for testing General Hypothesis 1. Although no hypotheses were stated earlier comparing the degree of heterophily among the common dyads and those engaged only in information- seeking or only in friendship communication, our expectations were that (l) the degree Of heterophily among dyads engaged only in information-seeking communication would be higher 90 than among common dyads, and (2) the degree of heterophily among common dyads would be higher than among dyads engaged only in friendship communication. The reasons for these ex- pectations are similar to those discussed in Chapter II. It was also realized that by comparing the degree Of heteroph- ily across information-seeking dyads, common dyads, and friendship dyads, one might be able to identify the specific dimensions or attributes on which (1) less heterophily (or more homophily) is necessary for friendship communication to occur; and (2) high heterophily is necessary for information-seeking communication to occur. Table 8 and Figure 3 show that the mean heterophily in status, change contact and movie exposure among dyads en— gaged only in information-seeking are greater than among the common dyads. The computed "t" values for differences in mean status heterophily, change contact heterophily and movie exposure heterophily are 4.25, 2.09, and 0.63, respec— tively (Table 9). Except for the differences in movie ex- posure heterOphily, the computed "t" values for differences in status heterophily and change contact heterophily between information-seeking dyads and common dyads are significant at the .05 level (Table 9) suggesting that there is a greater degree of heterOphily with respect to status and change 91 contact among information-seeking dyads than among dyads en- gaged both in information-seeking and friendship communica- tion. Average status heterophily, change contact heteroph- ily, and movie exposure heterophily are greater among the common dyads than among the friendship dyads (Table 8). The only heterophily dimension on which the calculated "t" value is found tO be significant at the .05 level, is change contact (t = 2.15). Thus, there is a greater degree of heterophily with respect to change contact only, among dyads engaged in both information-seeking and friendship communi- cation than among dyads engaged only in friendship communi- cation. With regard tO the degree of heterophily among (1) common dyads, (2) dyads engaged only in information-seeking, and (3) dyads engaged only in friendship communication, the findings suggest that low heterophily (or high homophily) in status is relevant and perhaps necessary for friendship communication to occur, and high heterophily (or low homoph- ily) in change contact is relevant and might be necessary for information-seeking communication to occur among rural peOple in India. 92 We found support for General Hypothesis 1, with re- spect to the three heterophily dimensions emerging out Of the factor analyses of the heterophily scores on the selec- ted variables. TO explore further the degree of heterophily among dyads with regard to the variables which constitute the heterOphily dimensions, we also analyzed the data to find out the degree of heterophily with respect to each of the eight variables among dyads engaged only in information- seeking, only in friendship communication, and among dyads engaged in both. Table 10 presents the mean heterophily scores, the standard deviations of the heterophily scores on eight variables, and the number Of dyads engaged in the different communication situations. Table 10 shows that the mean heterophily scores for the eight variables among information-seeking dyads are greater than (1) among common dyads, and (2) among friendship dyads. The mean heterophily scores for the eight variables are again greater among com- mon dyads than among friendship dyads. The calculated "t" values comparing the mean heterophily scores Of information- seeking (only) dyads and friendship (only) dyads, are sig- nificant at the .05 level (Table 11). This finding provides additional support for the General Hypothesis 1. 93 mmH 3H 3H mmH 3H 3H mmH mvH z n88? . . . . . . . . 83380 meanest mm 0 mo H mm H om o mm o mm o co H Hm o Onomcmum use mConom Hm.o MH.H mo.H mm.o vm.o hm.o MH.H mm.o can: LGOHumEHOmnH cuom own own om» mmh own was own ram 2 a a .0 o I a a a Gaga-“EU “20 mm 0 3 H N.H_, H Hm 0 mm 0 NH. o 00 H E. o Bungalow c938? vm.o mm.o ma.o mm.o om.o mm.o mo.H He.o coax mflcmucmwum 5mm 5mm 5mm 5mm 5mm 5mm 5mm mHm z . . . . . . . . anfiuflflmfi mo H NH H mm H mm o mm o mm o mH H mm o spnmncmum thO mcwxwom oo.o em.H hN.H vo.H mH.H m~.H m~.H Hm.o cams. I:OHumeuownH #03300 Ph.—”>3 OBRRde .flhfid oBRHde mHWMDflH mHRBEfiv .md menu NERO 35.02 mg nwmmmmswz IOQOEmOU Im>ocaH HOE Leona m m COH um so. H m .93 33m GOHOMOHGDEEOU womucou mmcmcu mdumum hHHcmououom .chHumsuHm coHuMOchEEOO unmuOMMHO cw mommo moose mOHnoHHo> OmuOOHOm any so hHHcmououoc cmmE CH mOOCONOMMHp How mosHm> :u: vmumHsonoll.0H mHmda 94 .08... 83.3 9:0 “OOH mo. B» pm “35335."? mend aHfiBeHh mm.o «mm.~ «mo.H 0H.H mo.o H~.o Hm.H «mv.~ .m> muflficCQEbO mnflncggfibo no.0 «HH.N mm.H mm.o «mm.e «mn.v OH.H «Nv.m .m> mqflmmwthumfififiCH was aflmfiaflu «mo.~ «om.m emm.e «mn.m amm.m «mm.o enm.m «Hm.s .m>u mGUEOHLaHumEHfiCH oBRHQXm uuflfiao mbfifidxm mmmeHH mQQHSHp mfiflndm HRH» c.2002 ”Home “0mg IOQQEmOU :9an mo Inga. mumou . mQHBU dflmq MHHCQOHOHOC Coo: CH mOOCmuomeQ How mOCHo> :9: I . .. HOHHHHBNEHO .mCOHumCuHm COHUMOHCCEEOO HCOHOMHHO CH mpmwo mCOEm mOHCMHHm> OouooHom on» C0 hHHCQOHOumC Coma CH mmOCmHomeo MOM mmCHo> a»: OOHMHCOHMOII.HH mHmCB 95 The calculated "t" values for differences in mean heterophily between information-seeking (only) and common dyads are significant at the .05 level with respect to caste (t = 2.42), level of living (t = 4.76), innovative- ness (t:= 4.35), and change agent contact (t = 2.11). The ”t” values comparing heterophily among common dyads and friendship (only) dyads are significant at the .05 level only for caste (t = 2.45), newspaper exposure (t = 1.68), and change agent contact (t = 2.52). Thus, (1) there is a greater degree of heterophily with respect to caste, level Of liVing, innovativeness, and change agent contact among dyads engaged only in information-seeking than among d yads engaged in voth information-seeking and friendship communication; and (2) there is a greater degree of heter- ophily with respect to caste, newspaper exposure, and change agent contact among dyads engaged in both information-seeking and friendship communication, than among dyads engaged only in friendship communication. 96 Heterophily in Information-Seeking and Friendship Communication and Village Modernity General Hypothesis 2: There is a greater degree of heterophily among dyads engaged in information- seeking_communication in more modern villages than in more traditional villages. Spearman rank-order correlation coefficients be- tween the ranks of the eight villages on modernity (from most to least modern) and on the degree of heterophily in information-seeking communication (from highest to lowest heterophily) were computed for each of the three heteroph- ily? dimensions (and for each variable also; Table 12). Frcnn Table 12, it is evident that the Spearman rank-order correlation coefficients between the degree of heterophily in all the three dimensions of heterophily in information- seeking communication and villages' modernity are not Significantly different from zero at the .05 level. Thus, General Hypothesis 2 regarding heterophily in information- Beekjdng communication and village modernity is not supported. The heterophily dimensions on which the relation- ships between the degree Of heterophily in information- seeking communication and village modernity are in the Predicted direction,. are change contact and movie exposure.. 97 TABLE 12.--Spearman rank-order correlation between the degree Of heterophily in information-seeking communication, and in friendship communication, and village modernity. ExeanmniRmfivOnka:Oanxflatfl31for1M0 Canmmddmjonfifixuatkns HeUxophfly'hi... Information-seeking Friendship Ccmmmication Status -.19 -.55 Change contact .50 .69* Movie exposure .29 .43 Caste -.37 -.60 Education -.36 -.69* Level Of living -.19 -.05 Innovativeness .07 -.21 Cosmopoliteness .55 .64* Newspaper exposure .48 .69* Change agent contact .57 .69* Movie exposure .29 .43 *Significantly different from zero at the .05 level, one-tailed test. NO significant relationships between heterophily in the eight variables (separately) in information-seeking com- munication and village modernity are Observed. General Hypothesis 3: There is a greater degree Of heterophily among dyads engaged in friendship communication in more-modern villages than in more traditional villages. 98 Among the three heterophily dimensions, only change contact heterophily in friendship communication is signifi- cantly related to village modernity (rS = .69, which is sig- nificantly different from zero at the .05 level) in the pre- dicted direction. Thus, General Hypothesis 3 is not sup- ported. Although the relationship between movie exposure heterophily in friendship communication and village modern- ity is in the predicted direction (rS = .43), the same be- tween status heterophily (in friendship communication) and village modernity is in the Opposite direction to the one predicted (rS = .55). Analyzing the relationships between the degree of heterophily in each of the selected variables and village modernity, it is Observed that heterophily in three vari- ables (cosmopoliteness, newspaper exposure, and change agent contact) which constitute the change contact heterophily dimension, are significantly related to village modernity (rs = .64, 69, .69, respectively; all significantly differ- ent from zero at the .05 level). HeterOphily in education (in friendship communication) is found to be significantly related to village modernity, but in the Opposite to the direction postulated (rS = 0.69, significantly different 99 from zero at the .05 level). The findings about the rela- tionship between the degree of heterophily in friendship communication and village modernity do not provide enough evidences of support for General Hypothesis 3. Communication Integration in Information- Seeking and Friendship Communication General Hypothesis 4: Communication integration is higher in information-seeking communication than in friendship communication. The average communication integration scores of the 680 respondents in information-seeking and in friendship communication are .08 and .46 respectively, with (l) a mean difference in communication integration of .38, and (2) a standard deviation of difference in integration of .38. Although the calculated "t" for differences between commun- ication integration in information-seeking and in friendship communication is significant (t = 26.0) at the .05 level, the mean communication integration score in friendship com- munication is found to be greater than the mean communica- tion integration score in information-seeking communication. Thus, the calculated t is significant but the difference in 100 the two means is not in the expected direction. General Hypothesis 4 is not supported. Communication Integration and Village Modernity General Hypothesis 5: Communication integration in information-seeking communication is higher in more modern villages than in more traditional villages. Spearman rank-order correlation coefficient between ranks Of villages according to their average communication integration scores in information-seeking communication, and ranks of villages in modernity (Table 13), was computed. The computed rs is .31, which is not significantly different from zero at the .05 level. Thus, General Hypothesis 5 is not supported. 101 TABLE l3.--Communication integration scores in information- seeking and friendship communication in eight Indian villages. I I Communication Integration Scores and Ranks Villages in Order of Information-Seeking Friendship Communication Modernity 0D h u>lcw) Integration Rank Integration Rank Scores Scores 1 Kanchumarru .1193 2 1.0521* 1 2 Polamuru .0558 6 .4940 3 3 Mulawa .0915 3 .5113 2 4 Pophali .0239 8 .4533 4 5 Manchili .1637 l .3948 5 6 Amdole .0913 4 .3338 8 7 Harishpur .0625 5 .3667 6 8 Laxmidanga .0468 7 .3641 7 *Communication integration index can be more than 1.00. For example, in a communication net A, B, C, D, E are engaged in communication as follows: C+-—A-—»B Total communication links for A are: l AtOB=l A to C = l D A to D = 1 \‘E AtOE=2 5 Thus, individual A's communication integration index = 5/4 = 1°25. 102 General Hypothesis 6: Communication integration in friendship communication is higher in more modern villages than in more traditional villages. The Spearman rank-order correlation coefficient be- tween ranks Of villages according to their average communi- cation integration scores in friendship communication, and their ranks on modernity is .90, which is significantly dif- ferent from zero at the .05 level. Thus, the General Hypo- thesis 6 is supported. .- e. 1,. . . '1-lfidfl\n CHAPTER V SUMMARY AND CONCLUSIONS The present chapter includes a summary Of the present study, conclusions, and suggestions for future research. Summary The present dissertation is the result of an attempt tO answer the following four questions raised about rural communication systems (villages): 1) DO heterophily relationships between sources and re- ceivers differ from one communication situation to another? 2) How are the heterophily relationships between sources and receivers related to the modernity levels Of the rural villages? 3) Does the extent of integration through interpersonal contacts differ from one communication situation to another? 103 104 4) How is the degree of integration through interper- sonal contacts related to modernity levels of rural villages? Thus, the major Objectives Of the present study were: 1) to compare the degree Of heterophily in information- seeking and friendship communication; 2) to compare the degree Of communication integration in information-seeking and in friendship communication; and 3) to find out the relationships between modernity levels of rural Indian vil- lages and (a) degree Of heterophily, and (b) communication integration. The major concepts dealt with in the present study are: l) Heterophiiy is the degree to which pairs of inter- acting individuals are dissimilar in attributes. 2) Information-seeking communication occurs when an individual interacts with another to Obtain information, advice, or evaluation for making certain decisions. 3) Friendship communication occurs when an individual interacts with another to have an informal and intimate affective association. 4) Communication integration is the extent to which the units of a social system are interconnected through interpersonal communication links. ti: OI t0 (9' In 105 5) Modernization is the process through which systems (individuals, villages, nations, etc.) change from a tradi- tional way of life, to a more complex, technologically- oriented way of life. 6) Level of modernity Of a system is its relative con- dition or state in the modernization process, as compared to others at a particular point in time. Data used in the present study come from the Indian part Of a larger study, "Diffusion of Innovation in Rural Societies," conducted in Brazil, India, and Nigeria, by the Department Of Communication at Michigan State University. The present study is based on the analysis Of the data from the second phase Of the Indian research project, in which the purpose was to determine the factors affecting the in- novative behavior of Indian farmers in rural settings. Data were Obtained from 680 farmers in eight villages located in three states (Andhra Pradesh, Maharashtra, and West Bengal) through personal interviews. Six general hypotheses were postulated, three deal- ing with heterophily among dyads engaged in information- seeking communication and in friendship communication, and the other three dealing with communication integration. Among the six general hypotheses postulated, only two were 106 supported by the data (Table 14). The question then arises, why were only two of the six hypotheses supported, and the others not supported? Methodological Factors 1. Sample Selection In the larger study from which the data for the present study came, the respondents were purposively se- lected. For a study Of homophily-heterophily and communi- cation integration, the ideal sample would have been all the household heads in the villages. When an individual respondent reported that he sought information from, or had friendship communication with, someone who was not in the sample, information about such dyads were not available. In rural India, where age and experience have been tradi- tional, the major criteria for deciding from whom to seek information, the age restriction (not Older than 50 years Of age) in selecting the sample resulted sociometrically- named receivers who were not in our sample, specifically in the case of information-seeking communication. Village communication integration, in the present study, is essentially the average integration of the 107 N N . 80£fi> HERBS» 98s :H SflneQRWHHH>HBBOQCMBOEHJUCQflCHHCCBfiHVmGUmom .moQBHH> 033.08 088 233503 5 80% megs 88s Evacuees: 8: NaoflHRb uoz IHumooo Imomxo oH>Q= mo oonoo noumohm o «H oHonB Gum mm 988 CH . Sofia? Endgames 88 fi can 55 mmmfldt, “snu CHBHCCTHECCA . HHHHCCVCHEXBHTHBCCHCaHquflBEEHumqflmomLsHu . seasoned 0.53 3.53 umHmflooA Inflows." CH 80mg megs 0:96 sHHsdououon some mcoflmoHthuMw uoz uoz noono ICOO omcmno mo ooumoo noumoum m mH onona .mum mm ICOHCH CH momma meH . momoHHH> HMCOHHHOCH. onoE CH CED momoHHH> CuoHVoE ICo momho mcofio ouQe CH Cmnu momoHHH> CHoHdE oHQe CH COHHCOHCEBOO .mHflanHouon mo ”fiaUflHflRHm Chan mCfifiXhiSflBmEBHFHLGHHomawm.mUGHOmCQENdefific “RMRoo.HHFOHm r uoz LOB rupees: mfifim no 8.6% Hummus m H when. CTN 0s m 3 mg “Nine \ CRHHOQUCECBO .COHHCOHCQEOO mHanCoHHH CH Cmnu mCHuHoom nm..n.an0CoH..fl monEom COHumEHOHCH CH nommmCo woman mCOEosmHHnmonouon on: CH conu mConom “CoonHcmHm uConcomonCH Imomxo oH>OE mo oonmoo nousonm m mH ouonB COIN mm LCOHuoEHOHCH mo ”COHHMOHCQEUO mHanCoHum CH Cons. mCHeEomICOHH CH powwfihmocumm A IQERHCHHHHOQERComHHSohwaEwtfiHngHMDonufihfl oopuoemdm HCoOHmHCmHm ozu ICOO omcmno mo oonmoo nouoonm o ms muons "muH mm IHHuuo CHounoo . awesome . B uncommon 5H3 oOCoCoHHHU .COHdMOHCQECO mHnmoCoHum CH Conn \HHHanHouon mo How mConomlCOHpmeuomCH CH nomomco woman mCoem0>HHneO oonmoo noumonm $803203 smash Lope: mBBm mo 83% Hummus m mH mg a: mm m .3 when. his dad uoz some \Cmtfimmaw lHflmom m .momflflbmbm mmfionflRéfln commonsense omen. H83 33m HBHCHQE Hag Hohgmm. EONHSB “1 .honum uComoum one CH ooumou mononuommn onp mo muHCmoH HO muoEECmII.vH mqm<9 108 Buuoaasm unmodHemHm HEoHoH Hmooo COHumHoHMOO noonOIquou Cgoomm . momoHHH> . momMHHH> HoCQHuHooHH oHQe CH Coco. HMCOHHHOMHH once CH Conn mommHHH> Cuo moooHHH> CuoHch 988 CH nooE oHOE CH COHHCOHCQEOO mHnmocoHHH . CH noann mH COHHoOHCHCCUo meHH COHHMOHCNEBUO HmCOmuomnouCH nag mHanCoHHH CH COHHon oouooCCopnouCH ouoE ono mHochH>HOCH .. .0 new IoHCH COHHMOHCQEOU .. .0 new Buuoaaam pamoH H908 noz uoz HCoHOHHHoOO COHuoHoHuOO Commotion Cmfioomm .mmmflH? H803 . momMHHH> HoCOHH IHHcmHH oHoE CH Conn momoH nHHcouu once CH CCnu. moOMHHH> Choer once IHH> Choooe oHoE CH nonmeHn CH COHHCOHSECCO mCHcHoomICOHugOmCH CH . mH COHHCOHCQEUO OConom meHH COHHMOHCQEUO HoCOmHomHoHCH nmconnu ICOHHQDOHCH CH COHuon oouooCCOOHouCH oHoE one mHochH>HOCH "mime IouCH COHCCOHSEBCO “mime COHHOoHHHV ouHm IOQQO on.» CH mH ooCoHow onEmm memo mo mCmoE 96. Coosfion .COHHOOHCQEUO mmflnm .COHuoOdCCBOO IHMOHHH CH non» COHuoOH mHancCoHHm CH Conn mCHnoomlCOHuQDOHCH CH ICQEUO ngCOHum—CHOH Rhonda 4H0 mfi use autonomous Ba: 0332:2580 HERHQRE £0285 in 5 .8202 H SUNS uoz uCoOHdeon How Hoopla oopooCCOPOHCH oHoE one mHosoH>HEH "mime IOHCH COHHCOHSEBU "mime N 358 N 3002?, .HSqBHRb Iammomo ouoE CH Cofi. momoHHH> CuopoE oHQe CH COHHCOHSE CCoOHHHCmHm . 150 eHanCoHum CH oomomCo mummo mco—Cm MHHnmououon .momoHHH> uoz COHu oncmomxo oH>oE HO oonoo Monmoum o .H oHonB "bum. mm HMCOHHHOMHH reHouhoo~ oHQC CH Conn .8OMHHH> HSOHUHUMHU mOOMHHAKw SHOE: N Hoop once CH cmnu momoHHH> CHoHOOE one CH COHHMOHCEBCO oHoE CH CQHu oopuom eHnmoCoH CH CH oomomco woman mCQCo >HHnmonouon Hosp long—Coo mdnm Imam uCCOHmHCmHm x58 ICOO omCmno mo oonoo noumonm m. H ouonm. "mum. mm Igwum CH momma #2 o WTUMHHHKV Hag I8 ME.” 8% CME HOME oHOE CH Conn mommHHH> Choooe oHOE CH COHuoO hHHanuouon HO uCoOH mHCmHm IHmon tHCdESO 9..ande Hm CH oomomCo momho mCOEM CHHHHEO ooumoo Houmon r uoz r Lopes Rheum mo 8.68 Honcho m E 03$ arm. as m H mg- Sec 109 selected respondents in a village. As there were several dyads for whom no data could be obtained, and the sample was not a random one, the measure of village communication integration represents only the degree of integration of the purposively-selected respondents. Similarly, the mea- sure of village modernity was also influenced by the pur- posive selection of respondents. 2. Sample Villages The states and districts in which the eight vil- lages are situated, were selected purposively. Thus, the eight villages are not representative of Indian villages in general. For four of the six general hypotheses, the units of analysis were villages. As mentioned earlier, only eight villages were studied. Such a small number of villages restricted the range of variability in levels of modernity, communication integration, and the degree of heterophily among dyads. 3. Measurements Information-seeking communication was measured by sociometric questions, such as, "If you needed advice on . . . from which one person . . . would you seek advice first?" Thus, the choices made are preferred or evaluative 110 or hypothetical, rather than actual. There was no pro- cedure to check whether a nominator did actually seek ad- vice from the nominee or not. Theoretical Factors In addition to these methodological factors, there is a need to reexamine the theoretical considerations re- garding homOphily-heterophily, communication integration, and modernity. Two of the hypotheses for which no support were provided by the data, deal with the relationship be- tween heterOphily and levels of village modernity. We postulated a positive relationship between the degree of heterophily in both information—seeking and friendship communication situation, and levels of modernity of vil- lages. The findings suggest that the degree of change contact heterOphily is positively related with village modernity in both information—seeking and friendship com- munication, but significantly only in the latter. Simi- larly, the relationship between movie exposure heterophily and village modernity was positive. A negative relation- ship between status heterophily and village modernity was found in both communication situations. This latter find- ing suggests that with the process of modernization, lll increased solidarity among caste groups and social classes may occur in rural India. Thus, the assumption that the attributes on which heterophily is relevant is the same in more modern villages, as well as in more traditional vil- lages, is questionable. The hypothesis regarding higher communication in- tegration in information-seeking communication than in friendship communication, was not supported. Tradition- ally, information-seeking communication has been limited to neighbors, friends, and extended family members. Friendship groups (also referred to as "gossip groups”) have also been sources of information-sharing among rural villagers. The assumption, implicit in the hypothesis re- garding higher communication integration in information- seeking than in friendship communication, was that with the introduction of agricultural innovations, new credit systems, etc., individuals would identify the few knowl- edgeable individuals in their village and seek information from them. Possibly, information-seeking as such is not a distinguishable phenomenon in rural Indian villages, and to a certain extent, information-seeking is included in friendship communication. 112 The argument for not finding non-support for the hypothesis about higher communication integration in information-seeking communication in more modern villages than in more traditional villages, is parallel to that mentioned just previously. Conclusions Studies dealing with source-receiver relationships in persuasion research, diffusion research, and in small groups, have produced conflicting results. In previous studies, heterophily has been studied without emphasis on the attributes and their relationship with the kind of interaction in which individuals are involved. In the present study, our emphasis on the questions: "Heterophr ily on what, and why?" provided answers about the different dimensions on which a greater degree of heterophily exists in information-seeking communication than in friendship communication. In any communication situation, the interacting individuals probably evaluate each other on the dimensions of safety, qualification, and dynamism. The attributes or the dimensions on which low or high heterophily is relevant 113 in a communication situation, differ from one communication situation to another, and from one system to another. rural systems modernize, there is no specific trend of in- creased or decreased heterophily among both information- seeking and friendship dyads. Within the limitations stated earlier, we conclude that rural communities in India are more integrated in their communication networks through friendship communica- tion links than through information-seeking communication links. Villages which are relatively more modern are more integrated communicatively than the less modern villages through friendship communication links. Implications for Change Agencies The findings of the present study offer the follow- ing suggestions for change agencies re3ponsible for the planning and implementation of technological change and the diffusion of innovations: 1) Information flow in rural societies: In rural societies information flows from individuals of higher status who are more innovative, cosmopolite, have more newspaper exposure, and change agent contact. It is 114 expected that these high-status individuals serve as sources of information in rural societies. A change agent's objective should not be to identify the indi- viduals of highest statuses only. A change agent should identify the different strata of status and then identify the higher-status individuals and the lower-status social groups for whom the former are sources of information. 2) Communication strategy as rural systems become more modern: As rural societies develop and become more modern, individuals who are more cosmopolite and/or have greater newspaper exposure become equally important sources of information to those with higher change agent contact. Thus after the initial stage of implementation, the communication strategy should be to diffuse informa- tion through newspapers and other city-oriented (mass) media instead of widespread face-to-face contact by the change agents. 3) Information-seeking and friendship communication: Information-seeking as such is not a predominant phenom- enon in Indian villages. These systems are highly inte- grated through friendship communication links. Thus, a change agent's strategy should be to use the existing friendship groups and attempt to transform the groups' objectives to innovation diffusion. 115 4) Integration and modernity: In more modern vil- lages, change agents do not have to make such widespread contacts. As the more modern villages are more integrated through friendship relationships, it is economical for the change agencies to contact fewer individuals who have higher outside orientation and newspaper exposure. As the villages become more modern, individuals communicate with others who are more different than they would in a more traditional village. Suggestions for Future Research To some extent, the present study is exploratory in nature. Several questions raised in earlier sections provide the lines of research needed on heterophily rela- tionships among interacting dyads, and communication inte- gration in communities. The suggestions for future re- search are summarized below: 1) Objective measures of heterophily were used in the present study. Research is needed on perceived heterophily among dyads, and its relationship with objective measures. 116 2) In the present study, heterophily was measured as absolute differences in attributes between individuals in a dyad. Thus, if heterOphily exists between a source- receiver dyad, the direction of the difference, i.e., whether the source scores higher than the receiver or not, is not known. In future studies, heterophily needs to be measured as differences in attributes of source-receiver dyads taking direction into consideration. We need to analyze dyads with sources scoring higher than receivers, and dyads with sources scoring lower than receivers, separately. This separate analysis may provide some meaningful insights into the nature of heterophily in different communication situations. 3) The limitation in the selection of respondents in the present study suggests that in future studies of heter- ophily and communication integration, attempts should be made to include at least all the individuals who receive a communication choice. 4) The question raised earlier is whether information- seeking communication can be differentiated from friendship communication, or not. The question becomes more important when one is studying rural villages. In other words, does some sort of information-Sharing also occur in friendship 117 communication in rural traditional societies, where the primary groups have been the sources of all information for a long time? To gain more insight into this problem, the participant observation method of research is sug- gested. S) No attitudinal variables were included in the heterophily measures. Research is needed to find out the degree of heterophily in attitudes in information-seeking and friendship communication. Attempts need to be made to identify the relevance of the kinds of attitudes in combin- ation with attributes on which heterOphily is necessary for a particular type of communication to occur. 6) Further research is needed to answer the question "Why heterophily?" more thoroughly. Attempts should be made to find out how the answers to the question differ ‘in different communication situations dealing with differ- ent types of information. 7) The area of heterophily in confidential or taboo communication needs to be explored further. Studies deal- ing with family planning communication, communication on abortion, etc., suggest that low heterophily exists be- tween sources and receivers on such taboo topics. Further studies are needed to find out the attributes on whiCh low 118 or high heterophily is relevant in such taboo communication as that dealing with venereal disease, homosexuality, etc. 8) The present study provided findings on heterophily among dyads engaged in two types of communication situa- tions. Further exploration is needed to find out whether a certain typology of dyads emerges among dyads engaged in different communication situations. BIBLIOGRAPHY BIBLIOGRAPHY Alleyne, E. P., and Coolie Verner 1969 Personal Contacts and the Adoption.of Innova- tions. British Columbia: Department of Agricultural Economics, University of British Columbia. Altman, Irwin, and Elliot McGinnies 1960 "Interpersonal Perception and Communication in Discussion Groups of Varied Attitudinal Compo- sition." Journal of Abnormal and Social Psy- chology, 60:390-395. Attah, Efiong B. 1968 "An Analysis of Polymorphic Opinion Leadership in Eastern Nigerian Communities." East Lansing, Michigan: Department of Communication, Michigan State University, Unpublished M.A. Thesis. Back, Kurt W. 1951 "Influence through Social Communication." Journal of Abnormal and Social Psychology, 46: 9-23. Barnlund, Dean C., and Carroll Harland 1963 "Propinquity and Prestige as Determinants of Communication Networks." Sociometry, 26: 467-479. Bennis, Warren G., and others 1962 The Planning of Change: Readings in the Be- havioral Sciences. New York: Holt, Rinehart and Winston. Berlo, David K. 1960 The Process of Communication. New York: Holt, Rinehart and Winston. 119 120 Berlo, David K. 1969 "Human Communication: The Basic PrOposition." (Mimeo), East Lansing, Michigan: Department of Communication, Michigan State University. Berlo, David K., and others 1969 "DimensiOns for Evaluating the Acceptability of Message Sources." Public Opinion Quarterly, 33: 563-576. ‘ ' Berscheid, Ellen 1966 "Opinion Change and Communicator-Communicatee Similarity and Dissimilarity." Journal of Personality and Social Psychology, 4: 670-680. Bettinghaus, Erwin P. 1968 Persuasive Communication. New York: Holt, Rinehart and Winston. Black, C. E. 1966 The Dynamics of Modernization: A Study_in Comparative History. New York: Harper and Row Publishers. Blau, Peter M. 1960 "A Theory of Social Integration." American Journal of Sociology, 65: 545-556. Blau, Peter M. 1962 "Patterns of Choice in Interpersonal Relaé tions." American Sociological Review, 27: 41-56. Bonney, M. E. 1946 "A Sociometric Study of Some Factors to Mutual Friendships on the Elementary, Secondary, and College Levels." Sociometry, 9: 21-47. Booth, Alan and Nicholas Babchuk 1972 "Informal Medical Opinion Leadership among the Middle Aged and Elderly." Public Opinion_guar- terly, 36: 87-94. 121 Bordenave, Juan Diaz 1966 "The Search for Instrumental Information among the Farmers of the Brazilian Northeast." East Lansing, Michigan: Department of Communica- tion, Michigan State University, Unpublished Ph.D. Thesis. Bose, Shanti P. 1967 "Social Interaction in an Indian Village." Sociologia Ruralis, 7: 156-175. Brock, T. E. 1965 "Communicator-Recipient Similarity and Decision Change." Journal of Personality and Social Psychology, 1: 650-654. Broxton, J. 1963 "A Test of Interpersonal Attraction Predictions Derived from Balance Theory." Journal of Ab- normal and Social Psychology, 66: 394-397. Byrne, Donn 1961 "Interpersonal Attraction and Attitude Similar- ity." Journal of Abnormal and Social Psychol- ggy, 62: 713-715. Carlson, Gosta 1965 "Time and Continuity in Mass Attitude Change: The Case of Voting." Public Opinion Quarterly, 29: 1-15. Cartwright, Dorwin 1968 "The Nature of Group Cohesiveness," in Darwin Cartwright and Alvin Zander, Group Dynamics: Research and Theory. New York: Harper & Row. Chou, Teresa K. M. 1966 "HomOphily in Interaction Patterns in the Dif- fusion of Innovations in Colombian Villages." East Lansing, Michigan: Department of Communi- cation, Michigan State University. Unpublished M.A. Thesis. Coleman, James S. 1958 "Relational Analysis: The Study of Social Organ- izations with Survey Methods." Human organiza- tion, 16: 28-36. 122 Coleman, James S., Elihu Katz, and Herbert Menzel 1966 Medical Innovation: A Diffusion Study. New York: The Bobbs-Merrill Co. Inc. Duff, Robert W. 1971 "Neighborhood of Residence and Knowledge, Atti- tudes, and Practices with Regard to Conception Control in a Philippine City." South Bend, Indiana: Department of Sociology and Anthro- pology, University of Notre Dame. Unpublished Ph.D. Thesis. Durkheim, Emile 1960 The Division of Labor in Society. Glenco, Illinois: The Free Press. Emery, F. E., and O. A. Oeser 1958 Information, Decision and Action: A Study of the Psychological Determinants of Changes in Farming Techniques. Victoria: Melbourne Uni- versity Press. Feldman, Sidney P. 1966 "Some Dyadic Relationships Associated with Con- sumer Choice." Paper presented at the American Marketing Association, Bloomington, Indiana. Festinger, Leon 1950 "Informal Social Communication." Psyghological Review, 57: 271-282. Festinger, Leon, Stanley Schachter, and Kurt Back 1950 Social Pressure in Informal Groups. Stanford: Stanford University Press. Festinger, Leon and John Thibaut 1951 "Interpersonal Communication in Small Groups." Journal of Abnormal and Social Psychology, 46: 92-990 Fliegel, Frederick C., and others 1968 Agricultural Innovations in Indian Villages. Hyderabad, India: National Institute of Com- munity Development. 123 Guimaraes, Lytton 1968 "Matrix Multiplication in the Study of Inter- personal Communication." East Lansing, Michi- gan: Department of Communication, Michigan State University. Unpublished M.A. Thesis- Guimaraes, Lytton 1970a "Network Analysis: An Approach to the Study of Communication Systems." A Prelim. Paper, Department of Communication, Michigan State University. Guimaraes, Lytton 1970b "Communication Integration: A Review of the Literature and some Emerging Hypotheses." A Prelim. Paper, Department of Communication, Michigan State University. Joyce, Richard E. 1970 "A Review of Literature on Source-Receiver Similarity and Inferences about Traditional- Modern Communication." A Preliminary Examina- tion Paper, Department of Communication, Michi- gan State University. Ho, Yung Chang 1969 "Homophily in the Diffusion of Innovations in Brazilian Communities." East Lansing, Michi- gan: Department of Communication, Michigan State University. Unpublished M.A. Thesis. Jones, E. E., and B. N. Daughtery 1959 "Political Orientation and the Perceptual Ef- fects of an Anticipated Interaction." Journal of Abnormal and Social Psychology, 59: 340-349. Kahlon, A. S., and M. Kaushal 1967 Adoption of Improved Agricultural Practices in the Plains of the Punjab. Ludhiana, Punjab: Department of Economics and Sociology, Punjab Agricultural University. Katz, Elihu and Paul S. Lazarsfeld 1955. Personal Influence: The Part Played by People in the Flow of Mass Communication. New York: The Free Press. 124 Klapper, Joseph T. 1960 The Effects of Mass Communication.. New York: The Free Press. Landecker, Werner S. 1950 "Smend's Theory of Integration." Social Forces, 29: 39-48. Landecker, Werner S. 1951 "Type of Integration and their Measurement." American Journal prSociologxo 56:a332-340. Larsen, Otto N., and Richard J. Hill 1958 "Social Structure and Interpersonal Communica- tion." American Journal of Sociology, 63: 497-505. Lazarsfeld, Paul F., and Herbert Menzel 1963 "Mass Media and Personal Influence." in W. Schramm (ed.), The Science of Human Communica- tion. New York: Basic Books. Lazarsfeld, Paul F., and Robert K. Merton 1954 "Friendship as Social Process: A Substantive and Methodological Analysis." in Morroe Berger and others (eds.), Freedom and Control in Modern Society. New York: Van Nostrand. Lee, Nancy H. 1969 The Search for an Abortionist. Chicago: University of Chicago Press. Lionberger, Herbert F. 1953 "Some Characteristics of Farm Operators Sought as Sources of Farm Information in a Missouri Community." Rural Sociology, 18: 237-238. Lionberger, Herbert F. 1955 Information Seeking Habits and Characteristics of Farm Operators. Research Bulletin #581, Columbia: College of Agriculture, University of Missouri. Lionberger, Herbert F. 1959 "Community Prestige and the Choice of Sources of Farm Information." Public Opinion Quarterly, 23: 111-118. 125 Lionberger, Herbert F., and Rex R. Campbell 1971 "Structural Influences of Personal Attributes on Social and Informational Relationships of Farm Operators in a Missouri Community." Rural Sociology, 36: 533-546. Lionberger, Herbert F., and M. C. Coughenour 1957 Social Structure and Diffusion of Information. Research Bulletin #631, Columbia: Missouri Agricultural Experimental Station. Loomis, Charles P. 1946 "Political and Occupational Cleavages in a Hanoverian Village, Germany: A Sociometric Study." Sociometry, 9: 316-333. Lott, Albert J., and Bernice E. Lott 1965 "Group Cohesiveness as Interpersonal Attrac- tion: A Review of Relationships with Ante- cedent and Consequent Variables." Psycholog- ical Bulletin, 64: 259-309. Menzel, Herbert 1959 "Social Determinants of Physicians' Reactions to Innovations in Medical Practice." Madison: University of Wisconsin. Unpublished Ph.D. Thesis. Menzel, Herbert 1960 "Innovation, Integration, and Marginality: A Survey of Physicians." American Sociological Review, 25: 704-713. Merton, Robert K. 1957 Social Theory and Social Structure. New York: Free Press. Newcomb, Theodore M. 1956 ”The Prediction of Interpersonal Attraction." The American Psychologist. 11: 575-586. Patel, Narsi 1964 "Shift from Primary to Secondary Reference in Interpersonal Communication." Paper presented at the Rural Sociological Society Annual Meet- ing at Montreal, Canada. g, . =._.__. 11% .53.: _§W§M 126 Patel, Narsi 1966 "Status Determinants of Interpersonal Communi- cation: A Dyadic Analysis." Lexington: Uni- versity of Kentucky. Unpublished Ph.D. Thesis. Pool, Ithiel de Sola 1963 ”The Mass Media and Politics in the Moderniza- tion Process," in Lucian W. Pye (ed.), Commun- ications and Political Development. Prince- ton: Princeton University Press.. Precker, J. A. 1952 "Similarity of Valuings as a Factor in Selec- tion." Journal of Abnormal and Social Psy- chology, 47: 406-414. Pye, Lucian W. 1963 Communications and Political Development. Princeton: Princeton University Press. Rahim, Syed A. 1961 Diffusion and Adaption of Agricultural Prac- tices: ‘A Study of Patterns of Communication. Comilla: Pakistan Academy for Village Devel- opment. Rahim, Syed 1965 Communication and Personal Influence in an East Pakistani Village. Comilla: Pakistan Academy of Rural Developm ent. Rahudkar, W. B. 1960 "Local Leaders and the Adoption of Farm Prac- tices." The Nagpur Agricultural College Maga- zine, 34: 1-2. Raju, K. S. S. 1969 . "Peasant Typology and Communication Behavior: A Factor Analytic Study in Six Indian Vile lages." East Lansing: Michigan State Univer- sity. Unpublished M.A. Thesis. Rao, Y. V. Lakshmana 1966 Communication and Development:,.AmStudypofTwo. Indian Villages. .Minneapolis: University of Minnesota Press. 127 Richardson, Helen M. 1940 ”Community of Values as a Factor in Friendships of College and Adult WOmen.” Journal of Social Psychology, 11: 303-312. Rogers, Everett M. 1962 Diffusion of Innovations. New York: Free Press. Rogers, Everett M., and Dilip K. Bhowmik 1971 "Homophily-Heterophily: Relational Concepts for Communication Research.” -Public Opinion gnarterly, 34: 523-538. Rogers, Everett M., and Rabel J. Burdge 1962 Community Norms, Opinion Leadership and Inno- vativeness among Truck Growers. Research Bul- letin #912, Wooster, Ohio Agricultural Experi- ment Station. Rogers, Everett M., and Frank O. Leuthold 1962 Demonstrators and the Diffusion of Fertilizer Practices. Research Bulletin #908, Wboster: Ohio Agricultural Experiment Station. Rogers, Everett M., and Johannes C. van Es‘ 1964 Opinion Leadership in Traditional and Modern Colombian Peasant Communities. Diffusion of Innovation Research Report #2, East Lansing: Michigan State University. Rogers, Everett M., with Lynne Svenning 1969 Modernization amonngeasants: The Impact of Communication. New York: Holt, Rinehart and Winston. Roy, Prodipto and others 1968 . Agricultural Innovation among Indian Farmers. Hyderabad, India: National Institute of Community Development. Runkel, Phillip J. 1956 "Cognitive Similarity in Facilitating Communi- cation." Sociometgy, 19: 178-191. 128 Savale, R. S. 1966 ”Technological Change in Agriculture: A Study of Sources of its Diffusion, Efficiency of these Sources, and the Economic Factors Af- fecting the Adoption of Improved Practices." Indian Journal of Agricultural Economics, 21: 199-208. ' Schachter, Stanley 1951 "Deviation, Rejection and Communication." Journal of Abnormal and Social Psychology, 46: 190-207. Schachter, Stanley 1968 "Social Cohesion," in David L. Sills (ed.), International Encyclopedia of the Social Sciences. New York: The MacMillan Co. and Free Press. Schramm, Wilbur 1963 "Communication Development and the Development Process," in Lucian Pye (ed.), Communications and Political Development. Princeton: Prince- ton University Press. Schuessler, Karl 1971 Analyzinngocial Data. Boston: Houghton Mifflin Company. Sen, Lalit K. 1968 "The Concepts of Tradition and Modernity.” Paper Presented at the Second World Congress of Rural Sociology, West Berlin. Sen, Lalit K. , 1969 Opinion Leadership in India: A Study of Inter- personal Communication in Eight Villages. Hy- derabad, India: National Institute of Commun- ity Development. Simons, Herbert W., and others 1970 "Similarity, Credibility, and Attitude Change: A Review and A Theory." Psychological Review, 129 Smend, Rudolf 1928 Verfassung und Verfassungs-Recht. Munich und Leipzig: Duncker und Humbolt. Smith, A. 1957 "Similarity of Values and its Relation to Acceptance and the Projection of Similarity." Journal of Psychology, 43: 251-278. Sorokin, Pitrim A. 1937 Social and Cultural Dynamics. Vol. I and Vol. III, New York: American Book Company. Sorokin, Pitrim A. 1957 Social and Cultural Dynamics. One Volume Edi- tion. Boston: Porter Sargent Publisher. Triandis, Harry C. 1959 "Cognitive Similarity and Interpersonal Com- munication in Industry." Applied Psyshology, 43: 321-326. Troldahl, Verling C., and Robert Van Dam 1965 "Face to Face Communication about Major Topics in the News." Public Opinion Quarterly, 29: 626-634. van den Ban 1963 De Communicatie van Niewe Landbouwmethoden. Assen, Netherlands: Royal Von Gorcum. Vick, Charles F., and Roy V. Wood 1969 "Similarity of Past Experience and the Commun- ication of Meaning." Speech Monograph, 36: 159-162. Weiner, Myron 1966 Modernization. New York: Basic Books. Winslow, C. N. 1937 "A Study of the Extent of Agreement Between Friends' Opinions and their Ability to Estimate the Opinions of Each other." Journal of Social Psychology, 8: 433-442. 130 Wood, Roy V., Joanne S. Yamauchi, and James J. Bradac 1971 "The Communication of Meaning Across Genera- tions." Journal of Communication, 21: 160- 169. Yadav, Dharam P. 1967 "A Comparative Analysis of Communication Structure and Innovation Diffusion in Two Indian Villages." East Lansing: Michigan State University, Unpublished Ph.D. Thesis. 1 ; ‘m-‘AJ APPENDIX 131 HOMHCOU oo.H am. am. so. NH.- me. am. am. mm. ucmma mmqmno mCH>HA oo.H hm. mo.| MH. mm. MH. mm. mm. m0 Ho>oq oo.H mo.| mo. mm. mm. mm. Hm. mmoCo>Hum>0CCH oHCmomxm oo.H mH.| mo. «H. Ho. mo.| oH>oz mCHCoumHA oo.H NH.I vH.| mo. Ho.| OHcmm oHCmomxm oo.H ow. mm. mm. Homomm3oz mmoC oo.H CH. ma. umuHHoCoCmoo oo.H Hm. CoHuooscm vomuCoo mCH>HA -mCH ousmomxm mmoCo>Hu onsmomxm mmoCouH Coau UComm mo . ICoumHA Comma . . oumou Im>0CCH oH>oz . IHOQOEmoU noosvm omCoCU Ho>oq . oHpom ImBoz QMHQDBm mH MAHmmommBmm mUHSB ZO mmqdem<> Omeomqmm mZHZ WEB ZMWZBmm ZOHfiddmmmOU MMQMOIommN XHszmmd r. .. Ill: "'TITliTILfifiqfljfllflafilLii'flt'ujfimfliflniWWW” 57 9324