7 5 I I I I l I i E l l g ‘V! ‘HOMOPHILY m INTERACTION PATTERNS IN THE “DIFFUSION 0|: INNOVATIONS IN COLOMB‘IAN VILLAG'ES Thesis for the Degree of‘ M. A. I 3 Teresa Kang Mei Chou MICHIGAN STATE UNIVERSITY 1966 u g: um; "12::ng gmw 1:! w: u Mug [a a L/ e HOMOPHILY IN INTERACTION PATTERNS IN THE: DIFFUSION OF INNOVATIONS IN COLOMBIAN VILLAGES BY Teresa.Kang Mei Chou A.THESIS Submitted to Rfichigan State university in partial fulfillment of the requirements for the degree of MASTER OF.ARTS Department of Communication 1966 ABSTRACT HOMOPHILY IN INTERACTION PATTERNS IN THE DIFFUSION OF MOVATIONS IN COLDMBIAN VILLAGES by Teresa Kang Mei Chou The present thesis was concerned with homophily in interaction patterns and the multi—step flow of comrmmication. ' Homophily is the degree of similarity on selected variables between pairs of individuals who are in interaction. The objectives of the present study were: (1) to examine some determinants of homophily (such as level of competence, communication contact, and status) in two kinds of interaction; friend— ship and information—seeh'hg interaction, and (2) to investigate the multi-step flow of communication. Homophily is expected because both the seeker and the sought seek cognitive cons onance through interaction with others like themselves in both friendship and information-seeking interaction. The research findings provide the evidence. Greater homophily is expected in the case of friendship interaction because people tend to interact with those who are more reliable and competent, and the seeker must perceive a superiority in the sought in order to consider him a useful and credible source, in information-seeking interaction. Two-step flow of communication stated that the first-step flow of communication was from mass media to opinion leaders , and the second step was from opinion leaders to followers. Since communication is a process without Teresa Kang Mei Chou beginning and ending, it is mere realistic to assume multiple steps in the flow of comrmmication. Since the soughts are more competent than the seekers, it is reasonable to assume that the soughts are exposed more to the mass media. The dependent variable was interaction. The operationalized de- terminant variables were functional literacy, innovativeness , mass media exposure, cosmopoliteness , social participation, age, and social status. The homophily of individuals under study was based on the variables mentioned above by using dyadic analysis. Dyadic analysis is a kind of investigation of the networks of relationships among pairs of individuals . The statistical method for the first two hypotheses was zero-order Pearsonian correlation; for the third hypothesis it was Z test; for the fifth hypothesis it was chi-square; for the fourth hypothesis observation was used. The data used in the present study grew out of the research pro- ject, "A Field Experiment of the Role of Opinion Leaders in Diffusing an Innovation in Three Colombian Neighborhoods." The project was directed by Everett M. Rogers in 1963, and the restudy was done by J. David Stanfield, Eduardo Ramos, and Elssy Bonilla de Ramos with approximately the same respondents in 1965. There were 160 interviews completed in 1963 and 136 in 1965. There were five hypotheses in the present thesis. The firt two hypotheses asserted that there is homophily of individuals engaged in friendship interaction and information-seeking interaction, respectively. The third hypothesis asserted that the homophily of individuals engaged in friendship interaction is. greater than the homophily of individuals engaged in information-seeking interaction. The fourth hypothesis Teresa Kang Mei Chou asserted that there are multiple steps in the flow of communication. The fifth hypothesis asserted that the soughts are exposed more to mass media than the seekers. The first, fourth, fifth hypotheses were confirmed. The second and the third hypotheses were not, but the findings agreed with the predicted positon. It was concluded that there is homophily of individuals engaged in friendship interaction; there are multiple steps in the flow of com- munication; the soughts are exposed more to the mass media than the seekers . ACKNOWLEDGMENTS The authoress is very grateful to her adviser, Dr. Everett M. Rogers , for his valuable suggestions , cooperation, and encouragement . She also wishes to express sincere appreciation to Dr. Hideya Kumata and Dr. Randall Harrison, who served on her M.A.. guidance committee. The authoress is indebted to Programa Interamericano de Informacion Popular, and Michigan State University, the sponsors of the research project, "A Field Ebcperiment of the Role of Opinion Leaders in Diffilsing of Innovation in Three Colombian Neighborhoods" in 1963, and the restudy in 1965, because they enabled the present research to be executed. To Nan Lin, Robbie Keith, and J. David Stanfield, she expresses a thank you for their helpful cormIents . Sandra Stanfield's assistance in editing this thesis and Ruth Langenbacher' s cooperation in typing it deserve an expression of gratitude. A final word of appreciation is expressed to her fiance Chuin- sheng Shen for his comfort, encouragement, and help all through her . graduate work . ii TABLE OF CONTENTS LISTOFTABLES . . . . . . . . . . . LISTOFFIGURES. . . . . . . . . . . . Chapter I INTRODUCTION . . . . . . . . . . Justification Objectives II RATIONALE AND HYPOTHESES . . . . . General Hypotheses Dealing with Homephily Choice of Variables General Hypotheses Dealing with the Multi- Step Flow of Communication III monomer _ . Data—Gathering Vereda and Munici io General Description of the Three Veredas Location Transportation Social Organization Method and Results of Data—Gathering Statistical Methods Operationalization of Variables Dependent Variable Determinant Variables Empirical Hypotheses IV FINDINGS . . . . . Test of Hypotheses Dealing with Homophily Summary of Hypotheses I, II, III Test of Hypotheses Dealing with Multi— Step Flow of Communication iii Page vi EC» 11 11+ 11+ 1” 1’4 15 15 15 15 16 19 19 19 2 3 27 27 3M 36 Chapter v SUMVIARY, DVIERPREI'ATION, IMPLICATION, AND SUGGESTIONS Summary Interpretation Implication fer Change Suggestions fer Future Researdh APPENDIX BIBLIOGRAPHY........ iv Page 1+6 1+6 1+8 51+ 5m 57 71 LIST OF TABLES Table 1 Results of data-gathering . . . . . . . 2 The zero-order Pearsonian correlations of characteristics on friendship interaction and information-seeking dyads . . . . . . 3 Z scores for differences between the information- seeking friendship homophily correlations for three Colombian villages . . . . . . . . 1% Differences in seeker-sought mass media exposure scores by steps in the multi—step flow of commotion............. 16 35 36 H2 Figure LIST OF FIGURES Paradigm of hypothesized relationships between seeker and sought in friendship and informa- tion-seeking interaction. . . . . . . Paradigm of the relationship of homophily and interaction . . . . . . . Steps of information flow among seekers and soughts in Clique l . . . . . . . Steps of infbrmation flow among seekers and soughts in Clique 2 . . . . . . . Steps of information flOW'among opinion leaders and fOllowers in Clique 3 . Mass media exposure scores of seekers and soughts in linked information flow in Clique l . . . Mass media exposure scores of seekers and soughts in linked infOrmation flow in Clique 2 . Mass media exposure scores of seekers and soughts in linked information flow in Clique 3 . Page 12 2O 38 39 H0 H3 44 us CHAPTER I INTRODUCTION JUstification Developing countries rely on imported technoIogy in order to modernize. Hopefully, for this imported technology to bring about change, ideas must (1) enter the social system from external sources, and ( 2) diffuse through the social system. A study of the introduc- tion and spread of new ideas is a study of commmication. How do new ideas enter a village system? New ideas can enter the village through mass media communication channels , extens iOn agent contacts , and villagers' contacts with outsiders in cities and other areas . The present thesis will only be concerned with the diffusion of ideas within villages after the new ideas have entered the system. In this case, the importance of mass and interpersonal communication Channels in difosing new ideas is not the same. Interpersonal come munication is more effective than.mass communication Channels in within- village diffusion. Schramm1(1962, p. 251) asserted that ". . . the boundaries of mass communication have become practically continuous with those of the entire field of communication study." It is difficult to study mass communication alone without taking interpers onal communication into account. One of Klapper's (1960, p. 8). generalizations is "Mass com— munication ordinarily does not serve as a necessary and sufficient 1 2 cause of audience effects, but rather functions among and through a nexus of mediating factors and influences." This is to say that mass cxmmunication are seldom.effective in Changing a person's behavior or attitudes. On the other'hand, it is often assumed that interpersonal communication plays an important role in influencing attitudes and be- havior. Roper (1960, p. xv) stated, “As the result of'my'own researoh into public attitudes I have come to the tentative conclusion that ideas often penetrate the public as a.whole slowly and--even.more important--very often by interaction of neighbor on neighborwwithout any apparent influence of the mass media." It seems that interpersonal communication is generally more effective than mass communication in changing attitudes and behavior. In the adoption process1 in the diffusion of innovations, there are five stages, awareness, interest, evaluation, trial, and adOption. According to Rogers, "Impersonal infermation sources are most important at the awareness stage, and personal sources are most important at the evaluation stage in the adoption process" (Rogers, 1960, p. 99). But this statement is not true fer developing countries. Mass media are not important Channels at the awareness stage (Rogers and Meynen, 1965; Rahim, 1961; Deutschmann and Fals Borda, 1962; Myren, 1962); inter- personal communication is most important at the awareness stage as well as at the evaluation stage, in villages in less developed countries. Possible explanations could be: (1) In villages, mass media exposure is low. lAccording to Rogers, the adoption process is the mental process through WhiCh an individual passes from.first hearing about an innova- tion to final adoption. 3 (2) The physical distance between cities and villages is a barrier to the geographic mobility of villagers. Villagers find it difficult to go to the cities where they might be exposed to mass media. (3) Because of low accessibility to urban.media facilities and.mass media.exposure, interpersonal.communication serves as a substitute (in terms of making people aware of innova- tions imported from external sources) fer:mass media. (4) Low levels of peasant literacy in developing countries. Rogers (1965, p. 217) found that the reason Why Colombian peasants were less able to obtain infOrmation from.the mass media received in their*homes was due to low levels of literacy. Interpersonal communication is extremely important in the dif- stion process in villages. A.few researChers (Rogers and van Es, 1964, Lionberger, 1963, and.War1and, 1963) have completed studies to investi- ‘ gate the dyadic relationships involved. Dyadic analysis is a kind of investigation of the networks of relationships among pairs of individ— uals. In this case, a pair is a.unit of analysis instead of an indi- vidual. Coleman stated (1958, p. 31) "Neither of the above kinds of analysis has required the use of sociometric—type data. An important kind of analysis whiCh does use suCh direct data on relationships is the analysis of pairs. . . . One of the most important problems WhiCh has been studied in this way is the similarity or difference in at- titudes or backgrounds between the two members of a pair." .According to publications available in the MSU Diffusion Documents Center, there are now about one thousand studies in the field of the difosion of innovations. About 37 per cent of the studies were done u outside of the U.S. Out of the 370 studies, none have investigated dyadic communication. Thus the present thesis will deal with peasant's dyadic interaction patterns in Colombian villages. In terms of the goal of interpersonal communication , friendship interaction and infermation-seeking interaction seem.to be two different and important types. Friendship interaction is a kind of interaction whereby associations among individuals are built. Information inter- action is a.kind of interaction whereby knowledge perceived as new to the individual is gained. FriendShip interaction is a part of everyday life, and almost everybody is engaged in it. Infermation—seeking interh action is important as innovations spread through a social system. Objectives The present thesis will be concerned with interaction patterns in Colombian villages as they relate to the process of the diffusion of innovations. The Objectives are: (1) To examine some determinants of homophily (dyadic similarity on selected Characteristics between people who interact) in interaction patterns by using three general kinds of variables: level of competence, communication contact, and status. ' (2) To examine the degree of'homophily in two kinds of communication situations: friendship interaction and infor- mation—seeking interaction. (3) To investigate the multi—step flow of communication in the diffusion of innovations in.terms of (a) number of 5 steps, and (b) discrepant mass media exposure scores between seeker and sought. CHAPTER II RATIONALE AND HYPOTHESES General Hypotheses Dealing with Homophily The first three general hypotheses are: General Hypothesis I: In_ the communication network in_ the diffusion o_f_ innovations , individuals with friendship rela— tionships have homophily pp the basis 93: level g competence , communication contact , and status . General Hypothesis II: _I_n_ the communication network a the diffusion g innovations , individuals with information-seeking relationships have homophily pp £119.. basis 51f level 91:: 00%- tence , communication contact , and status . General Hypothesis III: _Ip the communication network _ip t_he diffusion pf innovations, the degree pf homophily between _ip— dividuals with friendship relationships 13 gpeater than the de @e g homophily between individuals with information- seeking relationships . It is expected that individuals who interact are generally similar, whether it is a friendship relationship or an information relationship . There is an interdependence relationship between the homophily of in- dividuals on the basis of level of competence , communication contact , and status , and their degree of interaction . Zetterberg explained the inter- dependence relationship very clearly . According to Zetterberg (19 65 , 6 7 p. 73), "Thus, in an interdependent relation, a small increment in one variable results in a small increment in a second variable; then, the increment in the second variable makes possible a further increment in the first variable, WhiCh in turn affects the second one, and so this process goes on until no more increments are possible." So individuals interact with others like themselves; this in turn leads to greater similarity in certain social characteristics, whiCh leads in turn to further interaction, etc. Homophily of individuals Amount of who are in < > interaction interaction For some "static" variable like age, the interdependence rela- tionship does not hold. The possible rationale and empirical findings supporting the first two general hypotheses are summarized as fellows: First, Festinger's dissonance theory might support the first two hypotheses. Festinger (1958, p. 23) stated . . . (A) The existence of dissonance, being psyChologically uncomfortable, will motivate the person to try to reduce dissonance and aChieve consonance (or consistency). (B) When dissonance is present, in addition to trying to reduce it, the person.will actively avoid situations and irutmmations which would likely increase the dissonance. In friendship interaction, people with similar Characteristics would more likely be friends than people with different characteristics. Dissonance theory is mainly concerned with explaining ex_post facto behavior. In the present case dissonance theory is not the explanation of initiating a friendship between people, but the maintenance of 3 8 friendship between people.. The maintenance of friendShip assumes cogni- tive consistency or consonance between the two friends. It seems that individuals with very different Characteristics could interact once or twice, but probably would not over a considerable time. Does this notion also hold true fer the.rnfionmation—seeking situ- ations? When an innovation enters a village from.outside sources, disso- nance leads the individuals Who do not know about the innovation to seek information as a means to reduce this psychological discomfort . In order to maintain consonance, individuals would.avoid seeking infermation from those who would likely increase their dissonance. Thus, it seems that the establishment of infOrmation—seeking relationShips also assumes cog- nitive consonance between "seeker" and "sought."1 In order to maintain consonance, individuals would be similar in both friendShip relationShips and infbrmation—seeking relationShips. Heider (1958) discussed.how similarity can result in a balanced state in interpersonal relation. It is reasonable to assume the similarity between people with friendShip relationships and infbrmation—seeking relationships. warland (1963) found that individuals interact with those who have similar attitudes, similar'levels of competence, and similar socio— economic status. Katz and Lazarsfeld (1955) feund that the flow of influence is largely horizontal, and people on eaCh status level looked to their own corps of opinion leaders. The researChers did not use dyadic analysis. They identified four kinds of issues, namely: public affairs, marketing,4 movie-going, and faShions. Influence flow was feund to move across the lSeeker is the individual who initiates the interaction. Sought is the individual who gives infOrmation to the seeker in the interaction. ‘- 9 same age levels in marketing and movie-going, from slightly older to younger in public affairs , and from younger to older in fashions . Differential social status and innovativeness have been found to be barriers in the flow of communication (Rogers, 1960) . Wide dis- parity with respect to these two characteristics of people who interact can impede communication . Although the research to date has not investigated all the characteristics of individuals who interact in a communication network, it seems reasonable to assume that great dissimilarity in the character— istics of individuals who interact would be barriers to friendship inter- action and to information-seeking interaction . When great dissimilarity exists , individuals really speak different languages , and empathy with each other is more difficult . The previous reasoning could explain and support the first two hy- potheses . In explanation of the third general hypothesis, the following question could be raised; "Is there any difference in homophily between friendship interaction and information—seeking interaction?" It is hypothesized that the similarity of the individuals interacting in friend- ship relationships is greater than that of individuals interacting in information-seeking relationships . When talking about the characteristics of the individuals who interact, there are two kinds of hypotheses. One is the "like—me" hy- pothesis, that individuals will interact with people who are similar to themselves. Katz and Lazarsfeld (19148) and Merton (1957) supported the "like-me" hypothesis . The second hypothesis is that in the communication network , individuals being sought for information are more competent than 10 those who seek information. Lionberger (1953, 1957) supported this proposition. It is expected that in.informationeseeking interaction, the one sought would be higher in education level, more cosmOpolite, etc. than the seeker. This disparity is not expected to be great, however. Katz and Lazarsfeld (1955) feund that public affairs opinion leaders were better educated, had.higher status, and were more cosmopolitan in their mass media tastes than the followers. Roughly speaking, the seeker and the sought are still.much alike with respect to their Characteristics. It seems that in friendship interaction, the disparity is less than in infOrmation—seeking interaction. So it seems reasonable to assume the similarity of the individuals in friendship interaction is greater than in information-seeking interaction. In the situation of infOrmation-seeking individuals would like to interact with more competent people than themselves because in that case the goal of cormunication is to obtain information about innovations . liliritmmation-seeking interaction, people tend to interact with those Who are more reliable, legitimate, competent and teChnically accurate, but if the goal of the communication is mainly to socialize with eaCh other, or to establish and maintain friendships, individuals would be more likely to communicate with others who are more like themselves. Secondly, in information-seeking differences in education, cosmo- politeness, etc., relationShips establish a psychological inequality. The seeker must perceive a superiority in the sought in order to consider him.a useful and credible source. In friendship interaction, seekers and soughts wish to feel equal psyChologically; individuals therefbre interact 11 'with those who are as muCh like themselves as possible. Choice of Variables . UN Choice of variables to explain interpersonal commieation is cui'next consideration. Since the present analysis will deal with interaction patterns in the difosion of innovations, communication contact, and level of competence are certainly important variables. The variable status is selected because it indicates respect among villages in developing countries. It Should be understood that the Choice of variables are limited because the data used in the present thesis were not collected for the purpose of a dyadic interaction pattern analysis. The variables Chosen then, depend on their rele— vancy to the present thesis, as well as their availabilityu Figure 1 illustrates the interaction patterns hypothesized. General Hypotheses Dealing with the Multi-Step Flow of Communication In studying dyadic interaction, the universe is not the whole sample but the individuals who really interact. By studying dyads, the steps in information flow can be detected. The multi—step flow of come munication can then be investigated, since it is felt that the "two_step flow of communication" does not adequately explain the flow of innova- tions in peasant villages. The motion of the "two-step flow of communication" was ftmmulated by Lazarsfeld, Berelson, and Gaudet (1948) in the course of their analysis of the 1940 Presidential election. These investigators found that 'personal influence appeared to have been.more effective than the mass media in influencing voting decisions." In their study, the population Seeker Sought DI High C11 / 1 AF o_IB Variable 1 l ’ 1 Level of — - — - — _ — - _ _ _ - - _ - competence E F F1 > F1 lrl >Jrl Low Seeker Sought Variable 2 'AF2 rBF2 Communication — — — _ — _ _ _ _ _ - - _ _ _ contact H EF2 - >FF2 )L K12 / I2 Low IFZ 3>JE2 =======e>Friendship relationship ‘*““‘>lkfitmmation-seeking relationship IO-O'O-OIO‘IOQ-nfl Fig. l. Paradigm.of hypothesized relationships between seeker and sought in friendship and inferma— tion-seeking interaction. 13 was diChotomized into two categories: (1) Opinion leaders, and (2) fellowers. Opinion leaders act as bridges between mass media and followers. They claimed that "ideas often flow from radio to the opinion leaders, and frcnlthemlto the less active sections of the pop- ulation." These authors saw interaction between opinion leaders and followers as an intervening variable. Several shortcomings of the two-step flow hypothesis have be- come apparent. First, two steps are not enough. As Berlo (1960, p. 48) said "Communication is a.process, a dynamic process, with no beginning and no ending." So there are really multi—steps in the flow. Secondly, the population cannot be diChotomized into two categories. For'example, there is a group of individuals who never talk to others about a specific topic. The members in such a group are called "inactives" by Troldahl (1965). Troldahl found 75 per cent of the sample in his 1965 study in the "inactive" category. In addition to the assumption of the two—steps involved, there is another assumption whiCh states the opinion leaders are exposed more to the mass media than the fellowers. These will become two hypotheses for testing in the present thesis. The last two general hypotheses are: General Hypothesis IV: There are multiple steps ip_the_flow of; communication ip_the diffusion pf_innovations. General Hypothesis V: Soughts are exposed.more to_the mass media than the seekers. CHAPTER III METHODOLOGY _ Data—Gathering The data used in the present study grew out of the research project , "A Field Experiment of the Role of Opinion Leaders in Dif- fusing an Innovation in Three Colombian Neighborhoods . " The project was sponsored by Programa Interamericano de Informacicin Popular (PIIP) , Michigan State University , and Universidad Nacional de Colombia, Facultad de Sociologia, Bogota], Colombia . The project was conducted by Everett M. Rogers in 1963—64. Many of the same respondents were restudied by J. David Stanfield, Eduardo Ramos, and Elssy Bonilla de Ramos in 1965. Vereda and Municipio The data were gathered in three veredas in Colombia. A vereda is a neighborhood or a village. A vereda is a sub-unit of a municipio, defined as somewhat comparable to a county in the United States. General Description of the Three Veredas A general description of the veredas will help in interpreting later findings . Three aspects , the location, the transportation fa- cilities , and the social organization will be taken into consideration. 14 15 Location The three veredas are Pueblo Viejo, San Rafael, and Cuan Esquinas. Pueblo Viejo is in the municipio of Zipacc’m; San Rafael and Cuan Esquinas are in the municipio of Facatativa. They are located in the Departmento of Cundinamarca, near the center of Colombia, in the foothills of the Andes Mountains, on the edge of the plain surrounding Bogota, the capital of Colombia. The town of Facatativaf is the market place for each of the three villages. There is one—hour bus service from Bogota to Facatativa . Transportation In Pueblo Viejo and San Rafael, the transportation of farm products is by truck, mule or horseback; however, transportation by truck is quite limited in Cuatro Esquinas, especially in rainy weather. Pueblo Viejo has bus service three times a week; San Rafael has bus service on Sunday and Tuesday mornings . Cuatro Esquinas is with- out regular bus service. Social Organization Each village has an elementary school offering only the first two years of study. A program of the National Agricultural Extension Service has helped the peasants to organize a farmers ' council, a home economics club, and 4-H clubs for boys and girls. Method and Results of Data-Gathering The technique used in data-gathering was the structured personal interview. There were 192 eligible households, but only 160 out of 192 interviews were completed in 1963, and 136 were completed in 1965. 16 The distributions of the respondents in the three villages are shown in Table 1. Table 1. Results of data-gathering Results of data- No. of heads of house- No. of heads of'house- . gathering ‘ holds in 1963 ' holds in 1965 PV SR CE PV SR CE Household head was not a.farmEi~ 16 16 5 — - - Completed interviews 67 36 57 58 32 46 Incompleted inter— views 0 0 5 0 0 0 Unable to contact 6 4 8 9 4 11 Refusals _u _€>_ _2 .2 _2 _2 Total 93 ,61 V _ 75 67 36 57 Statistical Methods The present thesis is primarily concerned with two kinds of inter— action: friendShip interaction and information-seeking interaction. The interaction dyadic pairs were obtained from.the questionnaires in 1963 and 1965. The dyadic pairs of friendship interaction were constructed by combining the individuals specified by name. For example, if.A.names B as his friend, A.and B are treated as a pair for purposes of analysis. The 1965 sociometric question.was: "Do you.meet with your friends on Sunday and holidays?" No Yes What are the names of friends with.whom you most frequently meet? 17 (Use identification numbers) a) b) c) There were 56 friendship pairs in PV; 32 pairs in SR; and 29 pairs in CE. By the same procedure , the information-seeking dyads were obtained from a sociometric question in the 1963 questionnaire: "Have you talked with another farmer about agriculture in the last two months?" If yes, the next question was "with whom?" 1) ' 3) 2) 4) There are 68 seeker-sought pairs with information—seeking inter- action in PV; 27 in SR; and 69 in CE. The three villages are treated separately because the norms in the villages are different and by so doing, three replications can be obtained. "Following a survey of several villages, two were selected, one with traditional norms , Ouatro Esquinas, and one with relatively modern norms, Pueblo Viejo. . . . A third com- munity was chosen, San Rafael, which had relatively modern or non— traditional norms; . . . ." (Stanfield and Ramos, 1965). Differing village norms might cause the results to vary from village to village. Zero-order Pearsonian correlation will be used to measure the horophily (of the people who interact) for each variable . The reasons for using zero—order Pearsonian correlation are: (l) The variables are continuous . (2) Complete homophily is indicated by a coefficient of correlation of +1 . 18 (3) Complete non-hoIophily is indicated by a coefficient of correlation of —l. The null hypotheses in symbolic form for the first two general hypotheses are: 0 : rFé 0 H0 : rIé 0 The theoretical hypotheses in symbolic form for the first two . general hypotheses are: H1: rF>0 H1: rI>0 For General Hypothesis IV the transformation technique from r to Z will be used in order to test the difference between r]? (for friend- ship interaction) and r1 (for informat ion—seeking interaction). The formula is: \/ _L+ _1_. NF—3 NI-3 A Z table provides the appropriate probability of the Z scores . The null hypothesis and theoretical hypothesis for differences, re- spectively, in symbolic form, are as follows: H 0 ZF=ZI H1 " ZF>ZI When using Z transformations , independence of the two zero-order Pearsonian correlations must be assumed. The assumption to be made in the present case is that choice of a friend and an information source 19 are independent . The multi-step flow of communication is analyzed by graphing the steps of information flow in the cliques delineated by Stanfield and Ramos. For General Hypothesis IV observation of the figures for those cliques confirms this hypothesis. For General Hypothesis V chi square will be used. Operationalization of Variables Ependent Variable The major dependent variable of the present thesis is interaction . Interaction is defined as any interpersonal communication between two people with a purpose of seeking friendship and/or information. The bases of interaction will be investigated for three variables: level of con- petence , communication contact , and status . The relationship between homophily with respect to these variables , and interaction is shom in Figure 2 . Determinant Variables The first general category of variables is level of competence. Literacy and innovativeness are the two operational variables of this category. Literacy is measured by the number of underlined words read correctly in the sentence: "E1 hombre movio la manc’) rapidamente en 915121111“. de respeto."l According to Rogers (1960, p. 20) "Innovativeness is the degree to which an individual is relatively earlier in adopting new ideas than the other members of his social system." Innovativeness is measured by agricultural innovativeness scores , which are the summation of sten scores for time of adoption of twelve l‘I'he English translation is "The E moved his arm rapidly in a . gesture of respe ." 20 .Eafimapfi as... hangs; no magmaoflflfl mfi no assume - .N .mfl mBMPm HMHOOW 81.5.83 mEUHoom ow< 18.."ng toflboaoflhmo 300m wmmoovfloogfioo 8&8 Enema won: 205% I EHEozom mmméwpmxrog noun Emsmpfi mflmocmflfim /%U§Hmu.fl l Pomuboo no.3 859%0 no HmSoq . 21 agricultural innovations . The agricultural practices are: 1. 2. 10. 11. 12. 13. 14. 15. Insecticida (aldrin) - Insecticide Fungicida manzate (papa) - Potato fungicide Abono quiJ/nico - Chemical fertilizer Matamaleza - 2 , 4-D weed—killer Pumigadora (or aspersora) - Hand-sprayer Alimentos concentrados (Finca, Purina) - Feed concentrate Vacuna para cobra aviar - Chicken disease innoculation Vacuna contra carbon bacteridiano y carbon sintoma’tico para vacas - Black leg vaccination for cattle Vacuna contra aftosa para vacas - Vaccination for hoof and mouth disease Desinfeccion de tierra - Soil disinfection Usa tractor - Use of tractor Hortalizas - Home garden Diaco Narino (trigo) - New wheat variety Parda Pastusa (papa) - New potato variety Funza - New barley variety The higher the score , the more innovative the peasant is in adopting new farm ideas . Communication contact , the second category of variables , was measured by mass media exposure, cosmopoliteness, and social participation. A measure of mass media exposure includes exposure to radio, newspapers , magazines, movies, and TV. Seven questions (see questions 34, 356, 350a, 37d, 39b, 40a in Appendix A) tap this dimension. An individual's score on this variable is determined by summing the re- sponse categories for all seven questions after they are converted to 22 sten scores. The range of the scores is from 23 to 80. The higher the score, the more the peasant is exposed to mass media. Cosmopoliteness is defined by Rogers (1960, p. 102) as "the degree to which an individual's orientation is external to a particular social system." Cosmopoliteness is measured by the number of trips to Bogota per year. The range of the scores is from.0 to 99 or:more. The higher the score, the more cosmopolite the peasant. Social participation is measured by the formal participation score. The peasant (and.his family) are awarded.points for participa— tion in the following organizations: (1) community board; (2) Catholic Action; (3) night worship; (4) Christ brotherhood; (5) Chicken farmers' committee; (6) housewives club; (7) 4-S clubs; and (8) others. TWO points were awarded for each membership in an organization, except Catholic Action1 whiCh received one point; and one point was awarded fOr purchasing at the village cooperative. These points are then summed to obtain the formal participation score. The range of the scores is from 0 to 9. The higher the score, the greater the peasant's social participation. The third category of variables is status. Status is indexed by age and social status within each community. Social status is determined by an interviewer's rating of each peasant on the basis of the peasants' house, clothing, and.material possessions in comparison with other peas- ants in the community. The scores range from 0 to 4. The higher the score, the higher the respondent's social status, according to the interviewer's judgment. 1There is less chance for exposure to new ideas in a primarily religious group which is based on tradition. 23 Empirical Hypothesis General Hypothesis I: _In_ the communication network 1_n_ the diffusion g innovations , individuals with friendship relationships have horophily pn_ ph_e_ basis pf_ level p£ corpetence , ‘cormunication con- tact , and status . Empirical Hypothesis Ia: _In_ the Communication network E the diffusion pi innovations , individuals with friendship relat iOnships have homophily pp E basis p_f_ functional literacy. Empirical Hypothesis Ib: _Ip the communication network _13_ t_hp diffusion g innovations , individuals with friendship relationships have horophily @ E13 basis g the degree pg: innovativeness. Empirical Hypothesis Ic: E the communication network _i__1_1_ the diffusion p_f_ innovations , individuals with friendship relationships have homophily pp the basis p_f_ the degree pf_ mass media exposure. Empirical Hypothesis Id: _I_n_ the communication network _12 the diffusion pf innovations , individuals with friendship relationships have homophily pp Q13 basis p_f_ frequency p: trig tp cities. Empirical Hypothesis Ie: E the commUnication network 3'. the diffusion o_f_ innovations , individuals with friendship relationshipp have horophily o_n Elle basis pf the degree g social participation. Empirical Hypothesis If: II; the communication network 1_n_ _thp diffusion p1: innovations , individuals with friendship relationShips have him 'ly mega-easier. Empirical Hypothesis .Ig: _Ip the communication network 2 'Ee diff us ion p: innovations , individuals with friendship relationships have horophily pp the basis p: social status. 24 General Hypothesis II: _Ip the Communication network _i_n_ the diffusion pi innovations , individuals with information—seekipg relation- ships have horophily pp fie basis pi level pi competence, cormunication contact , and status . Empirical Hypothesis Ila: _Ip the communication network _ip the diffusion o_f_ innovations , individuals with infCImation-Seekipg relation— ships: have homophily o_n tfi basis g function literacy. Empirical Hypothesis IIb: _Ip the communication network 2 the diffusion g innovations , individuals with information-seeking relation- shipp have hompphily g1 fie basis g the degree pi innovativeness . EIpirical Hypothesis IIc: E the communication network ip the diffusion SE innovations , individuals with information-seeking relation- ships have homophily @ _tpe basis g the degree pi mass media exposure. Expirical Hypothesis IId: _Ip the communication network i_n_ ihe diffusion pi innovations , individuals with information-seeking relation- ships have homophily g1 pie basis o_f_ frequency pi trips t_o cities . Erpirical Hypothesis IIe: E the communication network pi the diffusion pi innovations , individuals with information-seeking relation- ships have homophily pp t_he basis g the degree pi social participation. Empirical Hypothesis IIf : E the communication network ip the diffusion g innovations , individuals with information-seeking relation- ships have horophily pp i_he basis pi age. Empirical Hypothesis Hg: in_ the communication network _ip the diffusion g innovations , individuals with information-seeking relation- ships have homophily pp ihe basis pi social status. 25 General Hypothesis III: E the communiCation network pi E diffusion pi innovations , the degree g hOmophily between individuals with friendship interaction i_s greater than the ”degree pi homophily between individuals with i_nformation-seeldng interaction . Empirical Hypothesis IIIa: E the commrnication network o_f_ the diffusion g innoVations , individuals engaged ' E friendship interaction have more horophily o_n_ ihe basis p_f_ functional literacy than the indi- viduals engaged E information-seeking interaction. EIpirical Hypothesis IIIb: E the communiCation network E the diffusion pi innovations , individuals 'epgaged E friendship interaction have more homophily o_n the basis pi the degpee "pi innoVativeness than individuals engaged E information-seeking interaction . Erpirical Hypothesis IIIc: E the cormunication network E _tlE diffusion E innovations , individuals engaged E friendship interact ion have more homophily pp Ee basis pi the degree pi mass media exposure than individuals engaged E informat ion-seeking interaction . Empirical Hypothesis IIId: E the communication network E _tpe diffusion pi innovations , individuals engaged E friendship interaction have more horophily o_n the basis pi the degree pi cosmopoliteness than individuals engaged E information-seeking interaction . Erpirical Hypothesis IIIe: E the communication network E the diffus ion _CE innovations , individuals engaged E friendship interaction have more horophiiy: pp Ee basis pi the degree pi social participation than individuals engaged E ‘information-Seekipg interaction. 26 Empirical Hypothesis IIIf: _I_n_ the communication network E the diffusion pi innovations , individuals engaged E friendship interaction have more homophily pp the basis pi pge than individuals engaged E E- formation—seeking interaction. Empirical Hypothesis IIIg: E the communication network E the diffusion pi innovations , individuals engaged E friendshE interaction have more hompphily pp the basis p_f_ social status than individuals _ep— . gagd E information—seeking interaction. Empirical Hypothesis IV: There are multiple steps E the flow pi communication E the diffusion pi innovations. Empirical Hypothesis V: Soughts are exposed more t_o the mass media than seekers . CHAPTER IV FINDINGS Test of Hypotheses Dealing with Homophily General Hypothesis I General Hypothesis 1: E the commurication network E the diffusion pi innovations , individuals with friendship relationships have homopIily pp the basis pi level pi competence communication, contact , and status . Empirical Hypothesis Ia: E the communication network E the diffusion pi innovations , individuals with friendship relationships have homophily o_n the basis pi functional literacy . The zero-order Pearsonian correlation for PV is .058, which is not significantly different from zero; for SR it is 0.21 which is not significantly different from zero; for CE it is .042, which is not significantly different from zero. Empirical Hypothesis la is not confirmed. Empirical Hypothesis Ib: E the communication network E the diffusion g innovations , individuals with friendslip relationships have homophiiy pp the basis pi the degree pi innovativeness . The product moment correlation for PV is . 304, which is sigrificantly . greater than zero. For SR it is .388, which is significantly different from zero. For CE it is -.087, which is not significantly different from zero. Empirical Hypothesis Ib is confirmed. 27 28 Empirical Hypothesis Ic: E the communication network E the diffusion E innovations , individuals with friendship relationshiLs have homophily pp _tpe basis pi the degree pi mass media exposure. The zero-order Pearsonian correlation for PV is .064, which is not signifi— cantly different from zero; for SR it is .767, which is sigrificantly different from zero; for CE it is .310, which is significantly dif- ferent from zero. Empirical Hypothesis Ic is confirmed. Erpirical Hypothesis Id: E the communication network E Ee diffusion pi innovations , individuals with friendShip ‘relationsl'ipp have homophily o_n Ee basis pi frequency E trips _tp cities . The zero— order Pearsonian correlation for PV is .063, which is not significantly different from zero; for SR it is .544, which is significantly dif- ferent from zero; for CE it is .336, which is sigrificantly different from zero. Empirical Hypothesis Id is confirmed. Empirical Hypothesis Ie: E the communication network E the dif- fusion pi innovations , individuals with friendship relationships have horophily on the basis of the degee g social Erticipation. The zero- order Pearsonian correlation for PV is -. 034, which is not significantly different from zero; for SR it is .285, which is not significantly dif- ferent from zero; for CE it is .598, which is significantly different from zero at the one per cent level. Erpirical Hypothesis Ie is not confirmed. Erpirical Hypothesis If: E the cormumication network E _1_:h_e iii- fusion g innovations , individuals with friendship relationships have _hp- is .075, which is not sigrificantly different from zero; for SR it is -.039, which is not significantly different from zero; for CE it is .267, which is not 29 significantly different from zero. Erpirical Hypothesis If is not confirmed. Empirical Hypothesis _Ig: E the communication network E the diffusion g innovations , individuals with friendship relationships have homophily pp the basis pi social status. The zero—order Pearsonian correlation for PV is .295, which is significantly different from zero; for SR it is .173, which is not significantly different from zero; for CE it is . 346 , which is significantly different from zero. Empirical Hypothesis _Ig is confirmed. Since four Empirical Hypotheses are confirmed, General Hy- pothesis I is confirmed. General Hypothesis II General Hypothesis II: E the communication network E the diffusion pi innovations , individuals with information-seeking relation- ships have homophily E the basis g level pi comEtence , communication contact , and status . Erpirical Hypothesis IIa: E the communication network E the diffusion pi innovations , individuals with information-seem relation- ships have homophily pp the basis pi functional literapy. The zero- order Pearsonian correlation for PV is .308, which is significantly dif— ferent from zero; for SR it is +.285, which is not significantly dif- ferent from zero; for CE it is .103, which is not significantly differ- ent from zero. Empirical Hypothesis 11a is not confirmed. Empirical Hypothesis IIb: E the communication network E ihp diffusion o_f_ innovations , individuals with information-seeking relation- ships have homophily o_n _tE basis SE. the degree pi innovativeness . The 30 zero-order Pearsonian correlation for PV is .088 ,9 which is not sig- nificantly different from zero; for SR it is -.259, which is not sig— nificantly different from zero; for CE it is -393, which is significantly different from zero. Empirical Hypothesis IIb is not confirmed. Erpirical Hypothesis 110: E the communication network E E3 diffusion o_f_ innovations , individuals with information-seeking relation- ships have homophily o_n E basis pi the degee o_f_ mass media epxposure. The zero—order Pearsonian correlation for PV is .263, which is signifi— cantly different from zero; for SR it is .274, which is not significantly different from zero; for CE it is .223, which is significantly different from zero. Empirical Hypothesis He is confirmed. Empirical Hypothesis IId: E the communication network E _tlE diffusion pi innovations , individuals with information—seeking relation- ships have homophily pp Es basis pi frequency g trips ip cities. The zero-order Pearsonian correlation for PV is . 090 , which is not signifi— cantly different from zero; for SR it is .219, which is not significantly different from zero; for CE it is .136, which is not significantly dif- ferent from zero. Empirical Hypothesis IId is not confirmed. Empirical Hypothesis IIe: E the Communication network pi the diffusion g innovations , individuals with information-seekipg relation— ships have hom0phi1y _o_p _tpp basis g the degree g social participation. The zero—order Pearsonian correlation for PV is .191, which is not sig- nificantly different from zero; for SR it is .653, which is significantly different from zero at the one per cent level; for CE it is .364, which is significantly different from zero. Empirical Hypothesis He is confirmed. Empirical Hypothesis IIf: E the communication netwOrk E the diffusion pi innovations , individuals with infonmation-Seeking 31 relationships have homophily pp ihp basis pi age. The zero-order Pearsonian correlation for PV is .014, which is not significantly dif- ferent from zero; for SR it is .221, which is not significantly dif- ferent from zero; for CE it is -.l28, which is not significantly different from zero. Empirical Hypothesis IIf is not confirmed. Erpirical Hypothesis IIg: E the communication network E Ep diffusion pi innovations , individuals with information—seeking rela— tionships have homophiiy E the basis pi social status. The zero- order Pearsonian correlation for PV is .132, which is not significantly different from zero; for SR. it is -.160, which is not significantly different from zero; for CE it is —.080, which is not significantly dif— ferent from zero. Empirical Hypothesis Hg is not confirmed. Since just two of the empirical hypotheses are confirmed, General Hypothesis II is not confirmed . General Hypothesis III General Hypothesis III: E the communication network o_f_ _tlE diffusion g innovations , the degree pi homophily between individuals with friendship interaction _ip gieater than the degree pi homoEily between individuals with information-seeking interaction . Empirical Hypothesis IIIa: E the communication network pi Ep diffusion pi innovations , individuals engagep- E friendship interaction have more homophily pp the basis pi functional literapz than the indi— viduals engaged E information—seeking interaction . The Z score for PV is -1.506, which is not significantly different from zero and it is in the direction opposite to that which was predicted; for SR it is 1.100, which is not significantly different from zero; for CE it is .316, 32 WhiCh is not significantly different from.zero. Thus, Empirical Hypothesis IIIa is not confirmed. Empirical Hypothesis IIIb: ip_the communication network 29:322. diffusion pi innovations, individuals engaged ip friendship interaction have more homophiiy_pp_ipp_basis piithe degree pi innovativeness than ipdividuals engaged ipLinfOnmation-seeking interaction. The Z score for PV is 2.217, whidh is significantly different from zero in the predicted direction; for SR it is 2.421, whiCh is significantly dif- ferent from zero in the predicted direction; for CE it is.-2.705, which is significantly different from.zero in the direction opposite to that WhiCh was predicted. iEmpirical Hypothesis IIIb is confirmetL Empirical Hypothesis IIIc: ip;the communication network ip_the difosion. i innovation , individuals engaged ip_friendship interaction have more homophily pp ipp_basis pi the degree pi mass media_p§posure than individuals engaged ip_infbrmation-seekipg_interaction. The Z .score fbr PV is -1.144, whiCh is not significantly different from zero, and it is in the direction opposite to that which was predicted; for SR it is 2.654, whiCh is significantly different from.zero and in the predicted direction; for CE it is 0.511, whiCh is not significantly dif- ferent from zero. Empirical Hypothesis IIIc is not confirmed. Empirical Hypothesis IIId: ithhe communication network ip_the diffusion pilinnovations, individuals engaged ip_friend3hip interaction have more homophily pp the basis pi the degree pi cosmppolitenessthan ‘individuals engaged'ip;informationéseeking’interaction. The Z score fOr PV is -.167, whiCh is not significantly different from zero, and it is in the direction opposite to that whiCh.was predicted; for SR it is 1.357, whiCh is not significantly different from zero but in the expected 33 direction; for CE it is 1.121, Which is not significantly different from.zero but in the expected direction. Empirical Hypothesis IIId is not confirmed. Empirical Hypothesis IIIe: ip_the communication network ip_the diffusion pi innovations, individuals engaged ip_friendship interaction have more homophilypp_the basis pi the degree pi social participation than individuals engaged ip_information-seeking interaction. The Z score for PV is —1.233, which is not significantly different from zero, and in the direction opposite to that which was predicted; fdr SR it is -1.700, which is significantly different from zero but in the direction opposite to that which was predicted; for CE it is 1.663, which is significantly different fromlzero and in the predicted direction. Empirical Hypothesis IIIe is not confirmed. Empirical Hypothesis IIIf: ip_the communication network ip_the diffusion pi innovations, individuals engaged.ip friendship interaction have more homophily pp the basis pi age than individuals engaged.ip_ip7 formation—seekinginteraction. The Z score for PV is .389, which is not significantly different from zero, but in the expected direction; for SR it is -0.943, which is not significantly different from.zero, but in the Opposite direction; for CE it is 2.147, which is significantly different from zero and in the predicted direction. Empirical Hypothesis IIIf is not confirmed. Empirical Hypothesis IIIg: ip_the communication network ip_ipp_ diffusion pi innovations, individuals engaged ip_friendship_interaction have more homophily pp the basis pi social status than individuals pp: ' gpged ip_information—seeking interaction. The Z score for PV is 0.989, which is not significantly different from zero, but in the predicted 34 direction; for SR it is 1.189,.Which is not significantly different from zero, but in the predicted direction; for CE it is 2.347,-which is signif- icantly different from.zero and in the predicted direction. Empirical Hypothesis IIIg is not confirmed. Since just one of the empirical hypotheses is confirmed, General Hypothesis III was not confirmed. Summary of Hypotheses I, II, and III One of the general hypotheses is confirmed, while two are not. The zero—order Pearsonian correlations are summarized in Table 2. 0f the fbrty—two zero-order Pearsonian correlations, nine are not sig— nificantly different from.zero at the 5 per cent level and negative, thirty-three are positive, and fifteen out of the thirtyethree are sig- nificantly different from zero at the 5 per cent level. The Z statistic scores of comparing two zero-order Pearsonian correlations are summarized in Table 3. Of the twenty-one Z statistic scores, fOurteen of twenty-one are in.the expected direction, and four of the fourteen are significantly different from.zero; seven of twenty— one are in the opposite direction, and two of the seven are significantly different from zero. .Hm>ma #ooo 9mg H one Hm onoN_aonm #omLMMMHp mapoMOHwHowwm+ .Ho>mH Pomo pom m one um 09mm 809% econoMMHp mHvomoHMMcmwmv. omo.- «mjm. omH.- msa. «ma. «mom. msPMpm awnoom .s mNH.- new. Hum. mmo.- :Ho. who. mm< .o «jam. +mmm. +mmm. mam. Hma. :mo.- conamanonyama Hmnoom .m mmH. «mmm. mHN. «jam. omo. moo. mmmamyHHOQQEmoo .: «mum. «cam. ska. +sms. «mom. :mo. mmsmoaxm macms mam: .m +wmm. smo.- mmm.- «mmm. mmo.- «ism. mmmcm>flpm>oqu .N men. was. mmm.- Has. «Hem. mmo. sumgmpfla .H COHPOEG Ammuzv Ammnzo Aswnzo Ammnzv Ammuzo “omnzo sameness; mo moflxmom ooapompopofl moflxomw soapomnmpofl moflxmmm coapomhopofl wpomoflanopmu conpmagomaH anamacmngm aonpmapoecH anamecmnmm conpmepoccH angmuamnmm mu Mm >m .mpm>p mcwxommICOEP Imsmowofl pom cowpomnmpow mfiamocmflnw co moflvmflnmpomnmno mo mCOMPMnghoo CMHoomnmom hmeOIOQoN mLH .m manme 36 Table 3. Z scores for differences between the infOrmation—seeking friendship homophily correlations for three Colombian villages. Villages PV SR CE FUnctional literacy -l.506 1.100 .316 Innovativeness 2.217* 2.421* —2.705* Mass media exposure -1.l44 2.654* .511 Cosmopoliteness - .167 1.357 1.121 Social participation —l.233 —l.700* 1.663* Age .389 — .943 2.147* Social status .989 1.189 2.347* * Significantly different from zero at the 5 per cent level. The Z statistic fOr comparing two independent zero-order Pearsonian correlations is: _l_._l_ NF-3 NI—3 Test of Hypotheses Dealing with the Multi-Step Flow of Communication In the present section, two hypotheses about the two-step flow will be investigated. General Hypothesis IV General Hypothesis IV: There are multiple-steps ip the flow pi; communication ip_the network pi diffusion pi innovation. 37 The cliques on this page were obtained through the visualization.method by David Stanfield and Elssy Ramos. The procedures are as fo1lows: (1) They constructed a sociogram with arrows drawn on an actual map of the community. (2) The construction of the map was based on the question regarding information-seeking mentioned previously . (3) On the basis of maps and arrows, the cliques were con- structed. Clique l Clique 2 Clique 3 004 007 073 008 023 076 012 024 -078 014 025 -079 017 027 ' 081 071 030 083 125 032 084 0351 088 037 094 038 098 040 .123 042 127 066 105 Figures 3,4, and 5 show the steps of information flow. The data show the information flow through interpersonal communication, the inter- action among opinion leaders and followers . In Figure 3 , the opinion leaders are not necessarily the infor- mation givers , and the peasants in the clique do not necessarily go to the opinion leaders for information . The data suggest that there are four kinds of information flow: (1) Information flow among opinion leadersl and followers , e.g., the information flow among 008, 004, 017. lOpinion leaders were determined by the sociometric choices they received. (Stanfield and Ramos, 1965) 38 Fig. 3. Steps of infOrmation flow among seekers and soughts in Clique l. NOte: The underlined numbers are opinion leaders. 40 Fig. 5. Steps of information flow among opinion leaders and fOllowers in Clique 3. Note: The underlined numbers are opinion leaders; /-_7 indicates individuals not in the clique. Lu (2) InfOrmation flow from.non—opinion leaders to opinion leaders, e.g., the step between 008 and 125. (3) Information flow from opinion leaders to followers , e.g., the flow from.008 to 014. (4) llfitmmation flow between followers, e.g., the information flow between 012 and 014. The data show that: 1. The infOrmation flow is not necessarily from Opinion leaders to fOllowers. 2. There is a.multi-step flow of infOrmation. In Figure 4, three kinds of information flow exist: (1) Opinion leader goes to non-Opinion leaders fOr information, e.g., 023 goes to 032, 073 and 098. The latter two are not in the clique. (2) NOn-opinion leaders go to opinion leaders fOr infOrmation, e.g., 027 goes to 023. (3) Non-opinion leaders go to non—opinion leaders, e.g., 040 _ goes to 035. The data show again, the information flow is not always from opinion leaders to fOllowers, and there are multi—step flow. In Figure 5, there are three kinds Of infOrmation flow. (1) Followers go to opinion leaders for information, e.g. , 127 goes to 073, and 123 goes to 073. (2) Opinion leaders seek infOrmation from other Opinion leaders, e.g., 083 goes to 073, and 098 goes to 073. (3) There is infOrmation flow among non—opinion leaders, e.g., 078 goes to 084. In this case, the data provide evidence fOr the multi-step flow as did the data of the previous two figures. So General Hypothesis IV 42 is confirmed. General Hypothesis V General Hypothesis V: Soughts are epiposed more ip the mass media than seekers . Figures 4, 5, and 6 in the following three pages show discrepan— cies in mass media exposure scores between seekers and soughts in the chain relationship of information flow. Table 4 indicates that the amount of mass media exposure scores of seekers differ from the mass media exposure scores of soughts, and the numbers of first, second and third steps. Table 4. Differences in seeker—sought mass media exposure scores by steps in the multi-step flow of commotion. Seeker— sought differences in Clique 1 Clique 2 Clique mass media Steps Steps ‘ Steps Total exposure lst 2nd 3rd lst 2nd lst 2nd 3rd steps Lower 20" 1 1 1 6 2 3 1 1 " 10+ 9 0 2 0 0 0 4 l 0 0— 9 1 0 0 1 1 1 0 0 27 Equal 0 1 2 1 0 0 0 0 4 High 0— 9 2 0 0 1 0 1 0 1 " 10:19 0 0 0 0 2 0 0 0 " 20 0 1 0 0 2 0 0 0 10 Totals 4 5 3 9 7 9 2 9 2 41 Chi—square is 7.81, which is significant at the one per cent level. So General Hypothesis V is confirmed. High Mass exposure Seeker Fig- 80 75 70 65 t 60 55 50 45 40 35 30 25 20 6. 1 4 1 J J J. b 07/ on! 43 Sought l 0/ 042 .20 0 I2— Sought 2 Sought 3 3%.; 43‘ 0,74 _‘P’Oy Mass media exposure scores of seekers and soughts in linked information flow in Clique l. Mass media Seeker 75 70 65 60 55 50 45 40 35 30 ”(can 25 20 Fig. 7. 1m Sought 1 fat 35" Sought 2 7.! .13 02! #37 531 Mass media exposure scores of seekers and sought in linked infOrmation flow in Clique 2. Seeker High Mass media exposure Fig- 801 75 25.. 20.. 8. I r Sought 2 73 013 Sought 3 073 71’ Mass media exposure scores of seekers and soughts in linked information flow in Clique 3. CHAPTER V SUMMARY, DIT'ERPRETATION , D’IPLTCATION , AND SUGGESTIONS Summary The present thesis is concerned with interaction patterns and the multi—step flow of communication. The objectives were (1) to ex— amine some determinants of homophily (such as level of competence, communication contact, and status) in two kinds of interaction: friendship and information—seeking interaction, and (2) to investi— gate the multi-step flow of communication in terms of the number of steps and differences in mass media exposure between seeker and sought. The data used in the present study grew out of the research project, “A.Field Experiment of the Role of Opinion.1eaders in Dif— sting an Innovation in Three Cblombian Neighborhoods." The project was directed by Everett M. Rogers in 1963, and.many of the same re- spondents were restudied by J. David Stanfield, Eduardo Ramos, and Elssy Bonilla de Ramos with approximately the same respondents in 1965. There were 160 interviews completed in 1963: 67 in Pueblo Viejo; 36 in San Rafael; 57 in CUatro Esquinas. One Hundred and thirty-six interviews were completed in 1965: 58 in PV; 32 in SR; and 46 in CE. HOmophily is the degree of similarity in selected characteris— tics between individuals in infOrmation-seeking interaction or friend- ship interaction. Dyadic analysis is a kind of investigation of 46 47 the networks of relationships among pairs of individuals. There were five hypotheses in the present thesis. The first two hypotheses dealt with the degree of homophily of individuals engaged in friendship interaction and in information—seeking interaction, re- spectively. The third hypothesis dealt with differences in homophily of individuals engaged in friendship interaction and information- seeking interaction. The fOurth hypothesis dealt with the number of steps in the flow of communication. The fifth hypothesis dealt with differences on mass media exposure scores between seeker and sought. The first, fourth, and fifth hypotheses were confirmed. The second and the third hypotheses were not. For the first two hypotheses, the findings agreed with the predicted position: (1) none of the nine negative zero—order Pearsonian correlations (indicating degree of homo- phily in dyadic interaction) were significantly different from zero, (2) thirty-three zero-order Pearsonian correlations were in the pre- dicted direction; and (3) fifteen of these thirty—three positive zero- order Pearsonian correlations were significantly different from zero. Seven out of twenty-one Z scores (which tested the difference between two homophily indexes) were negative, and two of the seven.were significantly different from.zero; fourteen out of twenty—one Z scores were in the predicted direction, and six of these fourteen.were signif- icantly different from zero. Thus, Hypothesis III was not confirmed. The failure to establish expected relationships in all cases in Hypotheses I, II, and III might be due to the following reasons: (1) the present thesis was exploratory in nature and there was little past re— search to guide its design; (2) the data were not gathered particularly 48 fOr the present study; (3) the variables were not crucial predictors in that the percentages of the variance explained were generally low. The fOurth and the fifth hypotheses were confirmed; they indi- cated that there are multiple steps in the flow of communication, and that the sought is exposed more to the mass media than the seeker. Interpretation The findings of the first three hypotheses will be explained in more detail. The norms of the villages may affect the results. SO the following interpretations are mainly based on differences in the village norms. General Hypothesis I General Hypothesis I: ip_the communication network.ip_the diffusion pi innovations, individuals with friendship relationships have homophily on the basis of level of competence,communication contact, and status. Empirical Hypothesis Ia was not confirmed. Functional literacy is not a determinant Of friendship interaction among villagers in the three villages. Generally speaking, functional literacy was not a very important determinant of seeker-sought homophily. The individuals in the villages may not know how literate their neighbors are, so literacy is not important in selecting a friend. Empirical Hypothesis Ib was confirmed. Innovativeness is a predictor of friendship interaction. The zero-order Pearsonian correla- tions in PV and SR.were significantly different from zero, and the cor- relation was low and not significantly different from zero in CE. In L19 villages with more modern.norm1(like PV), innovativeness is a better predictor of friendship interaction. In these villages with.modern norms, the individuals may knOW'hOW innovative their neighbors are. Thus, innovativeness is a criteria of choosing their friends. Empirical Hypothesis Ic, dealing with.mass media.exposure, was confirmed in SR.and CE, but the zero-order Pearsonian correlation in PV was low and not significantly different from.zero. The correlation was extremely high in SR (.767). It seems that mass media exposure is a better predictor of friendship interaction in more traditional villages. Mass media exposure might be a substitute for cosmopoliteness, which is rare in more traditional villages. Empirical Hypothesis Id, dealing with cosmopoliteness as a basis fOr friendship homophily was confirmed. The zero—order Pearsonian correlation.was relatively low in PV. Cosmopoliteness seems to be a better predictor of friendship interaction in.more traditional villages. Empirical Hypothesis Ie, dealing with social participation, was not confirmed . The zero-order Pearsonian correlation in (E was rela- tively high and significantly different from zero, and it was rather high but not significantly different from.zero in SR. The homophily correlation was low and negative in PV. Social participation seems to be a better predictor of friendShip homophily in more traditional villages. Empirical Hypothesis If, dealing with age, was not confinmai. Age is not an important predictor of friendhsip interaction. This result is different from Warland's (1963) finding among Iowa farmers, that people who interact are Of the same_age. The explanation for 50 the present findings might be: (1) the rather crude measure (by decades of age; and (2) the villagers in the sample are all adults, and hence there is relatively less variance in their ages. Empirical Hypothesis Ig was not confirmed. Generally speaking, social status is a relatively ineffective predictor in both modern and traditional villages. In summary, it was found that: (l) innovativeness, mass media exposure, cosmopoliteness, and social status are relatively effective predictors of friendship interaction; (2) there is a tendency in more modern villages fOr mass media exposure and cosmopoliteness to be poor predictors of interaction among villagers. General Hypothesis II General Hypothesis II: ithhe communication network ip_the diffusion pi innovations, individuals with information-seeking relation- ships have homophily pp the basis pi level pi competence, commppication contact, and status. Empirical Hypothesis IIa.was not confirmed. The zero—order Pearsonian correlation for functional literacy as a prediction of in- formation—seeking in PV is significantly different from zero, but it is the only significant correlation out of the three correlations for in— formation-seeking interactions. Earlier, it was pointed out that none of the three correlations for functional literacy as a predictor of friendship interaction were significantly different from zero. In a more modern village, functional literacy may be a more important de- terminant of homophily in information-seeking interaction. The finding is similar to Troldahl's (1965) and Warland's (1963) finding that 51 seekers and soughts were similar in formal educational level. Empirical Hypothesis IIb was not confirmed. The only significant zero—order Pearsonian correlation is in CE (. 398) . In a more traditional village , innovativeness may be a more important determinant of informa- tion—seeking interaction. Conversely , in friendship interaction in more modern villages , innovativeness may be a more important determinant of interaction. Empirical Hypothesis IIc was confirmed. It indicates that mass media exposure is a determinant of information—seeking interaction in the three villages . The only low zero—order Pearsonian correlation (.064) in PV in friendship interaction might suggest that mass media exposure is not as important a determinant of information-seeking inter- action in a more modern village. Empirical Hypothesis IId was not confirmed, indicating that cosmo— politeness is not a determinant of information—seeking interaction. The measurement of cosmopoliteness (the number of trips to Bogota) did not consider, however, the m of going to the city, thus perhaps re— ducing the validity of the measure. Empirical Hypothesis He was confirmed. TWO significant zero— order Pearsonian correlations (in SR and CE) might suggest that in more traditional villages , social participation is a more important determi- nant of homophily in information-seeking interaction. Erpirical Hypothesis IIf, dealing with age, was not confirmed, perhaps for the previously stated reasons mentioned in regard to Erpirical Hypothesis If. Empirical Hypothesis IIg was not confirmed. Social status is 52 not a determinant of homophily in information—seeking interaction. Evidently there is vertical communication about innovations across social status levels on the present villages. Overall, mass media exposure and social participation are de- terminants of homophily in information-seeking interaction. Cosmopolite- ness, age, and social status are not important determinants of homophily. Literacy is a determinant for the more modern village. Innovativeness is a determinant fOr:more traditional villages. General Hypothesis III General Hypothesis III: ip_the communication network pi the diffusion.pi innovations, the degree pi;homophiiy between individuals with friendship interaction i§_gieater than the dpgree pilhcmophily between individuals with infOrmation—seeking interaCtion. Empirical Hypothesis IIIa, dealing with literacy, was not con— firmed; however, in more traditional villages the degree Of homophily between individuals with friendship interaction is greater than the degree of homophily between individuals with infOrmation-seeking inter— action. Empirical Hypothesis IIIb, dealing with innovativeness, was con— firmed. Three tests fOr the difference between the homophily indexes were significantly different from zero, indicating that in.more modern villages the degree of homophily between individuals with friendship interaction is greater than the degree of homophily between individuals with.information—seeking interaction. Empirical Hypothesis IIIc, dealing with.mass media exposure, was not confirmed. It appears that in more traditional villages the degree 53 of homophily on the basis of'mass media exposure between individuals with friendship interaction is greater than the degree of homophily be— tween individuals with infOrmation—seeking interactions in the more modern villages. Empirical Hypothesis IIId was not confirmed. It appears that in more traditional villages the degree of homophily on the basis of cosmo— politeness between individuals with friendShip interactions is greater than.the degree of homophily between individuals with.infOrmation—seeking interactions. In the most modern village, the direction is in the op— posite, but the score is low. Empirical Hypothesis IIIe was not confirmed. It appears that in more traditional villages the degree of homophily on the basis of social participation between individuals with friendship interactions is greater than the degree of homophily between individuals with infOrmation—seeking interactions. Empirical Hypothesis IIIf and IIIg were not confirmed. In.more traditional villages the degree of homophily on the basis of age and social status is greater than the degree of'homophily between individuals with inflammation-seeking interaction. Generally, in the most traditional village, the degree of homo- phily on the basis Of social participation, age, and social status be- tween individuals with friendship interactions is significantly greater than the degree Of homophily between individuals with infOrmation-seeking interactions. For fUnctional literacy, mass media exposure, and cosmo- politeness, the tendency shows that in.more traditional villages the degree of homophily between individuals in friendship interactions is 54 greater than the degree of homophily between individuals in information- seeking interactions. But on the basis of innovativeness, in more modern villages the degree of homophily between individuals with friend- ship interactions is significantly greater than the degree of homophily between individuals with information—seeking interactions. In a.more traditional village the degree of homophily between individuals with information—seeking interactions is significantly greater than the homo- phily between individuals with friendship interactions. In the present thesis, the findings in the more traditional villages provided better support for General Hypothesis III. Implication for Change The different kinds of interaction patterns may suggest different strategies for a change agent planning the diffusion of an innovation. For example, if there is high homophily between individuals with infor- mation-seeking interactions, the change agent Should put his efforts on the "soughts" at each level of the different characteristics, e.g., social status. Then, the innovation.would spread rapidly. HOwever, in the case of low homophily, the Change agent can convert with his efforts on a few opinion leaders at the tOp of the social status. Suggestions for Future Research 1. It is reasonable to assume that kinship and physical dis- tance between individuals effect the probability of their friendship interaction and information—seeking interaction; these variables need to be investigated. This is especially true in developing countries, where transportation facilities are poor, and where physical distance and kinship are probably more important determinants of interaction. 55 2. Dyadic analysis could be used for gatekeeper studies, news diffusion studies, and fOrmal organization studies. As mentioned previously, it is a powerful method for analyzing fOrmal and infOrmal communication patterns. 3. In the present thesis, there are only two kinds of inter- actions under investigation, information-seeking and friendship inter- actions. It would be worthWhile to investigate other types of inter— action patterns, and compare them. 4. For continuous variables, zero—order-Fearsonian correlation is a proper homophily index. For discrete variables, however, Coleman's (1958) index ofhomophilyl may be more appropriate. 5. In order to investigate interpersonal communication by dyadic analysis an entire census of a community needs to be inter- viewed. It is often difficult to interview all of the people in the population. There are two ways of dealing with the problem. The first method is to interview those chosen in sociometric questions. This sampling scheme is called "snowball sampling." The second method is to obtain the sought's information from.the seeker. Obviously, the lThe fOrmula is: . all - 011 hi ' M1 - Cll where hi = homophily index, IMi = number of choice made by persons in subgroup i cii = expected number of choices from persons in subgroup i.to others in subgroup i aii = actual number of choices from persons in subgroup i to others in subgroup i. The index could be used for 2 x 2 or 3 x 3 tables. 56 information obtainable is limited to demographic information, for example, age, and race, which the seeker can perhaps accurately pro- vide about his sought(s). 6. The construction of an index of individual's homophily is also needed for certain types of analysis. It could be Obtained by subtracting seeker's score. Then, the relationship between the individual's homophily and other variables could be computed. The purpose of this type of analysis is to identify which individuals in a social system have a high versus a low degree of homophily. APPENDIX 58 NATIONAL UNIVERSITY OF COLOMBLA Interviewer: Dept. Of Sociology (TRANSLATION FROM SPANISH Jan. 30, Bogota, Colombia 1964) STUDY ON ADOPTION OF NEW AGRICUIEURAL PRACTICES Date: Checked by: Name: 23258. Number on the map: Community: Cuatro Esquinas San Rafael eblo Viejo y Zapaguya Note: The respondent should be a farmer or cattleman and also the head of the family. Good morning, I'm a student from the Department of Sociology. we are doing a survey to learn about the reasons that cause farmers to adopt new practices. 1. How many years have you been a farmer? (YEARS) 2. Do you have the fOllowing on your farm?: (Check those he does not have) Milk Herd How many? 2 a) Cows - 2 years or over 2 b) Other Meat Herd c) Cows - 2 years or over d) Others e) Horses or oxen f) Mules and donkeys g) Pigs h) Poultry i) Rabbits j) Sheep k) Others NNMNNNNNN 59 (ASK ONLY IF HE DOESN'T ANSWER IN FANEGADAS) How much did you plant last year? 4. Amount harvested last year? 3 a) Corn (FANEGADA) (bultos) 3 b) Wheat " (cargas) 3 c) Barley " h" 3 d) Potato " " 3 e) Bean " (pounds) 3 f) Vegetables " (packages) 3 g) Others " ‘ 3 h) How many fanegadas of pasture do you have (if he doesn't know, how many animals per fanegam ?) How many lots do you have? (CIRCLE BELOW) (Circle) 5 a) What is the size of each 6. Percentage of land op- separate lot? erated that is owned. Lots 1 (FANEGADAS) H g _— n 4 _— n 5 _— n 6 7 8 9 10 5 b) Give total area of farm (Fanegada§)7 Specify area exploited by each of the fOllowing: .Exploited by owner . Leased by person who exploits it Tenant farmer . Half-owner (shares) . Leased to others .In pay for services (When more than one use corresponding letter, 1. e. , P.D.) U)C1b1()3>tu Have you employed farmlhands in the last 12 months? No Yes a) Total days worked by farm hands (DAYS) How many members Of your family worked on your farm in the last 12 months? Have you used ( ) in your farm? NOT USED 60 When did you use the first time LOLOLDLOLDLDLOLD HHllll LOLOLOLOCOCOLDLO [_s O 10 10 10 10 10 10 10 12. 13. 14. 15. 17. 18. r) s) t) u) v) w) x) additional information about weedkiller? Uses at present Imwfid®(M®E) FUngicide manzate (potato) Chemical fertilizer weedkiller FUmigator (or sprayer) Concentrates (finca, purina) Vaccine fOr Avian cholera Selection of poultry (Rhode Island Red, New Hampshire, Hi-line Vaccine for Anthrax (for cows) Vaccine for foot 8 mouth desease Desinfection of land Uses tractor Vegetables Diaco Nariho (wheat) Parda pastusa (potato) Funza (barley) lllllllla I ||l|||||| lllllllll Have you used( )in your house .A new stove Toilet MedicineIEhest fOr your~house Flashlight Vhist watch Sewing machine Grain grinder Where or from Whom did you first hear about weedkiller? (SPECIFY) In what year? (YEARS) Once you were interested, where or fromlwhom.did you receive Who or what conVihcedSyou on usingfweedkiller in yOur farm Who has triad to convince you not to use weedkiller in your farm? Did you treat part or all of yourifermlthe first time you tried a.weedkiller? Part Total Where or from.Whom did you receive infOrmation on how to use weedkiller in your farmP Do you think agricultural credit is a good thing? 4 Yes: 5 Very gOod or little 3 Don't know: NO: 2 Little or very good _____. __jl_. . 19. 20. 21. 22. 31. 31 a) TO have 15 heads Of cattle is 61 What is your marital status? (READ ALL THE ANSWERS) 3- Single Married 3— Widower Free union 3— Separated 19a. DO you think it is important tO talk with your wife about agriculture? Yes: 5 Very or u Little 3 Don't know NO: 2 Little 1 Very Did you plant beans on your farm.in the last five years? NO Yes 20a. When did you plant and harvest (MO.) 20b. How much did you harvest? _——' 200. TO whom did you sell? 20d. Where? “ Do you know Of a bean appropriate tO this region other than the climbing variety? ‘ NO Yes — - - - — 21a. Have you heard about "col andino" bean? NO Yes 21b. From.whom? What is your Opinion on planting beans in this region? Good: 5 Very gOOd or 4 Little NO Opinion Bad: 2 Little or 1 Very bad The fOllowing are some phrases taken from other farmers. Please complete them with.your own Opinions. 31 b) If I lost an anm in an accident (or on the farm) 31 c) What I need most is 31 31 e) In order tO live better on my farm I should 31 f) I hope my eldest son 31 g) In the next ten years, I'nlgoing tO- 31 h) My major aspiration in life is 31 i) What my farnlneeds most is 31 j) To be successful in agriculture today, one 31 k) Farmers in our country need 31 1).A good farmer has tO 31 m) A.real:man is one that 31 n) To make a farm produce, one must I could— d) Not having enough land is 31 O) TO make good profits in.farming, the farmer must have ----- 31 p) TO have 6 fanegadas Of land is 62 31 q) In our countryagriculture should be 39331_ 31 r) What I hope tO do in my farnlin.the future is 31 s) If I didn't progress in my work, I 32. What dO farmers here think Of the person who is the first tO adopt new agricultural practices? Good 5 Very gOOd 4 Little 3- Neutral, don't know, nO Opinion Bad 2 Little 1 Very bad 33_ 34. DO you listen to the radio? NO Yes 3H a) How many times a day 3” b) Where dO you listen? (MOST IMPORTANT) 1 At his house 2 At a friend's house 3 At the store 4 Other: 3H c) Have you heard any farnlnews over the radio? NO Yes 35. Do you read the newspaper? NO Yes 35 a) Which ones 35 b) How many times a week? ' ' El Tiempo El espectador E1 Siglo La Repfiblica La Nueva Prensa E1 Figaro ' El Campesino What others 3n —---—— NO 35 c) Is the newspaper read to you by others? NO Yes 35 c a) How many times a week? 36. Have you read any farm news in the newspapers? 0- NO 1- Yes 63 37. Do you read magazines? NO Yes 37 a) Which ones? cromos VisiOn Familia Life Others: 37 b) How frequently during the month? NO (TOTAL NUMBER OF MAGAZINES PER.MONTH) 35_ 37 c) Are the magazines read to you by others? NO Yes 37 d) How many times a.month? 38. Did you read any fannznews in magazines? 0- NO 1- Yes 39. Have you seen any movies? NO Yes 39 a) Where? Vereda Facatativé ZipacOn Bogota §§_ 39 b) How many in the last 12 months? 39 c) Did you see any farnlnews at the movies? 0- NO 1- Yes ~ HO. Have you watched television? ___ NO Yes no a) How many times in the last 12 months? _- H0 b) Where? _—' In his house At his friend's house At the store Other " 40 c) Have you seen any farm.news on television? 0— NO _1— Yes 38,39 41. 42. 43. an. 64 Have you talked to local businessmen about agriculture? 0- NO 1- Yes Have you talked with another farmer about agriculture in the last 2 months? ' NO Yes 42 a) With whOm? (ONE OR.MORE FROM THE COMMUNITY). 42 b) Did you give them any news Or did they? NUMBER Received Gave (l) (2) (3) (4) Have you talked tO people from.the Extension Service in the last 12 months? NO Yes 43 a) How Often? (PER.YEAR) 43 b) When you talked with this person.were you specially invited by hinfl INVITED NOT INVITED Have you talked with the school teacher about agriculture in the last 12 months? NO Yes 44 a) How Often? (PER YEAR) 44 b) WhO started thisiconversation about agriculture? The teacher ' ou What kind Of farm.news in general dO you believe are useful? (READ ALL OF THEM AND CHECK.) Yes NO Radio Newspapers Magazines MOvies Extension Service school teacher Businessmen Other farmers 0- Neighbors l; Relatives ' Others 46. 47. 49. 50. 51. 52. 53. 65 Yes NO Others: (Specify) What kind Of farm.news dO you think are more useful? (CIRCLE ONE ABOVE) What is the name Of the representative Of this region tO the legislature? ' Right wrong Of what country is RomulO Betancourt a citizen? Venezuela Wrong What's the name Of the national colombian aviation company? Right.(£vianca) Wrong What's erosion? Right (Lose Of land caused by currents Of ‘ water, wind, or type Of soil) Wrong What purpose does the radiator serve in cars and tractors? Right (To cool the motor) Wrong On this card are a number Of words. We are interested in knowing what words are frequently used by people and what words are used less Often. Will you nOW'read what's written on the card? Whole phrase correct. "The min'shgnd moged sha%ply in a gesEure Of respect." 53 a) Are there any words on the card which.you can.read? (CIRCLE THE CORRECT LETTERS) 0- NOne HAND RESPONDENT CARD SHOWING LADDER. POINT TO TOP OF LADDER EACH TIME YOU MENTION IT. (TOP OF LADDER IS STEP NUMBER 10) POINT TO BOTTOM OF LADDER EACH TIME YOU MENTION IT. WHILE YOU ASK A QUESTION, MOVE YOUR FINGER UP AND DOWN RAPIDLY. 54. At the tOp Of the ladder is the farmer who adOpted new agricultural practices first. At the bottom is the farmer who was last in adopting new practices. On what step would you stand now? 55. 156. 157. 158. 159. 160. 66 NOw, at the tOp Of the ladder are people whose Opinion is consulted by the people Of the village. An important person whose Opinion is valued by others; at the bottom stands a person who is never consulted by others. On what step dO you think you stand now? At the tOp Of the ladder is someone who has all the Op- portunities and chances tO dO anything he wants. At the bottom stands someone who can't dO the things he wants tO do. I On what step Of the ladder dO you stand now? .At the tOp Of the ladder stands a person who has personal capacity tO make his life happier. At the bottom stands a person who has very little capacity to make his life happy. On what step Of the ladder dO you stand now? Suppose that at the tOp stands a person who lives in the best possible conditions Of life and at the bottom stands a person who lives in the worst conditions Of 11fe. 158a. On what step Of the ladder would you say you stand now? 158b. On what step Of the ladder would you say you stOOd five years ago? 158c. On.what step Of the ladder do you think you will stand five years from now? Some persons are more conservative in their way Of doing things than others. HOW would you classify yourself? I find it very easy to change my ways. I find it easy to change my ways. I find it difficult to change. I find it very difficult tO change. Don't know. DO you agree or disagree with this statement? Sometimes I feel all alone in the world. Agree 4 Slightly or 5 Strongly 3— Don't know ' ' Disagree 2 ' Slightly or 1 Strongly 161. 162. 164. 165. 166. 167. 56. 57. 67 People's ideas change so much that I wonder if we'll ever have anything tO depend on. fgree 4 Slightly or 5 Strongly 3- Don't know ‘ Disagree 2 Slightly or 1 Strongly There is little chance to get ahead in this life unless one can count with the help Of an influential person. Agree 4 Slightly or 5 Strongly 3- Don't know i ‘ Disagree 2 Slightly or 1 Strongly .A good son tries his best tO find a jOb where he can be near his parents even though he might have tO loose a gOOd jOb somewhere else? 7 .Agree 4 Slightly or 5 Strongly 3- Don't know ' Disagree 2 Slightly or 1 Strongly The government's most important job after running the country, is making educational facilities available tO all. Agree 4 Slightly or 5 Strongly 3- Don't know ’ Disagree 2 Slightly or 1 Strongly TO make plans fOr the future is tO make yourself unhappy because these plans never work out anyway. Agree 4 Slightly or 5 Strongly 3— Don't know I 1 ' Disagree 2 Slightly or 1 Strongly Health experts say adding certain chemicals to drinking water results in less decay in people's teeth. If yOu could add these chemicals tO your water, with little cost tO you, would you be willing to have the chemicals added? Yes Maybe Probably not NO Others: Are you a member Of the Cooperative? NO Yes Directive ' 56 a) Do you buy from the Cooperative? NO Yes Are you associated with: NO Yes Directive a) Community board b) Catholic Action c) Night WOrship d) Christ Brotherhood e) Chicken farmers committee f) Is your wife a member Of the housewives club 68 ,g) Are your children members NO Yes Directive Of 4-S Clubs 7 _— h) Others: 59. DO you have another jOb besides your farufl NO Yes 59 a) How many days in the next 12 months? 60. FOR EACH MEMBER OF THE FAMILY. a) KINSHIP OR RELAr b) SEX CI, d) e) f) TIONSHIP TO HEAD (M or P) AGE YEARS OF *CAN READ *CAN WRITE OF HOUSEHOLD SCHOOL NEWSPAPER A LETTER (Yes-NO) (Yes - NO) Head Of household 1. 2 3 4. 5. 6 7 8 9. 10. - -._A _ A; AA‘H Questions only for those 9 years Of age or over. 61. How many years Of school.would you like your Oldest son to complete? (or in case you had children Of school age) PRIMARY SECONDARY 'UNIVERSITY 0 i 7 i 13 I 8 14 2 9 15 3 10 16 4 ll 5 12 6 61.a. DO you consider this possible? Yes ' NO 62. What occupation.would you like fOr your Oldest son? 63. During your life have you ever lived outside the community? ‘ NO Yes 64. Have you ever thought Of moving from.here? NO Yes 65. 66. 67. 68. 69. 70. 71. 73. 74. 64 a) Where would you like to gO? 69 How many times a year do you go to Bogota? If you fOund an unknown disease in potatoes with what other farmer would you discuss it? (NUMBER FROM ATTACHED LIST) a) With whom else would you discuss it? (NUMBER) If you needed credit tO buy a plot, tO what farmer would you gO for information on how tO acquire it? (NUMBER) a) Who else? (NUMBER) If your son (or other relative) were sick with what farmer would you discuss their illness? ' ‘ (NUMBER) a) Who else? (NUMBER) If the Municipal Council Of Facatativé or ZipacOn.wished tO nominate a person from.the Vereda, which.of the farmers dO you think it could be?_ ' ‘ _(NUMBER) a) Who else? (NUMBER) If you were to decide someday tO sell your products in Bogota instead Of Facatativa, whichrfanner do you believe would be the most appropriate person tO accompany you there?__ (NUMBER) a) Who else? (NUMBER) DO people ask you for advice or counsel? NO Yes What are some Of the things they ask you? 1) 2) 3) In general, dO you think people turn.tO you for advise and counsel more than tO others in the vereda? l- ‘. You 2- Others Have you talked tO your neighbors about new agricultural practices during the last six months? ' 1- Yes i 0- NO' 75. 76. 77. 78. 79. 80. 81. 82. 83. 70 Have you requested advise on agricultural problems during the past six months? 1— Yes 0- NO Do your neighbors believe that you are the first tO have adOpted new agricultural practices? 1- Yes 0- NO If you were president Of the Community Board, what would you do next year? 2- _High 1 1— Medium. A 0- Low If you were Head Of the Agricultural Agency Of Facatativa,- what would you do tO inprove the price Of potatoes in this community? 2— High 1- Medium 0- Low If you were Mayor Of Facatativa (or ZipacOn), what would you do to Obtain a better highway for the community? 2- High 1— Mediunl 0- Low If you were Minister Of Education what would you dO for Rural SChOOlS in Colombia? 2- High AAA 1— Medium 0— Low If you were President Of the Republic what would you dO to fight against violence? 2— High 1- Mediunl 0— Low DO you think that the respectable persons in this community have changed in the last ten years? NO Yes -END INTERVIEW HERE— Status Of farmers Of this community: 4- Very high 3- High ‘ 2- Mediunl 1— Low 0- Very low BIBLIOGRAPHY Berelson, Bernard R., Lazarsfeld, Paul F. , and MOPhee, William.N. Voting: A.Study_ Of Opinion Formation in a Presidental Mpai Un1versity Of Chicago Press, Chicago, 111.,1954. Berlo, David K. The Process of_Communication. N.Y., Chicago, San Francisco, Toronto, London: ’HOIt, Rinehart and Winston, 1960. Coleman, James S. 'Relationa1.Analysis: The Study Of Social Organiza— tions with Survey Methodsl' Human Organization, 1958, 16, 28-36. Deutschmann, Paul J. and Fals Borda, Orlando. Communication and .AdOption Patterns in an.Andean Village. Programa Interamericano de Informac1on‘POpular, San Jose, COsta Rica, 1962. Festinger, Leon. ALTheory_ Of Cognitive Dissonance. Row, Peterson, and Company, Evanston,h11 I95 97 Hartman, Joel.A.. Validity Of Using Sociometric Questions in Determining Characteristlcs Of Personal Information Sources. Paper presented at the Rural Soc1olog1cal Society, Montreal, Canada, 1964. Heider, Fritz. The Psychology Of Interpersonal Relations. New YOrk: John Wiley, _1957, 174— 217. Horton, Paul B., and Hunt, Chester L. "Primary Groups are Relationship- directed, and Secondary Groups are Goal OrientedJ' Sociology, 1964, 183. Katz, Elihu. "The TWO-Step Flow Of Communication:.An Up-tO-Date Report on an.HypothesisJ' Public Opinion Quarterly, 1957, 21, 61-78. Katz, Elihu and Lazarsfeld, Paul F. Personal Influence. The Free Press, Glencoe, 111., 1955. Kelly, Harold H; 'COmmunication in Experimentally created Hierarchies." Human Relations, 1951, 4, 39-56. Klapper, Joseph T. The Effects pf_Mass Communication. The Free Press, Glencoe, 111., 1960. 71 72 Ktsans, Thomas and Ktsans, Virginia. "The Theory Of Complementary Needs In Mate—Selection." Winch, R. F., McGinnis, R., and Barringer (eds.) Selected Studies in Marriage and the Family. New York: Holt, 1962, 517—532. -—_ ' Lazarsfeld, Paul F., Berelson, Bernard, and Gauolet, Hazel. IE3. People's Choice. New York: Colombia University Press, 1948. Lazarsfeld, Paul F. and Merton, Robert F. "Friendship as Social Process: A Substantive and Methodological Analysis." Berger, M5, Abel, T., and Page, C. H. "Freedom and Control in MOdern Society." Lionberger, Herbert F. "Adoption Of New Ideas and Practices." The I: Iowa State University, University Press, Aries, Iowa, 1960. . 'Tkmnmufity’Prestige and the Choice Of Sources Of Eann Information." Public Opinion Quarterly, 1959, 23, 111-118. Legitimation of_Decisions tO.AdO t Faranractices and Purchase FaIUISupplies 1n TWO Missour1 Farm Communities: Ozark and Prairie. Columbia,—Missouri Agricultural Experiment Station Research Bulletin 826, 1963. Lionberger, Herbert F'. and Campbell, Rex R. The Potential Of Interb personal Communicative Networks fOr'Message Transfer— from Outside Information Sources. A.Study Of Two Missour1 Ckmnmnities. CO1umbia,Missofir1 AgriEfiltural Experiment Station Research Bulletin 842, 1963. ' . Segregating and Differentiating Influences Of Personal Attitudes on the Choice Of Persons as Information Sources and Assoc1ates in Two M1ssour1 Commun1t1es. wPaper Presented at the Rural SOEiOlogicaI'Society,NOrthridge, CalifOrnia. Lionberger, Herbert F. and Coughenour, M. C. Social Structure and Diffusion of_1nformation- Agricultural EXperiment Station Research Bulletifi'63l, 1957. Marsh, Paul (1 and Coleman, Lee IL "Farmers' PracticesAdoption Rates in Relation tO Adoption Rates Of 'Leaders.'" Rural Sociology, 1954, 19, 180-181. MCNemar, Quinn. Psychological Statistics. New York and London: JOhn Wiley, 1962. Merton, Robert K. "Patterns Of Influence." Lazarsfeld, Paul F. and Stantion, Frank N. , (eds.) Communication Research. New York: Harper and Brothers, 1949. Merton, Robert K. Social Theory and Social Structure. The Free Press, Glencoe, 111., 1957. 73 Myren, Delbert T. The Rural Cormunications Media as a Determinant Of the Diffus ion pf Information about Improved—Farming Practices— in Mexico . Paper Presented at Rural Sociological Society, Washington, D.C., 1962. ‘ Newcomb, Theodore M. "The Prediction Of Interpersonal Attraction." The American Sociologist, 1956, 11, 575-586. Patel, Narsi. Shift from Primary t_o Secondary Reference i_n_ Inter— personal Commmnication. Paper Presented at the Rural Soc10— logical SocTety, MonUeal, Canada, 1964. Rahim, S. A. The Diffusion and Adoption pi: Agicultural Practices: A Study E a VilIage 2 East Pakistan. a, Pakistan Academy for V111age Development, 1961. Rogers, Everett M. Diffusion.pf_Innovations. New YOrk: The Free . Press, 1960. \ Rogers, Everett M. and Beal, George M; "The Importance Of Personal Influence in the Adoption Of Technological Changes." 'Social Forces, 1958, 36, 329-335. Rogers , Everett M . and CartanO , David G . "Methods Of Measuring Opinion Leadership." Public Opinion Quarterly, 1962, 25, 435—441. Rogers, Everett M. and Meynen, Wicky L. 3' Communication Sources for 2 , 4-D Weed Spray Among Colombian Peasants. " Rural SOOiOlogy, 1965, 30, 213-219. Rogers, Everett M. and van Es, Johannes C. Opinion Leade hip in ’ Traditional and Modern Cblonbian Peasant Communities . Dif- fusion Of Innovations Research Report 2 , Department Of Com— munication, Michigan State University, East Lansing, 1964. Signorile , Vito and O' Shea, Robert M . "A Test Of Significance for the Homophily Index." The American Journal of SOciOlogy, 1965, 70, 467-470. Stanfield, J. David and Ramos, Elssy. The Diffusion g in Innovation in Three Colombian Communities - art III. MimeO Paper, Epartment OT Communication, Michigan State University, East Lansing, Michigan, 1965. Troldahl , Verling C. and Van Dam, Robert . "Face-tO-Face Comummication About Major Topics in the News." Public Opinion 'Ea‘rterly, 1965, 29, 626-634. Warland, Rex H. Personal Influence: The Degree pf Similarity ‘_O_f Those Who Interact. M.S. Tliasis, Iowa State Un1versity, Ames, Iowa, 1963. Zetterberg, Hans L. "On Theory and Verification in Sociology." New York: Tressler Press, 1954. RIES "ITIMTTIT’TT