THESIS FOR THE DEGREE OF P H. D. MICHIGAN STATE UNIVERSITY GUSTAVO MARTIN QUESADA 1970 This is to certify that the thesis entitled Patron-Dependence, Communication Behavior, and the Modernization Process presented by Gustavo M. Quesada has been accepted towards fulfillment of the requirements for Ph. D. degree in Communication hem/a? (I? 222%? Major professor Date 16 March 1970 0-169 LIBRAEE.’ " Michigan State University .. MILUIHILIUML/IHIHUH”!!!Will“!!!IIIHHUHW 10383 3731 camels of ear .‘ECELVer Chara. (c) the ef :I'flht ' . ‘ - s ‘s.’ ¢ “‘ tn “he VA A 1:“ I “filming rm \, " n.» :- \ - ‘-. '94; C L HXUra‘Q‘r ‘N ‘sfifi “\ ‘n‘ _ ‘fi‘k‘E‘Qd “‘ 6“ v‘ A I,. “an“: s v“\flanor ‘ V «(I . in . “'N ‘ a “*Eh th v \‘I i“: a ‘ .)(._M ‘ ‘Q‘y, ‘V m} , ‘ C. .l. ‘5‘ 'teay‘q‘r ck“ pyd ‘t._ C \_ ‘I ;4\‘\ \J‘ F. v. Q 5. 0“ J ‘V P. \ u u. \‘ N \i ABSTRACT PATRON-DEPENDENCE, COMMUNICATION BEHAVIOR AND THE MODERNTZATION PROCESS By Gustavo M. Quesada The present dissertation deals with (a) the channels of extra-system communication, (b) a particular receiver characteristic which is here called patron-dependence, and (c) the effect (modernity) that communication brings about in the receivers. The objectives of the study were (1) to define conceptually and operationally patron— ggpgndgngg_(PD), (9) to analyze empirically the relationship of communication variables with PD and modernity, and (3) to determine the possible intervening effects of PD on the association between communication variables and modernity. It was hypothesized that: (1) physical mobility, mass media exposure, cosmOpolite contact, and empathy (the fOur extra-system communication variables), while positively associated with modernity, were negatively associated with patronédependence; and (2) stratifying for patron—dependence (from high to low PD) will increase the relationship between the extra-system communication variables and modernity. The data for the present study are part of a larger research endeavor dealing with the diffusion of innovations. The present dissertation concentrates on interviews from a sample of 315 Minas Gerais farmers, also called Phase 2.5 of the Brazil Difussion Project. The test of the hypotheses tere all done .,..'t attest-order :..erencer~= b3 d. z,.( . 5 ‘ l p 1?. individual" t'tpr senor hie" Ice: _.c.Lred mt‘r - v +- were all done at the .05 level of significance, one—tail test. The statistics used were zero-order correlations, highest-order partial correlations, and Fisher's Z test for differences between two correlation coefficients. Patron-dependence, defined as the degree to which an individual's decisions are influenced by actors occupying superior hierarchical positions in the social system, was measured with a seven-item quasi-scale. Modernity (or the ability to COpe with change) was Operationalized in terms of agricultural knowledge and agricultural innovativeness. About half of the hypotheses were supported and about half were not supported. Physical mobility as measured in the present study did not correlate with the other variables in the model, while mass media exposure and cosmopolite contact contributed significantly to the vari- ances in PD and modernity. Empathy, which is not highly related to PD, associated with the cognitive but not behav- ioral aspects of the modernization process. In general, the association between extra—system commhnication and the cog- nitive aspects of modernity was a sound one. Theorywise, there are indications that patron— dependent relationships (rather than a suppressor variable) are antagonistic toward the norms and behaviors determining nmflern roles. In a way, patron-dependence acts as a nmchanism to maintain boundaries for mentally and physically isolated social entities, while modernity is part of a me’Chanism acting in the Opposite direction, i.e., the broad- enang 0f the reference-system boundaries of a social space :cludinr. more , .. &' accounting for :ruetonts on t ”new-we a d .m... :1 at? . including more heterogeneous and structurally complex reference groups. Validations with other populations accounting for systemic differences are advocated and provements on the conceptual and Operational side are recommended. and im- highly PATRON-DEPENDENCE, COMMUNICATION BEHAVIOR AND THE MODERNIZATION PROCESS By Gustavo M. Quesada A THESIS Submitted to Michigan State University in partial fulfillment of the requirements for the degree of DOCTOR OF PHILOSOPHY Department of Communication 1970 é ' gnaw /0 3f: 7.‘ Copyright by Gustavo Martin Quesada 1971 kKer- ; A ‘3‘qfim A: n” ‘W ‘ur‘v‘gv‘r‘y v'. V9,, ”y‘- N! .‘PO vu., Q '1 ‘ c 6-; . ‘N -i~.-§.2E.LL O- H Accepted by the faculty of the Department of Cormmmication , College of Conrmmication Arts , Michigan State University, in partial fulfillment of the requirements for the Doctor of Philosophy degree. Director of" Thesis 0 Guidance Committee: ‘7'] / . flares— ' , Chairman / W £7 alts—’1} ._ / I I A 4‘ 4’ 4 ’ To: Eugene Jacobson, for his incomparable love for the scientific process, and Patricia, who, like a number, never had a chance in the process. .‘r ..uC 2:; e p :9 SEC( ore, 19f. 6; ‘ 59. vs ‘ alsrls 1 Fiv .s' , C: e t 4» \Acc rt HA ‘LU. '.- t," ., ‘ ..v :ts. ¢.:J‘~ “Chg 9 (on in t «a hub V‘ 'z‘r‘ :zfi“ ‘5'!“ h r. any“ .',. IOOOS «at J 4i 8 1 i n. S p: A n.v .v afic h U . ‘1 « ‘. .7 » id. ‘ .1 o + V v l D a. O Q\ v f o v . by n _ .sn 1 M t T we 8 . _ n. a «I r .. ‘ V l i x «Q Mt. I . .1 sq 1 EH!» G .V. . n v +U A» c . n i. C; Q» zfika "A s ‘ N\b 9.» RV no ‘ .- h.‘ KV NNV ‘. at n h .A h h. x. W h“! r . Hrs .s n .\ s . ham AH . R U Ally NHL- ,Nl\ . fl ‘ an t .\ a» ‘ s Knew st « p9.“ In s\ u LIV .AW. Maori . Max. s n .u s «\V v s a r: s 2v. t ACKNOWLEDGEMENTS Five years ago, at the time I finished my Master's Thesis, I felt that this was the most unnecessary section in the entire thesis. So, in order to be fair, I acknowledged "to all of you that, either in Brazil or the United States, in one way or another, collaborated to make this thesis as it is." Today, either of two alternatives has happen: I have been socialized into the system, or the contributions to the present dissertation have been so highly commendable that I feel it is fair to cite spec- ifically, even at the risk of forgetting someone. I rather assume the second interpretation as valid. Observation tells me that the "primacy and recency effect" is operative even in dissertation acknowledgements. Therefore, let me proceed with the acknowledgements in a manner parallel to the time that the different phases of this dissertation were done. Financially the present study has been partially aided by the U.S. Agency of International DevelOpment, the Department of Communication at Michigan State University, the Brazilian National Research Council, the Organization of American States, and the Midwest Universities Consortium for International Affairs. I want to acknowledge Gordon Whiting, Bill Herzog, David Stanfield, VicenteckaPaula Vitor, Jose Armando de Souza, and the late Celb Nogueira da Gama, all members of the staff 0f the Diffusion Project in whose early contributions might lmye beenfbrgotten. Interviewers and "faceless" respondents' iv :infrzbutions ":39? and the ‘ . ; c- 2:9 sc-en 3710?. fiscutrze secr- ‘ I ‘ "3”“). n ‘ :fla‘Atlo T: 0 :“CV‘ ‘-VOCES rtCel“ 5‘ ”‘1 v “ { ‘ .4 am iiar I”: K. M a 5,: n D W‘J'Q . ‘ a“ ‘ y¥gb “'F L ‘& C '5 . l n.‘ ‘ ‘1’“:2“ ‘ “\An7‘:+ v‘ a U K, ' l “A flinging. ‘\ . h \v". a+ ‘. J1; C‘l‘ ~'- ‘ ‘fi, )~ 1:.“ .‘\;at contributions are also recognized. The involvement of the former and the hospitality of the latter, although loveable, made scientific impartiality much harder to achieve by the author. Special acknowledgements go to Renato S. LOpes, Executive Secretary, and his Associacao de Credito e Assistencia Rural (ACAR) in Brazil. Without the participa- tion of this efficient team of technicians, The Diffusion Project could never have gotten out of the planning paper stage and into field situations. To the author, they are not only competent professional colleagues, but friends as well. Criticism to earlier versions of the present dissertation is gladly recognized. Amoqgfellow Graduate Assistants, the interest shown by Niels Roling, Joe Ascroft, Ed Bodaken, Jerry Durlak, and John van Es was without parallel. The author is particularly indebted for editorial comments received from John Coggins, Linda Martin, Herman Struck, and Martha Jacob. The writer expresses his thanks to Janette Swanson, the typist, and to Betty Darlington, whose mental lapses forced him to learn a great deal more than desirable about the mechanics of computer programming. Both are dedicated enmfloyees who hardly fit the definition or common meaning Of'bureaucrats. Among the faculty of the Department of Communica— tion at Michigan State University, special recognition is dedicated to Drs. D. K. Berlo, H. Kumata, V. C.7fibldahl, V and 3. Greentf :‘n‘lna all tar,- ‘ayunafis of “V, ‘vu. k s, ‘ l“ l u. .. ‘ ~ .“r‘C’W‘ arr, «ted-yv“, “H J "Q'h‘n 0“ *.r‘ A 1’ 'W‘O . “In “ ”‘5‘ ‘2‘.” Y‘Y‘F‘A‘chqy Lila uad‘VV‘V‘ and B. Greenberg. To my fellow Brazilian, Lytton Guimaraes, I extend thanks for his example, advice and companionship during all these years of graduate training. The contributions and guidance received from the members of my Graduate Committee, Drs. Vincent Farace, Denton Morrison, and Larry Sarbaugh, is highly recognized. The chairman of this committee was Everett M. Rogers, my boss, major professor,and friend, undoubtedly the most important single contributor to the present dissertation. A recent typology equates the prOperty to defer gratifications with modern behavior. To my family I want to say that you have been too modern, too long. I thank Tanira, my wife, who was able to "process" three children while I "cooked" only one dissertation. To the children, one has to thank whatever inspiration is possible to be found in their smell, noises and excesses of energy. vi it""l" 'l" .5. “1"“! ._ .H',‘ r- “ q... ‘ 7 ‘ vti"~.“‘ >44 "‘\‘Y' r n‘.v - (AM (In 'II_J .. “" b A-‘. 51.. q.- .. J‘s-“fl ." 'H- u. “A 11?: Tl) PT. I? OF nf‘prrn EFIITL; ETXQ ACKNOWLEDGEMENTS iv LTST CF TABLE viii LIST OF FIGURES ix CHAPTER I TNTRODUCTTON ........ .. ....... . .......... l The Problem ard Its Relevancy... ..... 9 Possible limitations ................. S Organization of the Dissertation ..... 7 II THEORETICAL FRWE OR R.EF‘ERET~ICE.......... 8 Review of Literature ............ ..... 8 Conceptualizations ....... . ....... .... 13 Theoretical Hypotheses. .......... ..... ?5 Summary and Discussion ..... . ..... .... 41 TTT T‘quImHODO-IJFGY. oooooooo 00000000000000.0000. 44 Geographical Setting................. 45 Sampling Procedures.................. 46 Instrument Construction, Interviewing, and Coding......................... Sl Operationalization of the Variables.. 54 Empirical Hypotheses................. 79 Statistical Analysis................. 84 IV PRESENTATION OF THE FINDINGS. ..... ...... 90 Who Are the Patron~Dependents?....... 9O Hypotheses Testing 92 V CONCLUSIONS AND RECOMMENDATIONS......... 112 Summary and Discussion............... ll? Possible Reinterpretations........... 119 Recommendations for Research 121 Recommendations for Change Agencies.. 123 BIBLIOGRAPIWOOOOOo'o0900000900000...ooooooooooolooo l26 APPEIqDIXOOOOOIOOOOOOOOOOOOOOOOOOOOOO0.00.00.00.00. 1-31 vii ‘. I) 4.. 7" t .. v‘ , VQ‘I - “~ni ‘0 I! v O f‘ ,. T Y . ‘-."\f\‘ k. I (,I.’ TABLE III-l III-2 III-3 III-4 III-5 III—6 III—7 III-8 III-9 III-10 III-ll III—12 IV-1 IV-2 IV-1 c \ TV-4 IV-S V~l LIST OF TABLES Prorortion of Phase 2.5 Resnordents Der Community.................... Mass Media Exposure Index ' Intercorrelations................ Empathy Inter—Item Index Correlations..................... Intercorrelaticns Matrix of FDR Items............................ Factor Analytical Solutions with PDR Items........................ Patron—Dependence Items, Their Harginals, Guttman Scaling Errors Item to Total Score Minus the Ihmi Correlations and Factor Loading.. Frequency Distribution of the PDR Index. ........... ................ Social Integration Scale........... Example of Innovativeness Scores by Different Computational Nethods.......................... Measurements of Central Tendency and Variance of All the Variables in the Present Analysis.......... Reliability of Qome Phase II and Phase 2.5 Items.................. Statistical Methods for Testing the Hypotheses....................... Characteristics of High and Low PD Individuals...................... Zcro—Crder Correlation Matrix of All the Variables in the Present Analysis......................... Zero—Order and Third—Order Partial Correlations between Extra- System Communications and PD, Agricultural Knowledge and Innovativeness................... 7ero-Order and Partial Correlations Between Patron-Dependence and Agricultural Knowledge and Agricultural Innovativeness...... Zero—Order Correlations of the Extra-System Communication Variables with the Modernity Indicators for High Patron— Dependents....................... Summary of the Hypotheses—Testing Results in the Present Dissertation..................... viii Page 66 7O 72 75 77 80 88 91 104 107C 115 LIST OF FIGURES FIGURE Page II—l Paradigm Relating Personality Characierietics to Communication Behavior, Role Attainment, and Belief Systems Among Adults........ 14 II-2 Venn Diagram of Some of the Constitutive Linkages of Patron— Dependence....... ..... ............. 22 II-3 Different Roles of the Test Factor Y in the Relationship between the Independent Variable Y and the Dependent Variable Z............... 30 II-4 Model Relating Patron-Dependent Relationships to Extra—System Communication and Modernization.... 36 III-l South America, Brazil, and Minas GerajSOO0.0..OOOOOOOOOOOOOOOOOOOOOO 4 0\ ix II) M- “7 ‘11 .4 C :‘3 l ‘ l ‘ :escrLbed as c ':‘=’-“- $ ‘ .".~v‘.3 that vs :MFY‘ w "er-086's of o :f‘fiirea‘ .waifli‘ulOI’l Of ,‘gl‘lq . Eocel. :~ Na Chapter I INTRODUCTION ... Indoctrinating the youth with group standards in accordance with parental interpretations of them is by nature an authoritarian mechanism. (Barnett, 1953) Traditionally, the communication process has been described as composed of a source, or a person or group of persons that have Some reason for engaging in communication, who encodes his purpose in the form of a message via some medium or channel to an audience of decoding receivers (composed of one or more persons). Berle (1960) calls the combination of sources, messages, channels and receivers the SMCR model. Some authors, like Rogers with Svenning (1969, p. 49), add a fifth element or component to the communica- tion process. It is called the effect or the receivers' reaction to the stimulus from the source. The present dissertation deals with channels of extra—system communica- tion (interpersonal contacts with cosmopolities, mass media exposure, and trips to urban centers*), a_particular receive? Pharacteristic, here caIbd_patron-dependence1 and the Iggticular effect, modernity) that this communication brings Eflput in the receivers. The present study does not deal with sources and messages.** -_ . *Due to their theoretical relevance for communica- tlon Scholars, these concepts are kept separately in the model shown in Chapter II. c **Deutschmann (1963) suggeSted that independently of Ontent, media exposure induces conversation resulting in 800131 Consensus. l 5—3 5 , 1"”) “‘91“: anc 7:: d“\‘ L. e or ‘1' the Cu. 244‘ ‘ er ‘1? mccem h 0. "W3 of . , If, W‘s-system 1 c 2‘ :7~°“r“v~~ w \ ybwu Ud‘ “; L ‘ ' ~ ‘ ‘ ’ "h 2 The purposes of this dissertation are:(l) con- ceptually and operationally define patron—dependence, (2) to study the contribution of extra—systemic channels in determin— _ing modern behavior, and (3) to study the possible intervenru: effects of patron-dependence on the association between extra-systemic communication* and modernity. Several researches have studied the association between mass media use and modernization,** and diffusion researchers have emphasized the role of interpersonal relations in the less deve10ped countries. Beside the study of mass media exposure, we propose the study of cosmOpolite contacts, physical mobil— ity and Lerner's empathy in explaining modernization effects. THE PROBLEM AND ITS RELEVANCY Patron-dependence (PD) has been characterized, at different levels of analysis, as a family, organization, or societal trait. Kenny (1960) defines a patron as someone who is regarded, and who regards himself, as a protector, guide, a model to copy, and an intermediary in dealing with someone or something more powerful than oneself. For the purpose of this dissertation, patron-dependence is defined as the degree 39 which an individual's decisions are influenced by actors *A social system is a functionally differentiated Impulation of individuals, as it is used here, a community. figtra—gystem communication refers to the transfer of mes- sages that originate outside community boundaries. These two concepts are discussed in a later section. **For example: Lerner (1958), Frey (1966), Deutsch- nenn (1963), Rogers (1965), and Herzog (1967a). sniping sup- ::erarchical 1. am manager ‘ Izlltical boss is a shy-5*” are n ‘ ‘ ‘ :atron-aepence_ . w . “ variable" K“ Element type I; .u 'I .2: L. ‘- ‘ .u‘fid-onsh‘p . * *Denn: 1: ii, 7‘91 ‘ w s“. 3 4.», 3:410“ we; ‘. ‘ Maison ‘l t ‘I'vrs ‘ t..u.‘ons‘r‘1“ C” .. Patron) W :s “A.“ ‘ A nh'y-n ‘ " :y‘nuudlasm 1 I‘u. ‘ b n'fia * e~f~ ‘43:: u re la v.1 _ — n.‘ ‘ = ‘ I‘ .7 MN, ‘ ......Jnoulps ’ ¥*; 1 ' r‘ ’s.‘\ ‘ a b . '5 “mt nd at; ** 'y,‘ s,‘ N. n 1‘ F; mud llsed ‘8‘ Ifl- ‘ I": Un be t v": Q 1::‘Q‘vé " “A". ‘ L. *at; a {:l‘vu‘ 4“" v ‘ . 3.. Mauls VIN A“: D1“ -. 3;. “~e t9)».- .0“-.‘ ‘vu 3 occupying superior hierarchical postionsfi This superior hierarchical position can be theiather in a family, the bank manager in a business transaction, or the local political boss in a political decision.** As a general trait, patron dependent relation- ships*** are not particular to a single culture. Rather, patron-dependence is a widespread_phenomenon. The Japanese oyabum-kgbum¥*** employer-employee relationship is a patron- dependent type of relationship, as is the Indian jajmani- K§fl£fl***** and the Latin American EEEEEQZREEE relationship.****** *Bennett and Ishino (1963, p. 224) define paternahsm as "a relationship between the agents in any economic organ- ization in which the employer acts toward his employees in a manner somewhat similar to that of a father toward his childrenfl which conveys the idea that the hierarchical and authoritarian relationship contains formal obligations. The paternal figure (or patron) owes something to the subordinate, just as the employee is expected to give to his superior. To compromise between the family connotation of "paternalism" and the organizational nature of "authority- dependant relationships," we prefer to use "patron-dependent relationships," which is broader in scope than the other two. **Fals Borda (1961, p. 247) relates the importance of authority and respectfbr hierarchy in Colombia to the use of pronouns like Usted,or'Your Mercy"in son-father or tenant- landlord relationships. ***While patron-dependant relationships (abbreviated PDR and used by Hutchinson, 1966) refer to a link or inter- action between at least two persons, patron-dependence (abbreviated as PD) describes a role characteristic of the individuals who usually engage:h1patron—dependent relation- ship. The terms (PDR and PD are used as essentially synonymous in the present study. ****Ova-parent, ko-child, bum—status (Odaka, 1964; and Iknnett and Ishoni, 1963). *****Ja mani-patron kamin-client (Kolenda 1963' and Pocok, 19625 , ’ ’ ******Fals Borda (1961) and Freyre (1946). r\€ VJ. f .he Ch m nit? f cw. Ext 091‘; . Q ‘3' ""v‘v 9; e and e c p ‘v“‘ Ayn ll & ck flaw: relevant beca‘ CR‘J .. .r. kc s...1 wJ ‘ 3\.S\‘hU $b «N 0 .1 on 4 The study of patron-dependent relationships is relevant because (1) it can provide a potential for a high level of generalization in theory construction by linking together some past sociO-psychological theories Of social change and economic develOpment; (2) it can provide an Opportunity for using either a monadic methodology (as is the case in this dissertation), where the individual actor is the unit of analysis, or a dyadic methodology, where the linkage between actors is the unit of analysis;* and (3) it can provide insight into the relationships between communication and modernization variables, therefore contributing to the selection of adequate change strategies. We have shown that patron-dependent relationships exist widely. Bennett and Ishino (1963) and Bennett (1968) suggest that PDR tends to occur in the isolated and less developed nations of the world. SO it is important to study the intervening effects of BBB in communication and modern- ization behavior. The literature on patron—dependence (basically anthropological) provides conceptual definitions, but almost no Operational measures Of the concept. The present dissertation provides conceptual and Operational definitions of PDR. Bennett (1968, p. 473), in his summary on paternalism for the International EncyclOpedia of the Social Sciences, * A third probable methodology would be a systemic methodology where an average value for all the individuals in a social system is the unit of analysis. If results achieved by applying different independent methodologies are similar, then conclusions from these results are more generalizable than if the outcomes had been otherwise. says "there i Studying than ieverdenca to ‘"““*"°=ize h; ‘ ‘ovu ~e.~‘ "' i'di‘ni‘aal 4: : ”Y a": " It, d..'1 V 7‘ :4. 2. -A 157531911: 1‘0] TN“ “' ‘ a . .uilimitqtl .:VH Chr' r “‘5‘ \-Qt Q“ UV\\. ‘ «L 1",; 1 “Ethel ce --\~h‘ ‘ “IN‘LQ‘ES, C( \_“> . .4.\\“g elltgg :y; ‘ “Cf: S says "there is no detailed survey specifically aimed at studying change (in paternalism)." By relating patron- dependence to the modernization process, we will try to hypothesize how changes in PD might be related to the process of individual or collective change. Some of the potential contributions of the present research are: 1. To advance a more adequate understanding of the patron-dependent concept by developing an operational measure of it, and empirically testing theoretical propositions about PD. 2. TO determine the role of patron-dependence in the communicative and modernization processes. 3. To determine selected characteristics Of polarized categories Of high and low patron-dependent individuals. POSSIBLE LIMITATIONS This attempt to explore the tOpic ofpatron- dependent relationships, suffers several limitations. One such limitation is that this study is limited to a Latin American population; at no point do we deal with the Japanae and German cases of PDR, for example. It seems that in these countries, contrary to what happens in Latin America, the ruling elites and institutional channels favor innovation and change, and PDR has had some positive effects on modernization. A seco to a certain ext 2‘? content rat i or 222191 change e: " fab en a «l 0 “yr cax .I s. , mletariat to ‘V C t .er :LLS Stud" 13, “”3 "19337.39 Cat. zata—cclle C Li C IT‘M, . -\Cr‘t dlSSe] ‘l‘+n_ "U“Hr ray-‘1 ‘ " P-L .L ép,‘ 6 A second limitation, this one in the analysis and to a certain extend overlapping with the first, is the lack Of concentration on the preconditions leading to PDR. Bennett (1968, p. 476) emphasized that, in some contexts, social change encourages the decline of patron-dependence; in other cases, rapid change creates conditions exposing a proletariat to insecurity, which can create new forms of patron-dependency. Other limitations are methodological in nature. This study is, as far as the author knows, an early attempt to measure patron—dependent relationships. The probability of measurement error is great because of the lack of past experience in the subject. As said elsewhere, the field data-collection was not done for the sole purpose of the present dissertation, so we could not include in the inter- view schedule as many PD items as we would have liked. Further replications of the present (and amplified) measure- ments of PD are highly recommended. Another methodological limitation is the nature of our sample. The test of hypotheses assumes randomization in order to generalize to the population parameters. Complete randomization is very difficult to achieve in underdevelOped settings due to the almost complete lack of population lists. Our sample does not meet the randomization criteria, as judgment sampling was involved at certain stages of the samfim design. 7 ORGANIZATION OF THE DISSERTATION The first chapter of the present dissertation states the purpose of the study, the definition of the problem of inquiry, the relevancy and limitations of this problem, and concludes with a summary of the dissertation organization. The second chapter deals with the theoretical frame Of reference, consisting Of a review of the literature on the topic of study, the conceptualization of the variables and their constitutive linkages, the theOretical model of prediction, the theoretical hypotheses, and a summary. Chapter III, on methodology, contains a description of the geographical setting, sample, interviewing, the Operational- ization of the variables and the corresponding Operational hypotheses, and a summary of the statistical methods used for testing them. Chapter IV includes the presentation of the findings, the hypothesis-testing procedures, who are the patron—dependents, plus some considerations about PD and modernity. The last chapter presents a summary of the findings, a discussion of these findings, recommendations for change agencies, and suggestions for future research. A 89" it 1 organ? Even T rath€7 cannOt most what I given it is that i means relate Havirr .. n'a ‘- Hsbeer, F U .etion, if trntothe dies ”“9... creview of some cfcermnication m mci Chapter II THEORETICAL FRAMR OF RRRERRNCE A seience without a theory is blind because it lacks thateflement which alone is able to organize facts and give direction to research. Even from a practical point of view the mere gathering of facts has very limited value. It cannot give an answer to the ouestion that is most important for practical purposes—namely, what must one do to Obtain a desired effect in given concrete cases? To answer this ouestion it is necessary to have a theory, but a theory that is empirical and not speculative. This means that theory and facts must be closely related to each other. (Lewin, 1936) Having defined the purpose. of the present dissertation, its relevancy and curtailments, now let us turn to the dissertation's theoretical construction. First, a review of some of the pertinent literature in the fields of communication, anthrOpOlogy, sociology, social psychology, and social change will specify the state of our knowledge about patron—dependence and will help develop, later, some theoretical hypotheses. REVIEW OF LITERATURE Patron—dependency and modernity* can be considered as independent end-products of the process by which society socializes its new members. Patron-dependence deals with the acceptance of influence exerted by individuals occupying positions in the system that enable them to perform social control. On the other hand, modernity deals with the learn— ing of the rationality that equips to c0pe with a changing society. *Modernity_is the stats oflpcoming (more) modern, while modernization is defined as the process by which individuals learn to generate and cope with change. 8 Clause and social conti Eases for social 532:8 identical. :cral imperative which tend to i: which exemplify mg the norms t: :‘mi ....al norms ar 53ml fOI‘WS as lfia 31 Contro IAIOcal cen~., CGF‘rCl TES‘ ‘arg‘Er . Effie . ‘3 l neOrPOpn 4 C1 9 Clausen (1968, p. 6) says that socialization and social control*go hand—in-hand; they are complementary bases for social order and continuity, but they are by no means identical. In addition to the norms** that constitute moral imperatives, all societies have a variety Of sanctions which tend to insure the support of the moral order. Behavhms which exemplify the norms tend to be rewarded; those violat— ing the norms tend to bring a measure of punishment. But social norms are not monolithic, coercive imperatives. They differ according to time, place, and the characteristics of the person. They are enmeshed with the division of labor in society. The means of support or enforcement of accepted behavior include not only such highly institutionalized~ social forms as the religious and legal orders, but also the informal controls that Operate within kinship, occupation, and local community relations. The effectiveness of social control rests on three factors: (1) the transmission of the moral norms through the socilization process, (2) the recruit- ment and socialization of control agents, and (3) the wide— spread acceptance Of the legitimacy of the norms. As an underlying basis for social control, socialization efforts are designed to lead new members to adhere to the norms of the larger society or of a particular group into which they are being incorporated. The group's values are, hOpefully, to become the individual's values, or at least to be recognized *Social control denotes the means by which a person is conditioned in his actions by a social system to which he belongs. . **A norm is defined as the modal pattern Of overt behaVior in a‘given social system. t3: him as havin especially when :tclation of mo :gartance of t :5 individuals :13: crepares t :ecause chem ‘5‘ x 2 ..ever 3‘snugi ; -......'.-StiC I‘ Eli-ZS e‘tout P30 GQ .NII‘V ‘ LOckl :“h , ‘NVLLUr in ‘L Mg‘y C .n‘ ‘O“: 9? w 1‘ u .L . Lari . ‘E111 .fi‘ ,_ fade *- t n. O thQ ‘: r ..‘e 2. 1‘ plicl ‘3le+. ”(mile 10 by him as having legitimacy. The modes Of social control, especially when exercised in reaction to deviance or violation of moral imperatives, help to emphasize the importance of the norms and to strengthen the commitments Of individuals to those norms and to the group. Socializa- tion prepares the individual to accept or reject change, because change is one type of normative behavior. Several anthrOpOlogical studies dealt with paternalistic relations at the societal level. Freyre (1946) talks about PD's historical origins in Latin America by referring to that continent'sfirst "vertical settlers". These were Spanish or Portugese soldiers and colonizers that came to the Americas without their families, took Indian and Negro women as concubines, and were in a position of super- iority in these new "families" (De Azevedo, 1962). Looking for social and Clique relations and their function in the achievement of vertical mobility in Brazil, Hutchinson (1966) and Leeds (1964) also studied the patron- dependent relationships at a national level of analysis, concluding that patron-dependent relationships favor a verthfid type Of interaction pattern rather than a more horizontal and egalitarian pattern. Willems (1955) stated that in the Latin American rural communities he studied, the greatest Obstacle to the conversion from Catholicism to Protestantism was the implicit loss of the saints as paternal sources of assistance. Descri I cultures, Galjar that distinmis‘r itemization: l ”‘v 0 AA \ . w h:- -- ...—man. Flva' e socioeconc r efforts as c Ffiyllyo" ‘kv!1.¢s or on a E I .CWEI‘Q D9031 8 V . _L '7: Ut.&L|’ 01" Detror" ‘ $¢C m H ’_J (D <1 (D H ll Describing the patronic syndrome in traditional cultures, Galjart (1968, pp. 85—86) identified three elements that distinguish patronic relationships as deterrents to modernization: l. The assumption that any real improvement in one's socioeconomic situation depends not so much on one's own efforts as on favors granted by secular or supernatural powers or on a stroke of luck.* 2. The disposition to seek to establish patron- age relations with peOple who are, or in the future may be, able to do one good. 3. The absence of feelings of solidarity** toward peOple with whom one is not related by kinship, friend- ship, or patronage. ‘This absence of solidarity is associated with a disbelief inihe presence of such feelings of solidarity in others. A few socio-psychological theories have dealt with the relevancy Of the socialization process as a way of emphasizing certain personality characteristics at the indi- vidual level, which lead to economic develOpment at the societal level. Examples are McClelland's (1961) ——v—_ *The basic difference between PD, fatalism, and self-control is provided by answering the question "who has control over the enviornment?" Concept Who Has ContrdLOver‘Me Environment? 1. PD Others (above me) 2. Fatalism Fate (not me) 3. Self-control Oneself (me) **Solidarity is the cohesion or attraction that a group has for its members. ”achievement mot gersonality" . ** very little has :mication varia nor, indirectly, :raiitional sociz Speak sling with ado. e.g., achieveme: :Lon behavior of CC.“ Om‘ deCVMlng alpha!“ \ ”ACE“ OW “81‘ a K;; i‘f" :0 much for ' vu'. & . . ' to attain an ' n msl "Welland (1961 fit Izoti'vatioY1 i activity. « .' OPT **._CT‘ea'L1 I I“! ‘ idols ch? 1.: n 4 " *np I m. ‘33! l . e (“‘4 **Lemoh| .nghieal lelroi ‘04 “4.QI‘ ‘ . .\' aetifimlzatio: aelandgvelopin ‘ 9:; he 1 ‘ MOST)“ expOSL 12 "achievement motivation"* and Hagen's (1962) "creative personality".** But other than Lerner's (1958) empathy*** very little has been done to relate the effect of com- munication variables directly to the socialization process nor, indirectly, to develOpment and modernization of traditional social systems. Speaking in more general terms, we feel that, dealing with adult pOpulatiOns,personality characteristics (e.g., achievement motivation) help determine the communica— tion behavior of individuals in a manner that is functional to becoming aware of, knowledgeable about, and proficient *Achievement Motivation is the desire to do well, not so much for the sake of social recognition or prestige, but to attain an inner feeling of personal accomplishment. McClelland (1961, p. 63) says that a high level Of achieve- ment motivation predisposes society to vigorous economic activity. **Creative_personality is the type of personality that allows the individual to have a sense of world- Orderliness, i.e., every phenomenom is part of a system whose Operation can be understood and explained (Hagen, 1962, pp. 88-97). Hagen's theory states that in every society where individuals develop a creative personality, that society will achieve economic growth. ***Lerner's (1958, pp. 43-75) model states that a critical level of 10 to 25 per cent of urbanization or industrialization is necessary before a country can start develOping "satisfactory" levels Of literacy and mass media exposure. With time, literacy and mass media exposure will help to create empathy and Opinionatedness, and then social, political, and economic participation. in fulfilling t3. tith time, newllv" El} will alter s 1968) of the 1nd themselves have :haracteristics, is continuous C :riering of the ‘~ lscroft, 196 1 I a \() Before ., iY'V‘u Miss (1955) .... I: -‘ :T'Wth. V‘s“. “RCtiOH filer; ‘* A“ in»; - Vii-(“‘ar‘les of ‘1’"- V‘- -¢..CD¢L€S in the Misti a‘ o I ‘ "‘ .§:1~ a “ \' “513? w * hn‘re‘ filth “h Ted:- \“‘| o ‘ Q t i“ "firs Pub ‘ 0 Us r ‘ 1 oh ‘ 13 in fulfilling the roles prescribed by society (see Figure II—D. With time, newly—socialized roles (mainly those requiring less PD) will alter some of the more tanganfial beliefs (Rokeach, 1968) Of the individual's belief system.* These changes will themselves have a further effect in changing other personality characteristics, communication behavior, and role fulfillment, in a continuous dynamic process of self—improvement, self- ordering of the world, and self—control over the environment (Ascroft, 1969; Roling, 1969). CONCEPTUALIZATIONS Before building a theoretical model, Deutsch and Krauss (1965) suggest the negessity of a stage of concept construction and construct explanation to help delineate the boundaries of the theoretical system. Operational definitions of the concepts are described in the next chapter. The present section conceptually defines each of the relevant variables in the system as well as some Of their possible constitutive linkages.** *There is, of course, undefined conceptual overlap among personality characteristics, communication behavior, Iple attainment, and belief systems, but for the purpose of the present discussion, they are considered as ideal types in the sense that they describe "what ought to be". Some authors, for example, prefer to see beliefs as part of Imrsonality, but we prefer to regard them as separate because “mile beliefs are more changeable, personality characteristics are more enduring. **While Operational linkages link terms to physical lfimnomena, constitutive linkages link constructs to other terms. Thejinkage with nature guarantees the scientist is dealing with reality. Linkages with other terms permit the Selentist to manipulate his terms through symbolic activity 1n.ways which produce assertive propositions (Berlo, 1967). fi-‘Jre 1L1 F“ :3 30""r ‘ :9 “4.an C8310} :_W:t Aflults. .' Valle u: Sz'lrstc 14 ~ -- h‘.~.-—~.—o—.— » ‘—.--—_.»——— PERSONALITY _*W“”M~MWNMJ> COMMUNICATION 5 ; CHARACTERISTICS BEHAVIOR L i h \\ \ \ \ \ \ \l __ “iii“ * ......... , ROLE EESTERS (_ -__.--_l_ ATTAINNENT (PDR) Figure II- 1. Paradigm Relating Personality Characteristics to Communication Behavior Role Attainment, and Belief Systems among Adults. *Dotted lines indicate the indirect relations with the belief systems, which are not part Of the present dissertation. Commas asource to recs ; :39 of channel f :10: event sittert trey carry rein :: “filtl‘le struct‘ 1'! 15 Extra—System Communication Communication is the transfer of messages from a source to receivers. The content of these messages and the type of channel that carries them is what makes a communica— tion event eifler(l) within—system,* or (2) extra—system, if they carry reinforcing notions already existing in the cognitive structure of the receivers or not, and (l) instru— mental, or (2) consumatory, if they carry information that helps to perform tasks that are fuctional to society, or their content is purely used for entertainment and amuse- ment.** Being interested in the development of isolated rural social systems, let us now focus on extra-system communication, which we assume is mostly instrumental.*** Durlak (1969) says that before a person changes his norms or attitudes and behaves in a different way, he usually seeks knowledge of alternative attitudes or modes of behavior *System is any group of identifiable interdepen- dent elements in continuous interaction with the environment. In a social system individuals are the elements, and,in our case, a within system includes any channels that are intrinsic to the community. **Notice that we do not infer the notion of purposiveness that Hovland and some of his disciples imply for the defInEion of communication Movland gt 31., 1953). In this way, one can assume that messages that were originated with a consumatory intent can become instrumental for a different social system or in a later point in time. ***The other pole of the same assumption is that wihin— system communication is instrumental only for the purpose of system-maintenance, therefore it reinforces local norms and boundaries against the threat of change or incongruent mes- sages from outside. For a discussion of instrumental versus consumatory behavior, see Bordenave (1966). H - thrown what We. z'ten the ind-ivy? :henessages and sin overtly be" The nr' social systems (, estiientally 0th Enav: oral altern We w l i ‘5‘ 3 “Giving info tenant for 831} : *Al ‘92“ S , '3~.S (1962 0 CC 16 through what Waisanen (1969) calls dissociative experiences.* Then the individual evaluates the source and the content of the messages and decides to adOpt the new norms or attitudes, or to overtly behave in a certain way. The process of receiving information from other social systems (either by intentionally seeking it or accidentally obtaining it) is called extra—system communica— tion. These communication experiences function (1) to make a person aware of behavioral and attitudinal alternatives, (2) to facilitate the process of developing meaning for alternative attitudes and behavior, and (3) to provide behavioral alternatives within the new attitudes or norms. We will concentrate on the three most common ways of receiving information about other social systems: (1) by physical mobility,* (2) by exposure to the mass media, and (3) by contact with cosmopolites; and on a factor, empathy, influencing extra-system communication. Although physical mobility might also denote migration, for the purpose of the present dissertation physical mobility is restricted to visiting cities or towns which might represent different normative systems from the normative system at the community level. Mass media exposure and contacts with specialists such as school teachers or extension service agents are also important for gaining knowledge of other systems. Diffusion *Also called cosmOpoliteness by other authors. Rogers (1962, p. 1?) refers to cosmOpoliteness as an external "orientation." Here it is treated as physical mobility to avoid confusion with terms as cosmopolite contacts. of innovation I sang the init; :ieas in famir. sought before 6. experts or com in soeci ic an: :en‘t innovation. tence of interr countries, is e I'lCh *Ong 3". ”Ce c - 31”? a). "mun {‘3 .12“ d nQW ’Te ‘ “be Of ..~ 43" dgesfi‘gel‘ t}. 0 “£93 In 17 of innovation reserachers report that the mass media are among the initial sources providing information about new ideas in farming.* However, "fother] farmers are usually sought before deciding whether to adopt innovations, and experts or commercial dealers are frequently the source for specific and technical information about how to imple— ment innovations" (Wilkening, 1964). The relative impor—_ tance of interpersonal channels, true in more develOped . ...—RM:- ;—- countries, is even greater in the less develOped countries. I " The concept of empathy is another ingredient in the change process. Empathy is the ability to project one- self into the role of another person. It expands the ability of the individual to identify with others so that "others are incorporated fin.the mental system of the sub— ject] because I am like them" (Lerner, 1958, p. 49). Empathy leads the receiver to identify with certain parts of the message (or with different sources), thus affecting the quality of the extra-systemic experience. Patron—Dependence For the purpose of the presentcfissertation, patron- dependence (PD) is defined as the degree to which an indiv- idual's decisions, are influenced by actors occupying *One not-always-true assumption is that the content of these communication situations is instrumental in achieving change and new ways of life, but one can infer that thelarger the number of contacts with extra—system agents (or vehicles), the larger the probability of being exposed to instrumental messages of a pro-change nature. aperior hierarc he definition :sa role prescr attains of the iaracterisflg o; iecision-naking 1 :riiviiuals in 16 1‘, c ‘ 1 n. secono cnarac asen us proble: tale. It inplie 18 superior hierarchical positions in the social system.* This definition implies: (1) that PD, as a general class, is a role prescription that limits the decision—making patterns of the individual, and (2) that the peculiar characteristic of PD is that it helps to center this decision-making process in the hands of a few privileged individuals in leadership (usually ascribed) positions. The second characteristics of PD would seem to indicate a serious problem for the develOpment of a society as a whole. It implies that there is a large subset of citizens who participate very little in the national stream of decision making, a situation antagonistic to the politics of a democracy. If there is a series of different types of decisions which parallel the hierarchy of leadership positions in thedecision-making process, then the most crucial decisions (usually those affecting the large majority) are made only by a small proportion of the actors in the social system. What are some of the constitutive linkages of PD? In other words, how different is patron-dependence from authoritarianism, dOgmatism, and other similar concepts? First, let us categorize and define these concepts, and then let us see how they are constitutively linked to PD. . *Notice that this definition allows the use of different units of analysis, as we explain elsewhere. In 1':- stuiy on the 8‘.) p. 6) that nerr tent in which i personality is that has develo is capable of a» 1:.iividual Cha r: "twp ' v.1 .eSIStent r, 7 ”31‘ ll and the I o... ‘ ‘ ...e lazi-Jewi sh :tar‘. Inn nersora‘ ‘ ‘A\‘J extreme right . ‘3 0388C] UpOn 1.". HClgmtismv St Alert led, Says t' 551‘ l9 Authoritarianism In 1950, Adorno gt a1, published the now classic study on the authoritarian personality. They stated (1950, p. 6) that personality is a product of the social environ- ment in which the individual develops, bat, once develOped personality is not an object of the contemporary environment. What has develOped is a structure within the individual that is capable of acting upon the social environment. This individual characteristic, though modifiable, is frequently very resistant to fundamental change. It has been suggested that the influence of World War II and the immediate involvement of the authors with the Nazi-Jewish conflict biased their work on the author- itarian personality toward overly emphasizing the politimwly extreme right. The theory of the authoritarian personality is based upon the psychoanalysis of prejudice. Dogmatism Rokeach (1960) criticizes the political and ethno— centric biases of Adorno gt_§l. and suggests "dogmatism" as a psychological alternative to the authoritarian personality. He defines dogmatism as a measure of the extent to which a person's belief system is closed (1960, p. 169). The theory of dogmatism, structure—oriented rather than content- oriented, says that the ability to form new belief-systems is affected by: (l) the ability to keep in mind the new parts to be integrated, (2) the willingness to entertain new systems, (3) th‘ zopresent new ”1 and (5) the deal belief system (3 Bole-Faster}: Stewar tion of the autl‘. assiciates attri psychological an 33111: propose tI.< Boil ' ‘ I! ‘nhd'n \’ ' shaman 18 71885 tne individ; is m a personai vering '3 l (4. 594 l ..Hr‘c fid EHVirc iilrorhe «Ants shes.“ 9. sEre Open envi ror ROlE‘mE savior aSSOClai 3:110 . :31) that Pole 5, ‘1 ~ n l 4 Know how to 20 systems, (3) the experiences of the past, (4) the ability to present new beliefs as gradual parts of integral blocks, and (5) the degree to which there is isolation within the belief system (Rokeach, 1960, p. 398). Role-Mastery Stewart and Hoult (1959) authored a reinterpreta— tion of the authoritarian personality. While Adorno and his associates attributed the cause of authoritarianism to psychological and psychoanalytical reasons, Stewart and Hoult propose that the cause is sociological. Stewart and Hoult's theory is based on the assumption that authori- tarianism is negatively correlated with the number of roles that the individual has mastered (therefore authoritarianism is not a personality characteristic). They also argue that role-mastering ability is more difficult to attain in more restricted environments; therefore, peOple from restricted environments should be more authoritarian than people from more Open environments. Role-mastery includes both role-playing, overt behavior associated with a given position, and role-taking, the cognitive process whereby a person puts himself in the other persons' place.* Stewart and Hoult agree with Coutu (1951) that role—taking must precede role-playing, since one must know how to act before he can adequately fulfill the k *Also called empathy by Lerner (1958, p.50), as Will be explained later. expectations 0 ear that r0] 9- arerson has t :I: ‘Vv Lrecy an Off ‘r‘ ‘\. ’25‘ e 393‘ Dan L: "toause it ls r. i '11 (. rtt D ‘l/) I I ‘3 (N f 'D I F‘ d)(") r4 O HID . "a A 09 ‘1 21 expectations of a new position. But Stewart and Hoult also say that role-playing and role-taking interact because once a person has taken a role and learned how to play it, his increased experience will facilitate further role-taking. Nepotism Aikin (1964) defines nepotism as the practice whereby an Officer appoints one or more relatives to industry or public service, or confers on them other favors, in order to promote the family's prestige and income or to assist in building upgapolitical machine. Aikin notes that different relationships distinguish nepotism from the broader and closely related term of patronage. Nepotism is exclusively favoritism toward family members, while patronage includes favoritism toward both kin and non—kin. Constitutive Linkages How are authoritarianism, dogmatism, and role- mastery, conceptually related to patron-dependence? Figure II-2 shows in diagramatic form the theoretical constitutive linkages. Patron-dependence differs from authoritarianism because it is a role prescription rather than a personality characteristic. Patron-dependence deals with the acceptance and submission to the hierarchy of the system. Authoritariankmi deals with the intensity of reaction to the content of one- sided ethnocentric messages.* . . *For example: degree of agreement or disagreement with items like "in view of the present national emergency, it is highly important to limit responsible government 'Obs o native, white, Christian Americans" (Adorno at £11., 950, p.108). / ‘. Author— ;” /// " itarienism Role- r.‘,c Ma s t e ry I // Dogmatism Nepotism Figure 11-2. Venn Diagram* of Some of the Constitutive Linkages of Patron-Dependence. *The Venn diagram,named after the logician J. Venn, offers a helpful way to illustrate set relations. Iowever, in no way are we indicating the precision of the shown relationship. Dopt lmfllsbeliei partially oven Exhdeal wit} individuals In gerson might 1 :rindividuall 3301'9. 31 the indivh less of the s 3Fdefinit10n “st-WOT)» . {- \‘1 bhe E Y“: X“. ~ ~ n K r " “O‘tanF 5-1 xii 23 Dogmatism refers to the structure of the indiv— idual's belief system. In this regard, patron—dependence partially overlaps with close—mindedness (dogmatism) because both deal with the mental isolationism characteristic of individuals unable to entertain new belief systems. Yet a person might be quite dogmatic against certain ideas, objects, or individuals without necessarily relying on decisions from above. While dogmatism refers to the close-mindedness of the individual, role-mastery stems from the restrictive- ness of the social system to which the individual belongs. By definition patron-dependence also deals with individual performancewithin a social system. Both mastery over a restricted number of roles and patron-dependent relationships tend to originate more frequently in socially or geographcalw isolated and restricted envioronments (Bennett and Ishino, 1963). New concepts, like patron-dependence, need to be checked for validity,* conceptually and empirically. The Inevious description of the constitutive linkages of PD is ah.attempt to validate the concept by looking at the sim- ilarities and differences with already-accepted concepts. Iflthough the author of the present dissertation recognizes tme importance of the empirical checks on the internal \mlidity of patron-dependence, the circumstances surrounding *Validity is the extent to which the operational— ization taps the dimension that the researcher is trying to measure. the field rese concepts (like PD by differer to check some Campbell and F patron-(169931“ Diff :icnifferentl that modernize \ '18 modernizati silly cope wit :COUt the iImOV uguities pI‘ I h mo lita‘ ' . lS relat *ie :H A Sf‘a‘gml‘ott AS t30c;étlé%9 24 the field research made it impossible to measure similar concepts (like dogmatism) or to attempt to operationalize PD by different independent methods that could have helped to check some of the validational processes suggested by Campbell and Fiske (1959). Nevertheless, the validity of patron—dependence should be thoroughly checked. Modernization and Innovativeness Different research traditions define moderniza- tiondifferently.* Nevertheless, all the definitions imply that modernization is a process of continuous change. For us modernization is the ability to generate and success- fully cope with change. Change and a prOpensity to cope with change are the essence of modernization (Friedman, 1968). Change occurs in an effort to increase control over the environment.** But, as White and Lippit (1960) imply, in a complex modern world the individual must be selective about theinnovations he adopts in order to c0pe with the ambiguities produced by continuous change. Innovativeness is the degree to which an indiv- idual is relatively earlier than other members of his social system in adopting new ideas (Rogers, 1962, p. 20). Innova- tiveness has been used as a comparative measure of modernity _ (Salcedo, 1968). The previous definition does not take into * These different disciplinary approaches were abstracted by Weiner (1966, pp. 3-4) ** As suggested by McIver and Page (1957, p. 500) a¥d Ascroft (1969 Roling (1969) also refers to this type 0 ass001ation. consideration 1 discontinue us: Prev: iii not take ii if. is definl accurat infer: ..ice more rati mg (1) the :easurement of 311519 indicat achieved previ lilC‘IEthlis rs: "‘vu' . Nf‘ng “'1 th C‘n 95 consideration the degree to which an individual may discontinue using an innovation he has adOpted. Previous measurements of modernity by others also did not take into consideration the knowledge factor. Kggw- ledge is defined as the degree to which an individual possessxs accurate information(as defined by experts) enabling him to make more rational decisions. It is believed that by intro- ducing (l) the discontinuance of innovations,* and (2) some measurement of agricultural knowledge, we shall obtain better single indicators of modern rational behavior than have been achieved previously, assuming that knowledge and adoption of innovations recommended by experts is a successful way of COping with change. THEORETICAL HYPOTHESES Before develOping the theoretical rationale that will lead to the theoretical hypotheses, let us first analyze the type of possible relationships in research designs. This explanation will help the reader to understand the author's theoretical model, presented later. Types of Relationships** In presenting data in defense of a knowledge claim, a researcher should provide first evidence of the relationsnp' *The assumption here is that control over the envupn- ment is achieved by rational behavior and that, sometimes, rejecting an innovation while in possession of accurate know- ledge about it might be more rational than adOptin the same innovation without adequate knowled e about it, an then to have to discontinue it because of i s impracticality. in! ' ' - - and reasonigglgfsfiggéggeii (1966)? based upon the nomencmmne between two vari E)iecinrocal, specify which is enfile;and (3 iaafent variatf er able). Ihie ‘ . N i K 7:5? .* t ‘ :“NSI ‘ Causali 1 cer :‘w-‘u n. r" v i. av, Gaugils‘ & :iwn ‘ "it: ‘ Vr * v I: in the ,Rvofexperime: Kali - Les 1n SUI‘V; 26 between two variables. Such a relationship could be: (1) symmetrical, when neither variable is due to the other; (2) reciprocal, when there is no immediate possible way to specify which is the independent and which is the dependent variable; and (3) asymmetrical, when one variable (the in- dependent variable) may influence the other (the dependent variable). Due to its theoretical potential for determining causality, we should concentrate on the asymmetrical relatuxr- ships. Causality refers to the process by which events are linked in terms of cause and effect. Besides covari- ability, causality requires either (1) that one event is temporarily prior to the other, or (2) that the nature of the events allows logical inference of dependency into their relationship. In the present disseration we are using causality in the second meaning of the word. The cause-effect type of relationship, character- istic of experimental designs, encounters particular dif- ficulties in Survey designs due to the lack of specificity in determining pure stimuli under field conditions. Social research deals more oftenly with a disposition, a state or condition of the individual, or a characteristic, a relahmfly more enduring state of the individual, and a response, act or behavior. Asymmetrical relations lead to the second require— ment of research design which is to show evidence of causalmv Causality is a necessary connection between two variables, but a causal relationship represents but one among a much larger number of necessary connections among variables. fine-order help Ccepanel, or r solving the can test field res? ti:e.* Then, 1 enable deter :c-efficient‘? 10:01" detern: 11) calls ..1 earn we COnnr \ 9:; . i“KIN? Ga ’ 1‘9th 27 Time-order helps determine causality in experimental deswms. The panel, or re—interview technique is effective in re- solving the causality issue in survey design. Unfortunately, most field research gathers data only at a given point in time.* Then, how does the survey analyst determine which variable determines what on the basis of a correlation coefficient? The key criterion for understanding the direc- tion of determination appears to be what Rosenberg (1968, p. 11) calls "susceptibility to influence". For example, it is difficult to establish any temporal priority in the relationship between education and television viewing. During the years of education, one watched television, and during the years of watching tele- vision, one went to school. Yet the direction of deter- mination is clear. It is logical to see how level of educa— tion may determine one's preference for certain kinds of programs,but it is much more difficult to perceive one's preference for certain kinds of programs as determining one's educational level. Therefore, susceptibility to influence connotes the difference among variables in their "fixity, permanence, or alterability" (Rosenbeng 1968). By the same token, it is more logical to infer that exposure to the mass media and other extra-system communication would equip the individual with information about c0ping with champ *The present study consists of only one measure- ment in time. and increasmr cent (that is :iiemity proi nrroee of the 5“ ~. a“ - . n v.9. mas] o g _‘.._-‘ 3097‘0n01iie Cor C‘M .. ‘3‘) WHEN or: \Q 4 | K“ :33.“pr . ¢§v 4L4 qnab . C A the exhe“. E“ ‘1m9:‘+: U : in t . K. prl . .. qt‘ 1C: )0 a} . 9h V.“‘e qO'n k" ‘HE 0 Cone; cu 28 and increasing his share of control over the environ— ment (that is his modernity), than to infer that modernity produces exposure to the mass media. Fo3_the purpose of the present study: the extra—system communica- ..-—..— tion variables (physical mobility, mass media expgsurg, cosmOpolite contact, and empathy) are considered indepen— dent variables leading to modernityl_the dependent -—-‘.-“- variable. But such an assertion needs to be tested. The most important systematic way of examining the relation- ship between two variables is to introduce a third variable (called a "test factor" by Rosenberg) into the analysis. The introduction of test factors into survey analyses enables one to exploit some of the virtues of the experimental design (like ability to control certain variables) while avoiding the inappropriateness of experimentation (like its artificiality of social setting). The introduction of test factors into data-anlysis does not overcome all the problems of correlational analysis, but it does enable the survey analyst to approach the, characteristics of the after—only experimental design and to share some of the strengths of that scientific approach. Considering "any asymmetrical relationship between two variables [as] an abstraction from a never— ending causal chain", Rosenberg (1968, pp. 30—89) char— acterizes antecedent, extraneous, intervening, and sup- pressant roles that a test factor could perform. We have already prov1de extra-sys+em cc :ciernity indic mire-dice netrc 29 already provided some logical reasons for considering extra-system communication as independent variables and modernity indicators as dependent variables. Now we shal introduce patron—dependence as the test factor of the present study, and analyze its logical viability in each of the intervening positions affecting the relationship between the independent and dependent variables.* 1. Antecedent Variables - A two variable relationship is a truncated segment of an extended causal sequence, and any meaningful extension of the causal se- quence can only intensify our understanding of the larger process. The antecedent variable isan effective influence in the relationship between the independent and the depen— dent variables (see diagram in Figure 11-3). Rosenberg (1968, p. 67) says that "the analysis can go as far as the imagination of the theorist will carry him, without ever reaching the ultimate or first cause". Given a logical rationale for assuming an antecedentiariable, the statistical requirements are: (1) all three variables (antecedent, dependent, and indepen- I'" dent) must be related; (2) when the antecedent variable is *Chapter I dealt with the general relevancy of patron-dependence. The next chapters will concentrate on the relevancy of patron-dependency to a Latin American population. a . a“ MNQ 5.3 VRN n. ~2qu XEhHIDZO s L. «x Hausa .sma..fl..b EC. sunfish 3O .mmapsasm> Hosesoo same moanwaaw> cmaosaosm was "meanmsoapwams xmms amps mmsaa smepou “meagmsoapsams Heapsspmpsm some wsossw psmawsemx .N manwasm> psmwsmsmm one use w mapmHsm> pamcnmaousH mg» ammepmp gasmcospwamm was as x sesame same one so mmaom pumpmmmsn .muHH magmas N“? N M m m l W L 1 w Nelilllluxallllll.% AmQEmQOapmHmm Haw so.“ osmogv sommmsQQSm .v muesm>smpsH .m 3.. seems a s8 2mg 05 maemswapxm .m psmemompsa .a memma *mmHmmZOHB¢Amm N moeo . .- , iii - t -_ in tn semxpkcafi fizmaznlao £A£ monummmarba «.26 £28 .mrg ...:5 mm $2: :vS $2....:~.>> aomazoo. - r: I.i.i|.fl_l.i.c.ie Azspmem use .mpompsoo . _ mpeaonoEmoo .mASmomxm m _ deems mme .zpaaanos .Snfimems mama. me .25. me a amoamsfi ..mét ca 4 iiiiiiiiiiiiiiiiiiiiii w . . III noproaszEEoo mgm . a in as mmwoaosH . azmazmama mgm mommmmmmbm_ eme.mma.mme.xma mmnm ezmQZMmeZWE their own dec anity. Cf C ecce of soneo iszerent ext system that 6 fun the cop: hypothesis, '. to. retrain-d 91. N“ ‘ Thai enable ’2' ‘ i 800 at}: 37 their own decision—making by members of their own com- munity. Of course, they could go to cities under the influ— ence of someone else, but the fact that they went and had different experiences equips them with a broader reference system that diminishes the likelihood of future influences from the community. Therefore, our first theorotical hypothesis, TH1* is: Physical mobilfiy is pegatively related to patron-dependence. The mass media, including print (like newspapers and magazines) and electronic (like radio and television), carry messages from different worlds and different ways of living. Exposure to them equips the individual with a more heterogeneous base for decision—making; therefore, making him less dependent of influences from those in positions that enable them to exert such influence at the local level. TH2 states: Mass media exposure is negatively_related to --.-.... —- -m “C patron—dependence. ‘Those that possess acquaintances or reference groups who live, have lived, or oftenly travel to urban centers, sooner or later will hear about their acquaintance? travels and new worlds through the channel of interpersonal communication. This awareness may influenpe_the new knower *TH stands for Theoretical Hypothesis, and it is used in order to differentiate this set of hypotheses from the empirical hypotheses (EH1) presented in the next chapter. t ‘1 tv to 1 I COTTU n Therefc DD CV . n v I ‘7'" ”J 7’17)- 1 C t abili \ A - emuspred to ‘ \l Ul’ cc 4L Orv 1P4 e I. V“ f and, cheap f1 9!. “ AM“ - '." 2 u 4 -L V. Jen“ tin O ‘,’° ma. 0 38 to shift his reference group orientations from the local community to broader, more cosmOpolite, and egalitarian base. Therefore, we hypothesize (TH3): Cosmppplgte gpppag: is negatively related with_patppp:dependeppe. Empathic individuals, by definition, are those able to put themselves into a variety of situations by being able to assume other person's roles. They are individuals accustomed to thinking in terms of different alternatives. Such ability is similar to choosing among real alternatives in the decision-making process. This rationale leads to TH4: Empathy is negatively related towpaprpp:dppendence. Modernity as the Dependent Variable The findings of Lerner (1958), Frey (1966), Deutschmann (1963), Rogers (1965), and Herzog(1967a)sug- gest that extra-system communication is positively associated with COping with change or modernity. Those who are not afraid to leave their own com- munity and travel to different places are coping with change. Change more often occurs in urban centers where greater concentration of peOple and competition for jobs are present. Exposure to this type of world should equip the individual with mental resources that will allow him to cope with change in his own world. TH5 states: Physicgl mobility is positively related to modernipy. also usual iodide and (either h; and cope \ til/'9 1V 1‘9 pm ...! W State “Peg ‘ ”312-1431; 4.. Tim 1 owl‘se‘L‘yQ '9 39 The media carry some technical messages. They also usually carry messages about different and changing peOple and situations often happening in a more modern world than the receiver's own. Exnosure to these mes- sages can produce behaviors and attitudes in the receivers (either by imitation or learning) that allow them to create and cope with change. Th6 is: MESS media exposur:_is posi- --.—.-——-a——- ——-— ..- tively related to modernipy. As in the previous case of physical mobility, contact with cosmOpolites leads to dealing with persons who think and act differently. To a certain extent, cos- mOpolites are more modern individuals who have learned to adapt themselves satisfactorily to a changing way of life. TH7 states: Cosmopolite contact is positively related to modernity. Empathy, a characteristic of those able to put themselves in different and hypothetical situations, is likely to be present in more modern individuals who are constantly dealing with the ambiguities of a changing environment. Therefore, it is hypothesized (TH8) that: Empathy is positively related to modernit . PD as a Suppressor Variable Individuals with a low degree of patron- dependence are, by definition, among those less dependent oncxuxal hierarchy of influence. Low patron—dependent individuals are expected to place a high degree of cred: bili therefore 39932 ”as other hen dependenc iv; .13, f i entials. 1818. w: 11 fl.“ 4O credibility* on extra—systemic communication channels and, therefore, are among those more likely to accept pro—change messages carried by the extra-community channels. On the other hand, individuals with a high degree of patron- dependence are among those more dependent upon the select— ivity, filtering, and acceptance processes of local influ— entials. It is expected that high patron—dependent individ- uals will place more credibility in messages carried by local channels than in messages carried by extra—systemic channels. On the average, it should take relatively less time for low patron—dependent individuals to acquire know- ledge and adOpt new ideas than for high patron—dependent individuals who have to wait for these new ideas to be in- corporated into the community's way of living and social ' norms. The different degrees of credibility should result in different degrees of association between extra—system communication and modernity. Since the previous rationale applies in the case of each of the four extra-system communication concepts, we opt for presenting the four related hypotheses as a group, rather than separating each one with a similar rationale. *Credibility is the amount of trust or believ- ability that receivers place upon channels or sources. :1 tit, hit v jfc L‘ +ha“ .+‘ J. W M U :r‘n 87701” H H (1;. .. gm“ ‘3‘ . F‘M ..ti In..- U - “ Q L U-Le ‘. - __“¥\ 5 ¢.‘ 41 THg = Rb $1.98} -- 3013,1331. hafaaaiflrg- 36.13“ on.- ship_witb_mpdernity amgna low patron-dependencn individ- uals than among high patron—depgpdence individuals. TH103 Mq§§_media exposure has a higher_relation- —-.~ -.--~ ..--*-—-- -«—.- _.—-.—-..-— ship with_modernity ampng low patronfjepepdenpguipdivid- uals than among high patron—dependerce individuals. THll: Cosmgpolite contapt has a higher relation- ship with modernity amopg low_patrqn:depepdence individuals than.amppghighgpatron-dependence_individuals. TH12: Empathy has a higher relationship with modernity among low patron—dependence individuals than among high patron—dependence indiyiduals. ****** Let us reinforce once more the notion that extra— system communication to be instrumental must carry a con- notation of quality as well as quantity of contact. It is the content of the media, such as information on new Oppor- tunities for social mobility or editorial comments favorable to a political candidate from the Opposition party, that provides for changes in patron dependant relationships, rather than simply the amoupt of mass media exposure. SUMMARY AND DISCUSSION This chapter presented the dissertation's theoretical frame of reference. After a short review of the specialized literature in the field of socio— psychological models of develOpment, the author presented aconceptua] patron-depenr.. system comm: the main dew lzi'tages with satism, and r. :aradim was I ‘phrq‘. K‘Ju‘} ~lca1 mOb ...... Leg“ -. EYtra‘q‘ L! a £13,111. zed 1 4. PM} ‘Htla .. larger 42 a conceptualization of the variables relevant to the study: patron—dependent relationships, modernity, and extra— system communication. In the case of patron—dependence, the main dependent variable of the study, its constitutive linkages with other concepts, such as authoritarianism, dog— matism, and nepotism, were analyzed. Finally, a theoretical paradigm was develOped in which extra-system communication (physical mobility, mass media exposure, cosmOpolite contact, and empathy) were treated as independent variables, patron- dependence as suppressor variable, and modernity indicators as the dependent variables. Based upon this model, the author hypothesized that extra-system communication is negatively associated with patron—dependence and positively associated with modernity; and that among low patron- dependence individuals there is a higher relationship be- tween extra-system communication and modernity, than among high PD individuals. The prOposed conjectures allow the folbwing pos- sible advantages: 1. A new approach to the study of modernization, looking at it as the process through which individuals are socialized into broader social systems which provide them with a larger share of control over the environment, and therefore, a greater ability to c0pe with the uncertainties of change. R) psychologic such as per tenor, rol eni their 1 theories 01 variable , ] RIC}; an i] 43 2. A new integrative look at several socio- psychological theories of social change. Generic terms, such as personality characteristics, communication be— havior, role attainment, and belief systems,are advocated and their use should hOpefully produce more generalizable theories of social change. 3. A new conceptualization of the main dependent variable, patron-dependence (here defined as the degree to which an individual's decisions are influenced by actors occupying superior hierarchical positions in the social system), in a way that permits the use of different units of analysis in future instances. Either the individual, the communicative dyad, or the whole system can furnish the necessary elements to measure patron-dependent relationships. OWE? oraliza theeretieal a0 reliability of followed by tit! sic: of the ste e" The c later researc] 135:1: a1 8001 eitarment 0“ C and 4 «naheed bl. “3 The pres .xee Chapter III METHCDOLOGY Never measure the height of a moun— lfilhlflltll you have reached the top. Then you will see how low it was. (Hammarskjold,in Jantsch, 1957) This methodological chapter describes the geo- graphy of the research setting in the Brazilian State of Minas Gerais. The chapter presents the sampling procedures as well as explains the stages of questionnaire construc- tion, pretesting, interviewing selection and training, coding, and card punching. The chapter ends with the operationalization of the variables pertinent to the theoretical model and a consideration of the problem of reliability of measurements in less develOped settings, followed by the empirical hypotheses and a brief discus- sion of the statistical methods which were used to test them. The data reported here came from part of a larger research project on the Diffusion of Innovations in Rural Societies.* This research was conducted by the Department of Communication at Michigan State University and financed by the U. S. Agencyibr International Develop— ment. The present research concentrates only on data from *This chapter is heavily based upon the author's experience as Assistant Country Leader of the project in Brazil from November, 1965 to August 1967. 44 the Brazil " aram‘om ear limes Gerais where the looJ four years. research ende the-midile be Hortheast of EEO Paulo and The 7“ ‘ PM a DODula: ’+ M. at with only 45 ‘ the Brazil "Phase 2.5" of the Diffusion Project, which is a random sample of 315 farm operators from the state of Minas Gerais. The farm operators live in communities where the'bcal extension service worked for mere than four years. Interviewing was conducted during July, 1967.* GEOGRAPHICAL SETTING Minas Gerais was chosen over the other Brazilian states because (1) its economy is mainly dependent upon agricultural production, (2) there is an institutional infra-structure to provide institutional support for such a‘ research endeavor, and (3) the flingigg farmer is a man—in— the-middle between the most backward areas of the North and Northeast of Brazil and the most developed agriculture of Sao Paulo and the other southernmost states of the country. The state of Minas Gerais (shown in Figure III-l) had a pOpulation of more than 10 million in the 1960 Census Its area is larger than Texas and about the size of Spain, but with only a third of Spain's population. Minas Gerais extends from the Atlantic coastal range of mountains as far west as the Central Plateau to the border of the Federal District, where the recently built capital of Brazil, Brasflha is located. Neighboring stateseue Sao Paulo and Rio de Janeiro to the south, Espirito Santo and Bahia to the east, *For further information about field operations, the reader should see Stanfield 33. 31; (1968). 46 4’ . . BRAZIL” . ' MW. 99'7““ ‘\\ ,3.. y o"‘ Figure III-l. South America, Brazil, and Minas Gerais. Bahia attain Federal Di s‘ Farmers m f more copnerr sugar cane, of the state on field crr sperations a western reg: ‘l 46F? Bahia again to the north, and Goias, Nato Grosso and the Federal District to the west. The agriculture of Minas Gerais is transitional. Farmers in the southern regions of the state are relatively more commercialized, producing such market crons as coffee, sugar cane, and tobacco. The more isolated northern areas of the state have primarily a subsistence agriculture based on field crOps such as corn, manioc, beans, and rice. Cattle operations are of two types: beef on open ranges in the western regions, and many dairy herds in the central and southern regions nearer to the Rio de Janeiro and Sao Paulo milk markets. Many of the beef cattle are trailed from Minas Gerais for fattening in the State of Sao Paulo. Generally speaking, farming in Minas Gerais, either in terms of the type of crOps cultivated or in terms of the ways of handling dairy or beef cattle, is quite similar to farming methods in the rest of Brazil. SAMPLING PROCEDURES Our data came from a larger research endeavor, as explained earlier. Therefore, before explaining the sampling procedures for Phase 2.5 (from which the present data are taken),* it is necessary to explain some of the rationaleihat orientated preceding phases of the Brazilian Diffusion.Project** *The present data are taken from Phase 2.5 because (although there was some pretesting in Phase II with certain PD items) this was the phase that included the 10 original items measuring patron-dependent relationships. **The other two nations in the Diffusion Project are India and Nigeria. cess or fail communities. fie Credito 0 local office years. Sire: of change acre were consider 0 {‘me C11 ' H d r 10 atom » onp mam “1‘6;an . 47 Phase I was a survey of the causes affecting suc- cess or failure of change programs in 76 Minas Gerais communities. In 1965 the state extension agency, Associacao de Creditp'e Assistencia Rural (ACAB), was working in 126 local offices,* and had worked in 78 for more than three years. Since the Phase I aim was to determine the extent of change agency success or failure, these 78 local offices were considered as the total pOpulation of study. A 50 per cent sample was randomly drawn in each of three ecological regions. Then each of the randomly-selected ACAR local offices provided the name of the "best" and the "worst" communities in which they were working. The remaining 76 communities** constituted the Phase I sample and the start- ing point for Phase II. In Phase II the unit of analysis shifted from the community to the individuals living in the community. This Phase aimed at gathering initial measures of innovative- ness and opinion leadership, the two main dependent variables, before the introduction of experimental treatments in the *Roughly speaking, each local office works within the political boundaries of the municipio or county where it is located. The office consists—6f an agricultural and a home economics extension Specialist, a jeep, a secretary, and a desk. **Two ACAR offices in the Northeastern region of Teofilo Otoni were, at the time of data-gathering, working only one community each so they were withdrawn from the original sample. seidied cons :easurement s was obtained a different nmities. I carpi-etc sur . 'l - .ais flak: n; ‘ P 03 ACAR v ‘1 ‘5 \ LGTTJ‘I‘L" J“, O . / Tina . s_Q ._ ~ 111 . .r i-‘N 'L ‘-« i J5 Lk‘é c; y'?‘ ‘. ‘b..n'r‘\un7 t . Q-s‘ final. is: 32%,,“ “Jon 1 Um . ‘h‘: vL ‘*tlel ‘hamer 48 studied communities. These treatments and their "after" measurements constfiuted Phase III. The Phase II sample was obtained by ramdomly drawing 18 communities (each from a different local office) from the original pool of 76 com— munities. In each of these experimental communities a complete survey was performed of the pOpulation of individ- uals making farming decisions and the survey resulted in a Phase II sample of 1,307 farmers living in communities in which ACAR had worked for more than four years. Phase III called for the introduction in each community of either literacy or animation campaigns (also treated as pre-treatments) followed by either radio or print forums. The intermediary survey between the pre- treatments and the media forums is called Phase 2.5. The lists of respondents who were interviewed in Phase II were used to select respondents for Phase 2.5. Each list supposedly contained all landowners in a particukn‘community. Two types of samples were selected from these lists of respondents in order to cut down on mterviewing costs: random and purposive.* In each of the ten *The fact that not all the Phase II respondents were illiterates (or participants in the literacy campaign) was the cause of an "inflated" representation of literacy communities in the Phase 2.5 sample (The figures were: 3 animation communities, 6 literacy communities, and 2 control communities without exposure to either of the pre— treatments). purposefulli' about 30 far III-1). Ir. runity, TGf‘? arespondent on replaceme: “02" , .wycndents ‘ 0f roads, WU P90 me With ' 49 purposefully-selected communities, a random sample of about 30 farmers was selected from the lists** (See Table III—l). In the cases of death, migration from the com- munity, refusal to be interviewed, or unavailability of a respondent, the interviewers' supervisors were instructed on replacement procedures. The interviewers were instructed to try at least three callbacks at a farm before giving up and substituting another name. Substitution was made with the help of a table of random numbers and the identifica- tion number of the respondents remaining on the list. As is common in rural interviewing, certain respondents were extremely difficult tolocate. The lack of roads, widely scattered farms, isolated locales, and peOple with the same or similar names made the location of respondents difficult. Night interviewing, although easier for find respondents who might be working in dis- tant fields or at fie market during day hours also had its difficulties because of the lack of electric lights and general hazardous travel after dark. *With the exception of Corinto, where only 18 farmers were interviewed, (at the end of Phase 2.5) to in- crease the number of literacy participants in the Phase 2.5 sample. See Herzog (1967b) for further details. Cr MN LtihE it CTQ£2C1 OEQLZECLCZ thfi CCL .~ .-III I. Illtlll'- i .. .\.. Z kw.“ HC~pEEOC QC ,HrfikhU .HCUOQ MK —\HV~\ v \r... . f. ~,.C::Cmv .~ Tut; n.\~ ..u‘Hs~..L.§a.....~..~ a. o «um !\,., .\.~.~ -\ Md aka. .i'urh A...lq..r~.s 5 ~ I ~ ~ ~ .....~ .~ .9? 5O .sossm msewszop Op mac KOCH can» @905 we 85w me£e* *ccm mam mHgEmm Hmpoe .Qfll .mfil mLM¢H< oMmhsoo an: .OH ca mm messages mmesnom .m 0H cm ongoafisa mflaomoqe>em .m as mm enema mzsmpH .e 0 mad CAPSmQ mU OSCflHmHhSO OPQflhOO .0 as mm moves oeaomomq chews .m wm mm mcmoem mLmnomswm .v as cm mameou o>oz cam .m HH QM whhmm .mfl Owoh. owm ”IMO—:52“ mOPSwm .N CH mm mesa: ow ovmgoom osmossonmz Odom omm .H s z speszeeOo moemmo emcee mma mo. one we came Scam psmsmsssc aHPSwOflmfisme mew mpcmaofimmmoo seasmamssoo dmsfiasmccbxx map as smfiaswm cmoa>opm mm; msaoscs Empa mpmHsEoo onsx ii mo. mmq ss. sa. qu em. ma. m. mm. ss. mamas moaz pmzeo Haa so Haves was 632.25 mQOfiPmHmkhOo EmpH oo.a mo. oo. Ho.u mo.u so. so. as. mo. massages .oH so. «A4 Maw ma. mo. mo. mad mmq saws Heoacsoms .m mo. mo.u mo. mo.u Ha. mo. mmq mcaapoao .m Had mm. oo. mo. mo. wad moawpmomm nos .s os. mo. oo. no.1 mo. m.pmmaum .o mo. so. so. Ha. m>apmamm meam .m mad Hm. oa. maaxoem .e MMJ wad esopmsmgo .m *xMMIn mcfihfim mishaps .N xsoflmeSOoo m.:om .H oa m m s o m a m m H smpH Amfimuzv mamsH mos so sesame soapmfimsa°oampaH .s-HHH mfigms 64 .omsaahmoss mam Ewes =qmmao= a hoe assumes pmmgmfig mesx RNH Rwa Rafi Ram confidamxm moswssm> so soapsomosm sm. sa. mm. we. dams Hmoagzoms .m ms. Ho. ao.u om. msflapoao .m mm. mo. mm4 mm. mozmnmmmpm nos .s wa.n mo.n qu me. m.pmmapm .6 so. mo.: *mmq mm. m>apmamm mass .m mm.u wmq Ha. mm. moaeoem .e sH. mmd Ho.: He. msonmomso .m mo. mmq so.: we. mcflszm m.moae .m as. as. am. sm. possessooo m.som .H HHH HH H mmsaowoq sopomm pmsam ‘llu‘..-llll ‘I‘Allll‘lu soapzaom sopomh odewpomlxsaasm> soaPSHom mfixd Homeosapm amps mEmpH mom spas msoapzaom Haoapsamsa sesame .muHHH wanes 66 / other items, evenly divided, load on separate factors: one factor deals with the decision-making process in the nuclear—family group and the other factor deals with the non—nuclear family. The items about sons' occupa— tions and technical help are not "clean" items, but since they have some loading on either of the two meaningful factors, we retain them for the Guttman analysis.* In Guttman scalogram analysis, in order to <3btain a reasonable marginal distribution** of percent- eyges of respondents agreeing with the different items, int was necessary to eliminate the item about perfect aggreement with what the priest says. The remaining seven items (shown in Table III-6) yield a coefficient of Iwataroducibility of .85 indicating a quasi-scale according 'tC> Guttman's criteria.*** Further elimination of the TM3> Hopes map so pnmoams Hm msawaaxw noes; .mIHHH wanes as ompsmmmsa soapsaom Moscow maxmlammsosahm on» Op osomwmpaoo mmSHm>x **0.0mm whoahm Hopes sm. Hm. o.me ow posses sn soapomamm soapwmzooo .msom .H mm. sm. o.mv mm .tm>HpmHmm ammz now .s we. om. m.mm mm so>mm m we aHmm awoassome .0 mm. am. m.mm we m>apmamm Mo meanam .m mm. ma. m.am mm ‘ msoapoappmmm msaonm .msom awakens .c we. mm. o.s¢ Hm msoapoahpmmm . mcassm m.mmag .m as. mm. m.mm ss msopmomgo weaves .m Empa spa; . EmpH one apogee mnemohmm meson mama one % EmPH *mmsaomog maze: mmsoom msaHmOm so % Isoamms mo psmoamm sepomm Hopes Op EmpH so scapwamhsoo mammaws< Emsmoamom cosppzc .Amamnz~ mmsavmoq hepoma one .msoapmamssoo EmpH esp mama: mmhoom Hepoe o» EmpH .msoasm unflawom swappsc .mawsamsmz hamse .mEmpH mosmosmammlsohpwm .mIHHH wands 67 the two coefficients of reproducibility, it was Opted that the first solution was preferable as it included a larger number of items, (seven). It should be noticed that the seven-item PD scale do not include neither individual items contribut- ing less than six percent of the total variance, nor "consensual items"* that could have inflate the coef— ficient of reproducibility. The correlation matrix (Table III—4), the factor analytical solution (Table III-5), and the (}uttman scalogram analysis (Table III—6) of the PD items Inight be interpreted as presenting somewhat contradictory evidence for the use of a total PD index. Let us first eumflyze each individual piece of evidence and then, after 'the analysis, decide on which would be the better combina- tion of items to represent our PD index. The correlations of each of the nine remaining (dichotomous)PD items to the total score minus-the-item Eire all in the expected positive direction and all are ssignificantly different from zero. These correlation (zoefficients, although not unusually high, tend to pro- ‘ride some evidence for clustering the individual PD items iluto one PD index. \ *Consensual items are those with more than 79 Ifiercent of respondents' agreement, or less than 21 percent C>ff‘respondents' disagreement. These items tend to inflate 1She coefficient of reproducibility due to their consensual vVeight on the total scale. 68 The varimax-rotated factor analytical solution of the same nine dichotomous items seems to indicate the existence of twp independent dimensions in our measure- ment of patron-dependent relationships. This factor analytic evidence suggests the existence of two separate dimensions; one dealing with the decision—making structure in the nuclear-family group, and the other related to the decision making structure of the non-nuclear (extended) family. Finally, the Guttman scalogram analysis of the seven surviving PD items presents somewhat ambiguous evidence as tOIHndimensionality. The case rests on whether a quasi-scale coefficient of reproducibility is or is not enough evidence for unidimensionality. Nevertheless, on the basis that (1) conceptually, our definition of patron- dependence is not limited to strictly nuclear or extended family situations, (2) the factor loadings of the eight surviving PD items to the first factor of the unrotated principal factor solution are of fair (and nearly equal) magnitude,* (3) the marginal distribution of respondents agreeing with the items do not show either "consensual items" or the existence of items contributing less than five percent of the total variance (which might inflate the coefficient of reproducibility), and (4) we consider the present Operationalization as only a first attempt to empirically measure the concept, we decided to add the *The same rationale for using a si gle index for a conce t was used by Bordenave 0966, p. 104 under similar Circums ances. 69 :seven surviving items into a single PD index. The range sand the bell-shape of the unimodal distribution of PD :scores, shown in Table III-7, can be interpreted as (lesirable qualities of our PD index. Neverthelessfhture :research might attempt to empirically determine whether there is indeed a nuclear—family PD index and a non— nuclear family PD index. It is ouite possible that these two subdimensions were suggested by the present analyses only (1) because of the nature of the present respondents, or (2) as an artifactcfi‘the ten items that were originally included in our PD index. Control Variables As was said in Chapter II, individuals with higher rank and greater participation in the social system should be among those who conform to the norms of the system. Therefore, indicators of socio-economic status zand social participation should be controlled, i.e., the 'variance that they produce in PD is statiscally kept constant, in order 'M>have a "purer" patron-dependence effect on the relationship between extra-system communi— cation and modernity. thatus Statgs denotes position in a social system. Socio-economic status is the possession of physical Objects, that put an individual in a higher or lower 70 Tabls III-7. Frequency Distribution of the PDR Index (0"? o Score Frequency* Percentage (N=315) 0 -6 1.9 l 26 8.3 2 56 17.8 3 68 21.6 4 82 26.0 5 41 13.0 6 19 6.0 7 17 5.4 100.0 *The mode and median equal 4, the mean is 3.5 and the standard deviation is 1.6. 71 'position in relationship to his peers. Respondents were Eisked about a set of seven household items (water filter, Inlumbing, electricity, radio, inside bathroom, motorized ‘vehicle, and house in town)* that, aftereddition, con— stitute what from now on is treated as SES index. These items require economic capital to acquire them, and since these are not new to the studied communities, we perceive them closer to a measure of SES than home innovativeness. Social Integration Social integration is different from participat- ing in the system. Respondents were asked about the cuality and quantity of assistanceihat each one gave to and received from their peers in the community in terms of labor, money, tools and animals, and services in case of illness. Each respondent provided his annual frequencies (given and received) in each of the four subject-matters. Assuming that these items belong to the same conceptual universe of items, the author submitted them to a Guttman scalogram analysis (shown in Table III-8).** For each item, answers were dichotomized for "none" and "some." The items about ‘receiving tools and services in case of illness had to be dropped in order to maintain a marginal spread larger *These items had proven discriminatory enough in previous phases of the Brazilian Diffusion Project. **These results are based on a random subsample of 59 subjects (roughly 20 percent of the original sample). 72 .mm. ma mpaaanaosoossms mo unmaoamwmoo mgsx 0.0m "whossm mo Ampszz Hopes 0.0 mm msmog Ammo ca sm>ac da< .m o.m mm mason :wmo as om>amomm pa< .m o.v mm sense as omeamomm wa¢ .v m.v 0m sense ca Qm>au oa< .m o.w sm maooe ca sm>fio oa< .m m.m so Ammv mmmnHHH mo memo Ga sm>ac pea .H *EmPH whossm was msammmmmom msaHmOm mpseosoawmm mEmpH was Hmdpsz mo mmmesmohmm Ammnzv meom scappsmmpsH Hmaoom .mIHHH wanes 73 'than five percentual points from item to item. The :remaining six items constitute a perfect scale with a (coefficient of reproducibility of .92 and the dichotomized zanswers were added to form a social integration index for each respondent. Dependent Variables For the purpose of the present dissermnion, modernity, is the expression of the modernization process at a given point in time, is measured through the con- cepts of agricultural knowledge and agricultural innovativeness. lggricultural Knowledge Agricultural knowledge is the possession of accurate information that enables the individual to make rational decisions about farming matters. This type of iknowledge was measured through a battery of three to five items for each of the four Phase 2.5 recommended agricul- tural practices.* Each one of these four practices re— ‘present the least ambiguous practice, in a 2x2 table cell along with the cost and complexity, out of 32 prac- tices being promoted by ACAR.** ‘k *ACAR experts prepared these "funnel-type" ques- tions (i.e., correct knowledge in one question implies correct knowledge in the preceding one) about "home— 'Dharmacy," "erosion-controlling folliage," "planting ‘machine," and "trench—silo." **Twenty technicians Q-sorted these 32 practices along cost and complexity. Stanfield 23 a1. (1967, pp. 6—7) 1"911‘01"t that the Spearman correlation between the mean rank for each practice on cost and complexity was .73. 74 .Agxdcultural Innovativeness Agricultural innovativeness is the degree to which —-—--—..— ariindividual is relatively earlier than others in his social system to adopt new ideas or practices (Rogers, 1962, p. ‘l59). But when innovativeness is eouated with modernity, it is assumed that innovativeness corresponds to rational behavior. Rational behavior in this case means to weigh possible alternatives and check for applicabilities before adoption, in orderihat future discontinuances do not be- come economically and socially expensive. Traditionally, there have been two main ways of measuring innovativeness. ‘Firstly, by completely ignoring the discontinuance phenomena and being concerned only with adoption behavior. Secondly, 'by penalizing discontinuance and being concerned only “nth practices or innovations presently used. Considering 'both methods' extremes, we prOpose a conciliatory third Inethod which consists in giving a weight of "one" 6 the :number of years of discontinued innovations and a weight of "two" to the time ofuse of innovations still being used at the time of the interview. Let us compare the three methods of measuring innovativeness by analyzing the hypothetical example ‘presented in Table III-9. The first part of the table {gives the time since adoption of innovations X, Y, and Z, “mile the second half of the table presents the different innovativeness scores that farmers A, B, C and D will receive by rrlethod #1 (disregarding discontinuances), method 75 .mQOfipm>ozzfl dossfipcoomfic mmpmoficzhxx .OHIHHH magma mmm AHmuoE mg» :fi mmemHsm> segue esp mcflssmwwe mmxmwsfi smnpo map #09 mm Hng mmv mk cospms an cmpmoo>cw Kmosfi exp mo moswflsm> cam mpfiHmppsmo one psopw mosses w pom: ON OH OH N m m m NH N OH N **m **m O OH O OH **N **m **m m v N N N O O ¢ *m% NM H% N % K moonpmz mCOepm>chH pmshmm an mmsoom mo Gospmoc< mmmzm>flpw>osQH woman whmm% acossmz HwGOflpwngEoo psmswmwfim an mmhoom mmmsm>flpm>ocsH so mHLEcxm .mIHHH mHQwB 76 #2 (considering only the number of years of practices presently being used), and method 53(or the prOposed weighting system). Farmer A has only adopted practice Z, two years ago. Farmers B, C, and D have all adopted innovation X (5 years ago), innovation Y (3 years ago), and innovation Z (2 years ago), the only difference among them being that farmer B has discontinued all the three innovations, while farmer C still uses only innovation Z and farmer D uses all the three innovations. Method #1 can only differentiate between farmer A and the other three farmers, the other three farmers receiving higher innovativeness scores than farmer A. Method #2 discriminates a little more than method #1, but method #2 places farmer B at the lower end of the innova- tiveness continuum, irregardless that he had at least tried all the three innovations. Method #2 also allocated the same intermediary score to farmers A and C, while C has tried all the three innovations and A only one. Finally, Inethod #3 treats again farmer A as the farmer with the least innovativeness of the four, sequentially followed by :farmers B, C, and D. In order to build an agricultural innovative- Iless index, the respondents' years of use of reforestation, Thermite control, ant—killer, controlled breeding, and tick c30ntrol* were standardized and multiplied by the \_ *An independent set of innovations from the four I15sed for measuring agricultural knowledge. 77 Table III—10. Measurements of Central Tendency and Variance of All the Variables in the Present Analysis. Index Mean gggggiign Physical Mobility Scores 11.5 23,8 Mass Media Exposure Scores 19.9 2.8 Cosmopolite Contact Scores 29.3 18.8 Empathy Scores 7 3.9 2.8 PDR Scores 3.4 1.6 Socio-Economic Status Scores 4.8 _ 4.3 Social Integration Scores 2.7 1.8 Agricultural Knowledge Scores 4.2 4.0 Agricultural Innovativeness Scores 32.3 9.4 78_ discontinuance factor. The addition of the five scores con- stituted the respordents' innovativeness index, whose measurements of central tendency and variance are reported on Table III-10. Reliability Reliability denotes the reproducibility of em- pirical results or, in other words, the degree of response stability. Stanfield (1968, pp. 162-166) calculated the test-retest reliability of the measures repeated in Phase II and Phase 2.5, measures taken six months apart with the same 215 respondents. Stanfield's reliability coefficients range from .92 for age, to .08 for an achievement motiva- tion item. From these coefficients, those pertinent to the present dissertation (see Table III-11) range from .77 for the SES index, to .15 for the empathy question about municipjo mayor. Due to lower educational levels and lesser conditionings for question—answering lower response stability is expected in less develOped societies than in more deve- loped societies. Van Es and Wilkening (1969) confirmed this expectation, but they also discovered reliability patterns across countries. In a recent study comparing the percentage distribution of response stability, mean and median values for variables used in U.S. investigations and variables 'from a Brazilian sample, van Be and Wilkening show that reliability decreases in both countries from demographic variables to measurements of contemporaneous behavior, to 79 indicators of past behavior, to evaluative items. Stanfield's list of reliability coefficients confirmed the van Es and Wiflsning findings. Among the reliability coefficients reported in Table III—ll, the highest ones are for socio—economic indicators (or demographic vari- ables), followed by measures of mass media exposure (pre- sent behavior), and indicators of adoptive behavior and patron-dependence (indicants of past behavior), and, then, by some other items on PDR and empathy (evaluative items). EMPIRICAL HYPOTHESES Once Operational definitions for all the vari- ables in the theoretical model have been provided,it is possible to convert theoretical hypotheses into empirical hypotheses. EH1: The number of annual trips to cities with mope_than 4QLCOO inhabitants are negatiyely related tgfioatrqn- dgpendence scores. EH : Mass media exposure scores are negatively relatgd to patron-depgpdence scores. EH ° Cosmqpplite contact scores are nggatively related to patron-dependence scores. EH4: Empathy scores are gggatively related to patron- dependence scores. EH3; The number of trips to cities with more than AOJCOO inhabitants is positive1y_related to agricultural gppwledge scores. Table III—ll. Phase 2.5 Items.* 80 Reliability of Some Phase II and _— ' _ Reliability Concept Question Wording Coefficient SES Sum of items .77 Total area "What is the total area of your farm?" .72 Total no. of cows "How many cows do you have?" .72 Mass Media "Have you read (or has Exppsure somebody read for you) newspapers or magazines lately? How many timesa month? .64 Contact with ACAR "How many times have you talked to the ACAR agent last year?" .58 Schooling "How many years did you attend school?" .58 Mass Media "How often do you listen Exposure to the radio?" .44 Empathy Sum of items .39 Innovativeness Percent of practices adOpted .36 Empathv "If you were the ACAR agent, what would you do?" .34 PDR "When you wife needs to buy clothes or medicines... she has to ask you first or she can buy and then tell you?" .33 Empathy "If you were the person in charge of a factory, what would you do?" .22 Continuedwpn page 81. *Based on Stanfield (1968). 81 Table III-11 continued. Concept Question Wording Reliability Coefficient PDR Empathy Empathy "Would you wish that your sons follow an occupation . . . chosen by you or chosen by themselves (diff. wordsin T1 and T2)" "If you were the President of Brazil, what would you do?" "If you were the mayor of this municipkx what would you do?" .21 .19 .15 EH5b: EH6a: EH 6 b: EH, EH7b: EH83: EHBb: EHga: EHgb: 82 The number of annual trips to cities with more than 40,009 inhabitants is ppsitively related with agricultural innovativeness scores. Mass media exposure scorgs are_postively related to agpicultural knowledge scores. Mass media exposure scores are positively related to agricultural innovativeness scores. CosmOpolite contact scores areppositively related to agricultural knowledge scores. Cosmppolite contact scores are ppsitively related to agricultural innovativeness scores. Empathy scores are positively related to agricultural knowledge scores. Empathy scores are positively related to agricultural innovativeness scores. The number of annual trips to cities with more than 40,000 inhabitants has a higher relationship with agricultural knowledge scores among_individual§. with lowppatron-dependence scores than among individ- uals with high_patron-dependence scores. The number of annual trips to_gities with more tbsp 40,000 inhabitants has a higher relationship with agricultural innovativeness scores among individuals with low patron—dependence scores than among individ- uals with high_patron—dependence scores. EH EH EH EH EH 10b 11a: 11b: 12a: 12b: 83 Mass media exposure scores have a higher relation— ship with agricultural knowledge scores among individuals with low patron-dependence scores than among individuals with high patron-dependence scores. gpsmopolite contact scores have a higher relatippr ship with agricultural knowledge scores amOpg individuals with low patron-dgpendence scores than among individuals with highpatron—dependerce SCOTGS. Cosmppolite contact scores have a higher relation— ship with agricultural innovativeness scores among individuals with low patron-dependence scores than among individuals with high patron-dependence scores. Empathy_scores have a higher relationship with agricultural knowledge scores among individuals with low patron-dependence scores than among individ— pals with high patron-dependence scores. Empathy scores have a higp§p_relationshippwith agricultural innovativeness scores among individ- pals with low patron-dependence scores than amppg find i flayelejziihlh 1 sh pa tree-:6 199253.929 .62. _s_C..0._1:.e§ - 85 Nevertheless, since Hays is talking about correlations in general, one has to be more restrictive when using a correlation statistic for descriptive and for hypothesis-testing purposes, as we are. The Pearson product—moment correlation, being a linear model, requires linearity, and an interval level of measurement (Siegel, 1956, p. 30), and that the two correlated variable, X and Y, be at least Operationally independent. Scatter-plots were performed for all the vari- ables in the model to be tested in association with patron-dependence.‘ None of the tests for curvilinearity* *Using a non-lineal model (also advocated by Whiting, 1967, and Herzeg, 1967b) Waisanen and Kumata (1969) found it theoretically fruitful to plot years of education versus several indicators of modernity and to check for what they called the "take-off effect." Parallel to the previous tests for curvilinearity, we also divided our communication variables along standard deviation units and found the patron-dependence mean for each of these sub-groups. The plotting of these means allowed us a quick check on the take—off effect for decreases in patron- dependent relationships. The curves for physical mobility and mass media exposure did not follow the "S-shape" of the take-off pattern. 0n the other hand, the curves for cosmopolite contacts and empathy did show a certain take— off effect. Early shifts in reference—groups from the community to the outside world produce larger decreases in PD than later on when the individual has more contact with cosmOpolites. Yet, for empathy there are the individuals ranking high in the empathy index (let us say between 8 and 10) i.e., those able to assume roles far away from their local system like being "President of Brazil", that, on the average, account for decreases in PD away from linearity. Under the present interpretation, contact with compololites and empathy present theoretical consistency with Coutu (1951) conclusion that role—taking must precede role-playing. 84 STATISTICAL ANALYSIS The next chapter will present the hypotheses testing procedures. It will be apprOpriate to finish the present methodological chapter with a note on the statistical methods to be applied in these testing proce— dures. The Operationalization of the variables show, that we are dealing with continuous variables and the previous hypotheses are about the covariance between two or more of these continuous variables. Therefore, the Pearson product-moment correlation coefficient (zero-order and partial correlations) will be the statistic used to test Theoretic Hypotheses #1 through #8 and Fisher's z to test Theoretic Hypotheses #9 through #12. that are some of the underlying asmmnptions that should be kept in mind in performing such cor— relational analysis? Hays (1963, p. 510) says: It is not necessary to make any assump— tions at all about the form of the distribution, the variability of Y scores within X columns or'arrays,'or the true level of measurement represent- ed by the scores in order to employ linear (regression and) correlation indices to describe a given set of data. So long as thenaare N distinct cases, each having two numerical scores, X and Y, then the descriptive statistics of correlation (and regression) may be used. In so doing, we describe the data as though a linear rule were to be usEd for prediction.* *Parentheses provided by us. 86 (i.e., for significant differences between eta and the Pearsonian coefficient) were significant at the not- very-conservative .01 level* Also, during the process of index construction, intervality was achieved by standardiza— tions and additions of scale-forming items. However, even if in some instances intervality was not reached, this assumption is only relevant when dealing with relatively small sample sizes. The computing procedures for the Spearman rank correlation and the Pearson product-moment correlation are identical for samples larger than 50 observations, as in our case. Basically speaking, our data are approximately at the interval level of measurement, we have a reason- ably large sample (N=315), the variables were measured independently, and they associate non-curvilinearly with the other variables in the model. Therefore, the assump— tions for using the Pearsonian correlation coefficient in hypotheses—testing—procedures are met by the present study. *In this particular case the .01 level is less conservative than the, for instance, .05 level because, testing if eta, (the curnilinear solution) significantly explains a better association or more variance than the linear solution (or product—moment correlation) then a larger accepted value for the statistic "F" represents a less restrict test on the null hypotheses that eta is enual to the Pearsonian—r. patror patron respon commur obtain 87 The sample was stratified at the median of the patron-dependence scores allowing subsamples of 159 high patron-dependence respondents and 156 low patron—dependence respondents. Zero—order correlations between extra-system communication variables and modernity variables were obtained for these high and low patron-dependency subsamples. The Michigan State University CDC 3600 computer facilitated the analysis of the data. Associations are tested through zero—order correlations (product moment) and partial correlation analyses. A relationship was con- sidered as evidence in support of the hypotheses when the statistical test indicated that the association could be due to chance only in five percent or less of the cases. Table III-12 shows the statistical method that was utilized to test each of the empirical hypotheses. With theoretic Hypotheses #1 through #8 where zero—order and partial correlations could be used for testing procedures, we Opted for the highest-order partial cor- relation because it is a purer indicant of an associa— tion than the zero-order correlation. Besides, not knowing the degree of conceptual overlapping among the extra-system communication variables, we considered it theoretically preferable to test the associations of each of the communication variables free of the effect of the other three communication variables. In the future, when Table I Theore: Hypothe I. II. PD H Table III—l2. 88 Statistical Methods for Testing the Hypotheses. Theoretical Empirical Hypotheses Hypotheses Statistical Method I. PD as the Dependent Variable THl EHl Third-order partial correlation TH2 EH2 Third-order partial correlation TH3 EH3 Third-order partial correlation TH4 EH4 Third-order partial correlation II. Modernity as the Dependent Variable TH5 EHSa Third-order partial correlation EH5b Third-order partial correlation TH6 EH6a Third-order partial correlation EH6b Third-order partial correlation TH7 EH7a Third—order partial correlation EH7b Third-order partial correlation TH8 EH8a Third-order partial correlation EHBb Third—order partial correlation III. PD as suppressor Variable TH9 ' EHga Fisher!s ztest EHgb Fisher's ztest TH10 EH10a Fisher's satest EHlOb Fisher's ztest TH11 EHlla Fisher's zntest mllb Fisher's ztest TH12 EHl2a Fisher's ztest EH Fisher's ztest 12b meanir the q: of thi syster consid thesis was s: direCT theses corres to e: in 0rd for ti: 89 the qualitative and quantitative contributions of each of these variables have been well mapped, it might be meaningful to try to measure the degree of overall extra- system communication. 80, in conclusion, the decision criteria for considering empirical support toward a particular hypo- thesis was met if the highest-order partial correlation was significant at the .05 level, one tail-test due to the directionality of the hypotheses. For Theoretic Hypo- theses #5 through #12, where each theoretical hypothesis corresponds to two empirical hypotheses, it was necessary to determine significance for both empirical hypotheses yin order to consider whether sufficient support was present for the theoretic hypothesis. patFOr SYSfer (3953 a: Chapter IV PRESENTATION OF THE FINDINGS Through prolonged closecontact and friction with the objects of their study, the minds of experts finally acquire a pictoral, mothlike, fiddling perfection. (Sterne, in Jantsoh, 1967) Chapter III presented the index construction of the variables dealing with extra-system communication, patron-dependence, and modernity. The present chapter summarizes the research findings specifically dealing with (l) the type of relationship between extra—system communication and the intervening and dependent variables (PD and innovativeness), and (2) the suppressor effect of patron-dependence on the relationship between extra- system communication and the indicants of modernity. WHO ARE THE PATRON-DEPENDENTS? Before starting the hypothesis-testing procedures, let us summarize some of the differences among individuals scoring high and low in the patron-dependence scale. Means were obtained along several dimensions (see Table IV—l) for the highest and lowest deciles on the PD scale.* * *"Low PD" corresponds to a score of one or less, and "High PD" corresponds to a score of six or more, or, in each case, ap roximately ten percent of the total sample (see Table III-7 . 90 Table Indepe F" C) o 1 ‘4" £1) 0') UT?! 91 Table IV—l. Characteristics of High and Low PD Individuals Means on Independent Variables Low Entire High t* PD Sample PD Respondents Means Respondents Independent Variables (N=32) (N=315) (N236) 1. Years of Age 43 44 A4 0.29 2. Annual Trips to Large Cities 16 12 10 0.99 3. Years of Education 3.1 2.3 2.0 2.21 4. Number of Contacts with ACAR in Past Years 11 7 4 2.24 5. Functional Literacy . Scores** 41 30 21 4.19 6. Farm Size in y Hectares*** 78 51 38 1.17 7. Number of Cows ' Owned 20 12 7 2.32 8. Agricultural Know- ledge Scores (0-16) 7 4 2 4.91 9. Socio Economic Status , Scores (0-7) . 7 5 4 2.81 10. Opinion Leadershi Scores (0.00-1.00 *”“* 0.16 0.06 0.01 4.40 11. CosmOpolite Contact Scores (0-240) 33 29 20 2.61 12. Mass Media Exposure Scores (0-40) 21.6 19.9 20.0 1.98 13. Agricultural Innova— tiveness Scores (0-99) 35 32 27 4.21 *Underlined t values (for the difference between the means of low and high PD respondents) are significant at the .05 level, one-tail test, with 66 degrees of freedom. **Number of correct words read out of a standard paragraph containing 50 words. ***2,2 acres =1.hectare or 10,000 square meters. ****Total number of nominations received, divided by the number of interviewees in the respondent's community. 92 Within our sample of Minas Gerais farmers, patron- dependent individuals are those with less education, with less contact with ACAR specialists, with less functional literacy, and fewer cows, with less agricultural knowledge and innovativeness, with lower socio—economic status and Opinion leadership, and with fewer cosmopolite contacts and mass media exposure. 0n the other hand, age, trips to large cities, and farm size do not differentiate between individ- uals with high and low patron-dependence. So, for the purpose of our sample of farmers, we generalize that the uneducated, the socially isolated, the economically powerless, and the laggards will be among those that will rely more often on the decision—making abilities of those occupying superior positions in the system. HYPOTHESES TESTING Basically, the first eight theoretical hypo- theses deal with the types of relationships between the extra system communication variables and patron-dependence (for the first four hypotheses), and agricultural know- ledge and agricultural innovativeness (for the next four hypotheses). It is predicted that extra—system communica- tion scores are negatively related with PD scores, and positively related with the two indicators of modernity. Table duct in t} the p; are s: coeff; syster by the 93 Table IV-2 provides a correlation matrix (zero-order pro- duct moment correlation coefficients) of the nine variables in the model. It shows that all the correlations are in the pgedicted direction. However, in order to test if the relationships are significantly different from zero, partial correlation coefficients were obtained in which for each of the extra- system communication variables, the variance contributed by the other three antecedent variables was statistically removed. The corresponding zero-order and third-order partial correlation coefficients between each of the antecedent variables and the intervening and consequent variables are shown in Table IV-3. The testing of the hypotheses follows. PD as the Dependent Variable THl: Physical mobility is negatively related to patron-dependence. EH1: The number of annual trips to cities with more than 40,000 inhabitants is pegatively_related to_patron—dependence §£2££§~ The zero-order correlation between the number of trips to cities and patron—dependence scores is -.05, which is less than the -.10 required for significance at the .05 level (Table IV-2). Tablr Iari: Varia 1. MC an Q; r VUC TESpO at t}. COTre 94 Table IV—2. Zero-Order Correlation Matrix of all the Variables in the Present Analysis (N=315). Variables l 2 3 4 5 6 7 8 1. Mobility 2. Agricultural .01 Knowledge 3. Socio-Economic Status .05 .39* 4. Empathy .02 .41 .31 5. Social Integration .10 :29 .09 .28 6. Cosmopolite .07 .29 .14 .29 .21 Contact 7. Mass Media Exposure .20 .42 .62 . 8 .22 .21 8. Innovativeness .10 .48 .43 .22 .36 .20 .37 9. PDR -005 -o32 -020 -011 -021 ““017 -018 -020 *For a sample size equal to or larger than 300 respondents, a correlation higher than .10 is significant at the .05 level of confidence, one tail—test. Significant correlations have been underlined. mass 7 is ~,l contac refiuir EH2 isl 95 The third-order partial correktion between the number of trips to cities and patron-dependence is-201, when controlling on mass media exposure, cosmopolite con- tacts, and empathy, which is less than the —.10 required for significance at the .05 level (Table IV—3). EHl is not supported. Therefore, THl is not supported. TH2: Mass media exposure is negatively related to patron-dependence. EH2: Mass media eXposure scores are negatively related to patron-dependence scores. The zero—order correlation between mass media exposure scores and patron—dependence scores is -.18, which is greater that the -.10 required for significance at the .05 level (Table IV—3). The third-order partial correlation between mass media exposure scores and patron dependence scores is -.13, when controlling on physical mobility, cosmopolite contacts, and empathy, whiCh is greater than the -.10 required for the significance at the .05 level (Table IV-3). EH2 is supported. Therefore, TH2 is supported. TH3: Cosmopolite contapt is negatively related to patron-dependence. EH3: CosmOpolite contagt_scores are negatively related to_patron-dependence scores. Eflwnwlmknmllflwa UZXNN Hth Cg”? fiver“, mzo H. ##v. H WurFLAMC x Fink. T .LVIL. EFL. {fitmxln C'erxfi .F S»? Warn 7% (a Qh\ahrfl\||Ab.L Psehvu I N ||\H N. «UV Nah Rhux.‘ .pmssnsoo mmHQstp> psmosmampSs adow exp an meanwssm> pampSmsmp esp mo Scam as pmsHmmem mousshs> fleece mzexxx .pmsaasmcsa ammo. were: manmamssoo pswosMssmfim .emmplaflwp mac .mQSmpsmsoo mo Hm>ma mo. one pm espoamfismsm ms 0H. sage spams: Soapmampsoo w .mesmpsoamms ooom sane pmmsma so 0p HmSww musm panama m somxx .mmanwfism> soamefiQSEEoo Empmmmlmppxm omens soapo esp sow msflaaospsoox c .... c \ Ox . . .5... a tee ewe em eexee masses 5 6 G; _ so. mm. em. as. No.1 Ha.n savages .e a. .om. ss. mm. ma.u es.u neoepeoo epeaoeosnoo .m .QMu .mmq Hm. we. ma.u *xwa.u anemones eases ewes .m mo. awe wo.u so. Ho.n mo.u speaepez Heoanssm .H *smcsc Amuse xsmpso posse *hmpso sopso unease noeeN lessee nonem sneeze noses mmmzo>spm>ossH mmcmasosz - HMMSPHSoesm< amazeaaossm< mm .Amamnzv mmmsm>spm>ossH HmQSPHSOasma was .mmpmasosx HmHSPHSoflsm< .mm was .sOspmossSEEoo Ememhmlmspxm cmmspmn msospwamssoo *Hwapswm smUHOIpAsne was smpHOIoth .mu>H eases 97 The zero-order correlation between cosmopolite contact scores and patron—dependence scores is -.l7, which is greater than the -JO required for significance at the .05 level (Table IV-3). The third-order partial correlation between cosmOpolite contact scores and patron-dependence scores is-al3 when controlling on physical mobility, mass media exposure, and empathy, which is greater than the -.10 required for significance at the .05 level (Table IV-3). EH3 is supported. Therefore, TH3 is supported. Th4: Empathy is negatively related to patron— dependence. EH4: Empathy scores are negatively related to patron-dependence scores. The zero-order correlation between empathy scores and patron-dependence scores is -.11, which is greater than the -.10 required for significance at the .05 level (Table IV-3). The third-order partial correlation between empathy scores and patron—dependence scores is -.02)when controlling on physical mobility, mass media exposure, and cosmOpolite contact, which is less than the -.10 required for significance at the .05 level (Table IV-3).EH4 is not supported. 98 Therefore, TH4 is not supported. ***** In general terms, the overall prediction that extra-system communication is negatively related to patron- dependence seems to be confirmed by the data. All the zero- order and third-order partial correlations between the extra— system communication variables and PD (in Table IV-3) are in the predicted negative direction. And with the exception of physical mobility, all the zero—order correlations between extra-system communication and PD reach statistical signifi- cance. Nevertheless, in terms of social significance, these correlations are not unusually high and extra-system com— munication as a whole accounts for only 5 percent of the variance in patron-dependence. (Table IV-3). Modernity as the Dependent Variable TH5: Physical mobility ispositively related to modernity. EH§§ The number of annual trips to cities with more than 40,000 inhabitants is positively {elated to agricultural knowledge scores. The zero—order correlation between the number of trips to cities and agricultural knowledge scores is .01, which is less than the .10 required for significance at the '05 level (Table IV-3). The third-order partial correlation between the 1'lumber of trips to cities and agricultural knowledge scores 99 is -.08.when controlling on mass media exposure, cos- mOpolite contact, and empathy, which is less than the .10 required for significance at the .05 level (Table IV— 3). EH is not supported. 5a EH b: The number of trips to cities with more 5 than 40,000 inhabitants is positively related to agricultural innovativeness scores. The zero-order correlation between the number of trips to cities and agricultural innovativeness scores is .10, which is equal to the .10 required for significance at the .05 level (Table IV-3). The third-order partial correlation between the number of trips to cities and agricultural innovativeness scores is .03 when controlling on mass media exposure, cosmopolite contact, and empathy, which is less than the .10 required for significance at the .05 level (Table IV- 3). EH5b is not supported. Therefore TH5 is not supported. TH6 : Mass media exposure is positively related tg_modernity. EH6a: Mass media exposure scores are positively related to agricultural knowledge scores. The zero-order correlation between mass media exposure scores and agricultural knowledge scores is .42, “*HiCh.is greater than the .10 required for significance at the .05 level (Table IV-3). 100 The third-order partial correlation between Inass media exposure scores and agricultural knowledge sscores is .31 when controlling on physical mobility, (cosmOpolite contact, and empathy, which is greater than ‘the .10 required for significance at the .05 level EH6a is supported. EHsb: Mass media exposure scores are postively related to agricultural innovativeness scores. . The zero-order correlation between mass media exposure scores and agricultural innovativeness scores is .37, which is greater than the .10 required for significance at the .05 level (Table IV-3). The third-order partial correlation between mass rnedia exposure scores and agricultural innovativeness scores is .30 when controlling on physical mobility, cos- rnOpolite contact, and empathy, which is greater than the .10 required for significance at the .05 level ( Table IV- 3). EH6b is supported. ' Therefore, TH6 is supported. TH7 : ggsmOpolite contact is positively related to modernity. EH73: Cosmopolite contact scores are positively related with agricultural knowledgg_scores. The zero-order correlation between cosmopolite (lomtact scores and agricultural knowledge scores is .29, 101 *which is greater than the .10 required for significance :at the .05 level (Table IV-3). The third-order partial correlation between cos- rnopolite contact scores and agricultural innovativeness scores is 17 when controlling on physical mobility, mass Inedia exposure, and empathy, which is greater than the .10 required for significance at the .05 levéL (Table IV-3). EH is supported. 7a Therefore TH7 is supported. TH8 : Empathy is positively related to modernity. EHBa: Empathy scores are pqsitively related to agricultural knowledge scores. The zero-order correlation between empathy scores and agricultural knowledge scores is .41, which is greater than the .10 required for significance at the .05 level (Table IV-3). The third order partial correlation between empathy scores and agricultural knowledge scores is .26 when controlling on physical mobflity, mass media exposure, and cosmOpolite contact, which is greater than the .10 required for significance at the .05 level. EHBa is supported. EH8b: Empathy scores are positively related to agricultural innovativeness scores. The zero—order correlation between empathy scores Eind agricultural innovativeness scores is .22, which is égreater than the .10 required for significance at the .05 ilevel (Table IV-3). 102 The third-order partial correlation between enmmthy scores and agricultural innovativeness scores :is .06 when controlling on physical mobility, mass media (exposure, and cosmOpolite contact, which is less than the .10 required for significance at the .05 level (Table IV—3). EHBb is not supported. Therefore, TH8 is not supported. Generally speaking, the broad prediction that extra—system communication isflpositively related with modernity seems to be supported by the present analysis. Table IV-3 shows that all the eight zero-order correla- tions between the four extra—systemic communication vari- ables and the two indicants of modernity are in the expec- ted positive direction. Seven of these eight correlations are statistically significant, and, in terms of social significance, these medium-size correlations account for more than one fourth of the variance in agricultural know- ledge (28 percent) and about one sixth of the variance in agricultural innovativeness (16 percent). PD as a Suppressor Variable In Chapter II it was predicted that extra- system communication would correlate negatively with patron— dependence, which would also correlate negatively with Inodernity. The present section will analyze first the second part of the afore—mentioned association which is not 103 converted into a theoretical hypothesis, and then test the suppressor role of patron—dependence on the relation- ship of extra-system communication to modernity. The present chapter has been dealing with the two sides of a triangle, i.e., the relationships between extra-system communication and PD, and the relationships .between extra-system communication and modernity. What about the relationship between PD and modernity, the third side of our triangle? Table IV—4 shows that, as predicted, the associa- tion is indeed in the negative direction, which means that individuals scoring higher in the PD scale tend to have less knowledge about agricultural innovations and to have adopted fewer (or done so more recently) agricultural innovations. This negative relationship between patron- dependence and the indicators of modernity is solid enough to survive after controls for socio—economic status and social integration were introduced in the analysis. The last four theoretical hypotheses deal with the suppressor effect that patron-dependence is expected to have in the extra system communication variables and the variables measuring modernity. It was hypothesized that the extra-system communication variables will have higher relationships with the modernity indicants among low patron-dependence individuals than among high patron- dependence individuals. The sample was stratified at the 104 Table TV-4. Zero—Order and Partial Correlations Petween Patron—Dependence and Agricultural Knowledge and Agricultu- ral Innovativeness (N=315). Patron—Dependence Zero—Order First—Frder Partials 1* II** Agricultural Knowledge -.32*** :ng _.27 Agricultural Innovativeness -.2O —.l4 -.l3 *Controlling for social integration. **Controlling for socio-economic status. *** For a sample size equal to or larger than 300 respondents, a correlation higher than .10 is significant at the .05 level, one tail-test. Significant correlations are underlined. 105 median of the patron—dependence scores and zero-order correlations between extra-system communication and modern- ity were obtained for the high and low patron-dependency subsamples. Table IV-S presents the correlations between the extra-system communication variables and the modernity variables for the two sub—samples. Fisher's 2 was utilized to test the significance of the difference between two comparable correlations for the high and the low patron- dependents.* Support for both empiric hypotheses was requir— ed for support of any of the theoretical hypotheses. TH9: Physical mobility has a higher relation- ship with modernity among low patron- dependence individuals than among high patron-dependence individuals. THga: The number of annual trips to cities with more than 40,000 inhabitants has a higher relationship with agricultural know- lgdge scores among individuals with low patron-dependence scores than among igggyjduals with high_patron-depegdenca SCOI‘GS . . *The formula for computing the z score for the difference between the two 2 scores, representing the two COITelations is: - 1 f Nl-3 N2_3 ThiS z score for difference must be greater than 1.65 to be significant at the .05 level. 106 The correlation between the number of trips to cities and agricultural knowledge scores is -.10 for the high PD category and .08 for the low PD category (Table IV- 5). The difference between these two correlation coefficients is in the predicted direction, from negative to positive. Fisher's z is 1.59 which is less than the 1.65 required for significance at the .05 level. EHgais not supported. Equ: The number of annual trips to cities 4 with more than 40,000 inhabitants has a higher relationship with agricultural innovativeness scores among individuals with low patron-dependence scores than among individuals with high patron- dependence scores. The correlation between the number of trips to cities and agricultural innovativeness scores is .07 for the high PD group and .14 for the low PD group (Table IV—5). The difference between these two correlation coefficient is in the predicted direction, and Fisher's z is .66 which is less than the 1.65 required for significance at the .05 level. EHgbis not supported. Therefore, TH9 is not supported. THlO: Mass media exposure has a higher relation— ship with modernity_among low patron- dependence individuals than among high patron—dependencegindiyigpals. 107 .pmmelaflmp mco Sm Mom poo Hm>mH m.o map pm pamOflcflawflm ma mm.a nag» pmgmfls to 0p Hmsom N m.gmgmfim axx .umnfiapmcss Comp m>w£ usofipwamppoo pnwcfiwfinmfim .Hm>ma mo. mgp pm unwofl%flzmfim we ma. none hmsmfis Soapmawhpoo w .mpcmcnommoh OmH swap pmmhma so 0» HwSUm msflm mapfimm m pom .wmapewmlnzw cap me smegma am one macaw vmflMprspm mm; mpommDSw mam mo mHQEwm Hmnflmepo onyx xxm>.m be. 0m. mm. mv. mm. zgpmmsm .v xxmm.m mo. 0m. xxom.a 0m. 0m. pompzoo mpflaoaofimoc .m *xmv.m mmq Hm. xxma.¢ mm. ea. mnsmocxm . deems mmmz .m. mm. «A4 so. mm.a mo. oa.u ssfiaflpos Hmoflwegm .H N mm mm N am pm haemae aog zmflm pmgmfim goq swam mmem>pr>oczH mmcwaaosx awhSpHSOHhm< HmpSpHSOflhm< .mpsmHOflmmmoo Soapmamhpoo 039 mgp smmspmn monLmMMem map pom N m.gm£mflb cam .* emanzv mpsmwnmpmm Cchpmm Bog cam Ammanzv mpCccchmmlcOppwm swam hog whopmoeccH pechmcoz exp Ape; mmapmflpm> SoaprficseEoo Empmhmldhpxm mSp mo m20fipmamhhoo thAOIOamN .mI>H mamas 103 EHlOa: Mass media exposure scores have a higher relationship with agricultural knowledge scores among individuals with low patron— dependence scores than among individuals with high patron-dependence scores. The correlation between mass media exposure scores and agricultural knowledge scores is .17 for the high PD group and .56 for the low PD group. The difference between these two correlation coefficients is in the predicted direction, and Fisher's z is 4.19 which is greater than the 1.65 required for significance at the .05 leveL EH is 10a supported. WHlOb: Mass media exppsure scores have a higher relationship with agricultural innovative- ness scores amone individuals with low patron-dependence scores than among individ- uals with high patron—depencence scores. The correlation between mass media exposure scores and agricultural innovativeness scores is .21 for the high PD group and .53 for the low PD group. The difference be- tween these two correlation coefficients is in the predicted direction, and Fisher's z is 3.43, which is greater than the 1.65 required for significance at the .05 level. FHlOb is supported. Therefore, THlO is supported. 109 THll: Cosmopolite contact has a higher relation- ship with modergity amOpgglow patron- dependence individuals than among high patron-dependence individuals. EHlla' Cosmopolite contact scores have a higher relationship with agricultural knowledge scores among individuals with low patron- dependence scores than among individuals with high patron-dependence scores. The correlation between cosmOpolite contact scores and agricultural knowledge scores is .39 for the high PD group and .20 for the low PD group. The difference between these two correlation coefficients is not in the predicted direction, and Fisher's z is 1.90, which is greater than the 1.65 required for significance at the .05 level. EH 11a is not supported. EHllb: CosmOpolite contact scores have a higher . relationship with agricultural innovative- ness scores among individuals with low patron-dependence scores than among individuals with high_patron—dependengg was. - The correlation between cosmOpolite contact scores and agricultural innovativeness scores is .30 for the high PD group and .06 for the low PD group. The difference 110 between these two correlation coefficients is not in the predicted direction, and Fisher's z is 2.27, which is greater than 1.65 required for significance at the .05 level. EH is not supperted. 11b Therefore, THll is not supported. THlpz Empathy has a higher relationship with modernity among low patron—dependence individuals than among high patron— dependence individuals. EH Empathy scores have a higher relation- 12a: ship with agricultural knowledge scores among individuals with low_patron- dependence scores than among individuals with high patron-dependence scores. The correlation between empathy scores and agricultural knowledge scores is .35 for the high PD group and .43 for the low PD group. The difference between these two correlation coefficients is in the predicted direction, and Fisher's z is 0.86, which is less than the 1.65 required for significance at the .05 level. EH12a is not supported. EH Empathy scores have a higher relationship 12b' path agricultural, innovativeness scores among individuals with low patron— dependence scores_than among individuals with high_patron—dependence_§pgres. ...—...- -.--y. 111 The correlation between empathy scores and agricultural innovativeness scores is .20 for the high PD group and .47 for the low PD group. The difference between these two correlation coefficients is in the predicted direction, and Fisher's z is 2.79, which is greater than the 1.65 required for significance at the .05 levd. EH12b is supported. Therefore TH12 is not supported. The general prediction that patron—dependence has a suppressor effect in the relationship between extra- sgstem communication and modernit: seems to be supported in light of the present data. While half of the Fisher's 2 scores reach the significance criterion, the general tendency is that the relationship between the extra-system communication vari— ables (withthe exception of cosmOpolite contact) and the two indicants of modernity is, as predicted)higher for the low PD individuals than for the high PD individuals. Chapter V CONCLUSIONS AND RECOMMENDATIONS We must instill in our students the expecta— tion of tedium and disappointment and the duty of thorough persistence, by now so well achieved in the biological and physical sciences. We must expand our student's vow of poverty to include not only the willing- ness to accept poverty of finances, but also a poverty of experimental results. (Campbell and Stanley, 1963) SUMMARY AND DISCUSSION The present study focused upon the intervening effects of patron-dependence relationships in the communica- tion and modernization process. The main objectives were: (1) to define patron-dependence, conceptually and Operational-1 1y; (2) to analyze empirically the relationship of com- munication variables with PD (patron-dependence) and modern- ity, and (3) to determine the possible intervening effects of PD of the associatidn between communication variables and modernity. Patron-dependence was defined as the degree to which an individual's decisions are influenced by actors occupying superior hierarchical positions in the social . _. * svstem. PD was measured With a seven item scale. Our patron-dependence index items seem to be oriented toward the nuclear family. Out of seven items finally retained in the scale, four items deal with situations involving the *Typical PD items are "not to allow daughters to date without a chaperone" and "to prefer'to hire always a relative on respondent's farm." 112 113 " _ ‘- nuclear-family. These nuclear family items do not cor- relate very highly with the non-nuclear family items, per- haps suggesting that with the inclusion in the future of a larger number of items, two or more sub—scales might emerge. Nevertheless PD (as measured in the present study with a sample of Brazilian farmers) had a reasonable range and the unimodal distribution approximated a bell-shaped distribu— tion. The retained PD items also have equal loadings of a magnitude of .35 to .57 on the first factor of the (un- rotated) principal factor solutiOn, and item to total score- minus-the-item (positive) correlations that range from .18 to .31. Thus, the seven items in the PD index demonstrate a modest degree of common variance. It was expected: (1) that extra-system communica- tion variables while positively associated with modernity variables are negatively associated with patron-dependence, and (2) that stratifying along patron-dependence will in— crease the relationship between extra-system communication variables and modernity variables. Physical mobility, mass media exposure, cosmOpolite contact, and empathy were ex— pected to be negatively related to PD. These four extra- system communication variables were expected to be positive- ly related toihe modernity variables of agricultural know- ledge and agricultural innovativeness. Lastly, we expected that among low PD individuals it would be a higher relation- ship between the extra-system communication variables and modernity variables then among high PD individuals. 114 The data for the present study come from part of a larger reserach endeavor dealing with the diffusion of innovations in Brazil. The presert dissertation con- centrates on interviews with a sample of 315 Minas Gerais farmers, elsewhere called Phase 2.5 of the Brazil Diffusion Project. Table V-l shows a summary of the hypotheses- testing procedures. Half ofihe theoretic hypotheses were supported and half were not supported. Nevertheless, with one exception, the direction of all the relationships in the model supports our theoretical expectatives. There are two ways to explain these results: methodologically and theoretically. We shall now present methodological and theoretic reasons supporting the behavior of each of the independent variables. 1. Physical Mobility - The data on physical mobil- ity do not present statistical support for any of the theoretical hypotheses (Table V-l). Nevertheless, all the ' relationships with physical mobility are in the predicted direction. Physical mobility was measured with a single direct question. Physical mobility was Operationally de— fined as the number of trips to cities with more than 40,000 inhabitants. Such urban centers may be too far removed from the reference system of our respondents. At the time of the interviews, 1967, there were only four cities of such size in Minas Gerais. It could be that these cities were o: ...—..-- “ copsoagam poz .>Hpm>ossH nmmm Umpsoaazm pox cmpsoggsm .mwmasosm wwmm aspwasm mme wmpsomazm .>pr>onH nsmm copycaQSm UmpsomQSm .mvmasosx puma pompsoo mpfiaomofimoo Pma copsogmsm .>fipw>ost nmmm cmesOQQSm UmpMOQQSm .mvmasosx wmmm whamoaxm deco: mam: ems popsopmsm poz .>flpm>ossH pmmm amppogazm poz ampnOQQSm poz .mamaaons mmmm spaaflnos amOflmsns mme . mapwfisw> mepCmme one wmiNeflssmpoz .HH 5. l . i . a 1lwmpsonc5m poz awesompSm poz mm vim >£ngam me cappopQSU awesompsm mm mmm powpcoo mpfiaopoawoo «as a cmpsopssm omppoaaam cm mmm msSmcsxa mficmz mmmz was amateggSm poz empnoamzm poz mm, Hmm spaaanos Hmoamsam axe . 2 m was Bangles data ......a HmOflpmsomxe Hmofisfipam mmapmwsm> mmmaspoamm pgmosoo mmmwgpoamm pflzmas mswpmmEImflmmnpcpam psmcsz¢m Hwoflsflssm enmcsmpmrsfi ospmpomge .sofipmpsmwmflm psmwmsm one 2a mpaammm msfipmmelmmmm£pommm one mo zMMEESm .HI> mapma . _ DNH . meMOQQSm .>va>ossH mm “coppopQSm poz sophomQSm poz .momasosx amflmm knewasm Name sophommdm poz .>flpm>ossH naamm 6 Hunmpsooezm poz cmPLOQRSm poz .MUmfisosM maamm powpsoo mpHHOQoEmoo dame cmpsopgSm _.>flpm>ocsH noamm ompsoaasm omphoapzm .mcmflsosx woamm madmopxm wflwmz mmmz came Umpsomazm poz .>apm>osQH pmmm empsomgzm poz amphegdSm poz .mamaaons mmmm spaaflnos amoawsnd mme manmfism> sommmAQQSm m mm as .HHH Hmoflpmsomzs Hwofisfiasm mmapwflsw> mommnponhm ppmosoo mommSpoomm pasmmm msfipmmEImammgpoamm pamuswmmm Hwoflsfiaem pamccmmmcsH oflpmsomse .amscapcoo Hu> magma 117 physically and perceptually so far from the farmers' communities that they traveled to them only in cases of extreme necessity, looking for certain specialized services not immediately available in the local, small- er centers. In these circumstances, the individual is so uncomfortable in the large cities that he does not look for instrumental experiences, or if he dies, they seem so intangibkathat their applicability to his life situation is strongly doubted. 2. Empathy - The data on empathy do not support any of the theoretical hypotheses but show support for two of the five empirical hypotheses, and all of the empirical relationships involving empathy are in the predicted direction. The frequency distribution of the empathy index is hardly bell—shaped, and this distribution may have affected the statistical results of the analysis. Previous comments (Chapter III) about the "take-off effect" of empathy in modernization imply that it is necessary to tap more roles at the national and international level. We believe that improvements in the operation would produce a more discriminating empathy index, and perhaps a more bell-shaped distirbu-' tion. Whiting (1967, p. 148) in a recent study with a similar sample of Brazilian farmers found that empathy, as a predictor of modern orientations and attitudes, was less important than mass media exposure, literacy, or general intelligence. Nevertheless, our findings suggest 118 that empathy is a more important predictor of the cognitive aspects of modernity than of the behavioral aspects of modernity. 3. CosmOpolite Contact - Now, let us explain the behavior of cosmOpolite contact in the present analysis. First, two of the three theoretic hypothesis dealing with cosmOpolite contact were supported and the three correspond- ing empirical hypotheses when tested yielded relationships in the expected direction. CosmOpolite contact, as pre- dicted, is positively related with modernity variables, and is negatively related with patron—dependence (Table V—l). But, contrary to prediction (and unlike mass media exposure physical mobility, and empathy) among high PD individuals there is a higher association between cosmOpolite contact and modernity than among low PD individuals. It seems that low patron-dependents (the ones that travel most, read most, and with better economic resources) talk more and most often with the cosmopolites in their community (who are more like themselves), but these local cosmopolites do not seem to exert any influence or act as a source for information about new ideas for the low PD farmers. On the other hand, high patron-dependents might have some selected contacts with the local cosmOpolites, regard these cosmOpolites as patron figures, and are influenced by the few pro-change messages that could be exchanged in such sparse contacts. 119 4. Mass Media Exposure — The data on mass media exposure behaved in the predicted directions, and supported all the hypothesized relationships. These findings suggest that the content of the media is, on the average, instrumen- tal for the modernization and development process.* Even when the mass media may not be carrying much information that could be directly considered as instrumental for agricultural development (such as information leading to the diffusion of innovations), exposure to the media seem to be teaching the farmers about the sources for such information. POSSIBLE REINTERPBETATIONS On the basis of the data from the present study, what are some of the possible theoretical reinterpretations and future testable hypotheses? First, the present study seems to indicate that, in patron-dependent societies, farmers do not make farming decisions individually. It seems that in certain less developed countries social structural differences affect decisions by the individual. Hodgdon and Singh (1963), in a study of the diffusion of innovations in India, show that "external" factors are much more important in explaining *McNelly (1964), writing about mass communication and the climate for modernization in Latin America, sug— gested that "much of the content in all of the media, including advertising, is informational, educational, or propagandistic in nature, designed to inform or persuade peOple about various kinds of modernization." 120 adoption than the individuals' decision to adOpt or reject the recommended innovations. Rogers (1966, p. 388) shows the relative importance of patrons in adOption decisions in a particular Colombian community where 95 percent of all the arable land belonged to five large landowners. The "diffusion model" (Rogers, 1962) developed in the U.S. with U.S. samples, assumes that individual farmers should be the unit of analysis since, in this less patron—dependent society, farmers individually take adOption decisions in most cases. But what will happen in other societies where hierarchical differences may affect adOption decisions? It seems viable to hypothesize that under the previously men- tioned conditions, social systems with more innovative patrons (or"elites," fin'national analyses) will have a faster rate of adOption than systems with less innovative patrons. Secondly, perhaps patron-dependence is antece- dgpt to extra-system communication and modernity, rather than intervening. There is both a logical and an empirical reason for this possibility. l. Patron—dependence is an enduring trait of the individual PD, learned through the childhood socialization process, might be considered to temporarily precede extra- systemic communication. 2. Only one of the three theoretic hypotheses testing the suppressor effect of PD was supported (Table V-l). 121 Further, when PD is not partitioned (Table IV-3), seven of the eight relationships between extra-system communica- tion variables and modernity variables are significantly different from zero. When PD is partitioned at the median (Table IV-S), 12 of the 16 empirical relationships between extra—system communication variables and modernity variables are significantly different from zero. In other words, the relationships between extra-system communication variables and modernity variables do not differ strikingly whether or not PD is mathematically present or not. Thus, we have not very convincing evidence that PD intervenes between extra—system communication variables and modernity variables. In order to check the antecedent position of PD in the relationship between the independent and dependent variables, it would be necessary to test the hypothesis that the relationship between PD and modernity variables is reduced by contrgllingaon extra—systemwggmmunication variables. RECOMMENDATIONS FOR RESEARCH We now proceed to make some research recommenda- tions that will help with the processes of knowledge creation and knowledge utilization. Based upon conclusions and evidence from the present study, we recommend the following types of research: 122 " ”" "“' 1. Research dealing with the validation of the present results with different pOpulations. This type of research, aimed at providing a wider hase for theoretic generalization, should also test whether the present re- sults hold across relatively modern and traditional sys- tems. Such analysis could provide evidence for the con- sistency of the present results during the sequence of develOpmental stages of a society.* 2. Research dealing with measurement improve— ment. Further validation of the present PD scale is neces- sary but it is also necessary to elaborate the index by including more diversified items (about non—nuclear family relationships and about patron figures outside of the immediate social systems). These items might provide insight about the possible existence of two or more PD subscales. 3. Research dealing with the amount of PD variance explained. It is possible that differences in PD could he better accounted for by intra-community variables than by extra-systemic variables. This type of study might provide evidence for further amplifications of the causal sequence proposedtm'de present dissertation. u -.— -‘”--*“"H mp-.. - - *sneh are the stages in economic development advocated hy Rostow (1960). 123 4. Research dealing with the so-called "take- off" models (Perzog, 1967b; Whiting, 1967; Waisanen and Kumata, 1969) of modernization. Change agencies are particularly interested in knowing not only what the important ingredients for change are, but also the con— centration and interaction of such factors in the Optimal- mix recipe of planned change. For example, working with such non—linear models, planners might determine not only that mass media exposure at the community level is relevant for develOpment, but also whether it might be more efficient to provide a few selected interpersonal contacts (outside the community) for a few leaders than to provide exposure to tlme media for the entire community. 5. Research dealing with the qualitative aspects of message content that producg message-acceptance by high andibw patron-dependents. Change agencies, in their inter- mediary role between sources (scientists, planners, and politicians) and receivers (clients or citizens), need to be effective in "filtering" which messages to choose for certain audiences under specified conditions. We are yet so far from this goal. RECOMMENDATIONS FOR CHANGE AGENCIES The findings of the present dissertation tend to suggest that: 1. In patron—dependent societies, change agencies should work through the hierarchy of patron—figures. l24 Acceptance of modern practices by innovative patrons should accelerate the adoption process among those who tend to base their adOption decisions on what others (in superior. social positions) have dOne. 2. Change agents should concentrate their efforts on high patron dependent individuals because our findings suggest such persons are especially influenced by cos— mOpolite contacts in the community. Change agents are local cosmOpolites and able to influence patron-dependent individ- uals through interpersonal contacts. While new change agents may profit in being iden- tified as a patron figure, they should be careful with such identification in order to avoid, after their departure, being replaced by less change oriented patron figures. Change agents should encourage their clients to have con- tact with other change-oriented cosmOpolites. 3. Change agencies should concentrate their mass media campaigns in reaching an audience of low patron- dependence individuals. It appears that low patron— dependents are more eager to accept pro—change messages carried by the media, and in the long run they should influence the less persuasible high patron—dependence individuals. 4. Change agencies should have a socio-educa- tional approach parallel to their technological approach. Agricultural extension agents for example should not only l25 work to increase their clients knowledge of hybrid corn, but should also try to increase their clients' knowledge of other social roles and behavioral alternatives far removed from the immediate community. Thus attempts to increase empathy and other variables might be a long—run objective of change agencies, in addition to their shorter range objectives of diffusing innovations. ****** In conclusion: if modernization is perceived as a sequence of steps representing different reinterpre- tations of the world and shifts in the reference system of an individual from the family to the community, to the nation, to the planet, to the cosmos. Then what com- munication inputs are necessary to overcome the thresholds separating these steps becomes a question needing a prior— ity answer. BIBLIOGRAPHY 126 BIBLIOGRAPHY Adorno, T. W., E. Frenkel-Brunswik, D. J. Levinson, and R. N. Sanford (1950), The Authoritarian Personality, New York: Harper. Aikin, c. (1964), "Nepotism", in J. Gould and w. L. Kolb (eds.), A Dictionary of the Social Sciences, New York: Free Press of Glencoe, 461-465. Ascroft, J. (1969), Modernization and Communication, Ph.D. Dissertation, East Lansing: Department of Communication, Michigan State University. Barnett, H. G. (1953), Innovation, New York: McGraw-Hill. Bennett, J. W. (1968), "Paternalism", in D. L. Shills (ed.), International Encyclonedia of Social Sciences, New York: MacMillan and Free Press, Vol. XI, 472-477. Bennett, J. W. and I.Ishino (1963), Paternalism in the Japanese Economy, Minneapolis: University of Minnesota.Press. Berlo, D. K. (1960), The Process of Communication, New York: Holt, Rinehart, and Winston. (1967). "The Process of Social Research", East Lansins: Department of Communication, Michigan State University (unpublished paper). Bordenave, J. D. (1966), The Search for Instrumental Tnfgg- mation Among Farmers of the Brazilian Northeast, Ph.D. Dissertation, East Lansing: Department of Communication, Michigan State University. Campbell, D. T. and D. W. Fiske (1959), Convergent and Discriminant Validation, Psychological Bulletip, 56:81-105 Campbell, D. T. and J. C. Stanley (1963). Experimental and Quasi-Experimental Designs for Reaserch, Chicago: Rand McNally. Clausen, J. A. (1968), "Socialization as a Concept and a Field of Study", in J. A. Clausen (ed.), Socialization and Society, Boston: Little, Brown 3-17 0 (ed., 1968), Socjalization and Spciety, —-~‘- ~v~---w“-M‘ ‘-. _ Boston: Little, Brown- 127 Coutu, W. (1951),'Role¥p1aying versus Role-taking", Amggicgn Sociological Revi ew, 16:180—187. DeAzevedo, T. (1962), Social Change in Brazil, Gainesville: University of Florida Latin American Monograph #22. Deutsch, M. and R. M. Krauss (1965), Theories of Social Psycholggy, New York: Basic Books. Deutschmann, P. J. (1963), "The Mass Media in an UnderdevelOped Village", Journalism Quarterly, 4(h62—75, Durlak, J. T. (1969), "Social System Identification, Com— munication and the Modernization Process". East Lansing: Department of Communication, Michigan State University (mineo). Fals Borda, O. (1961), Campesinos de los Andes, Bogota: Ioueima. Frey, F. W. (1966), The Mass Media and Rural DevelOpment in ngkgy, Cambridge: Center for International Studies, Massachusetts Institute of Technology. Freyre, G. (1946), The Masters and the Slaves, New York: KnOpf Friedman. J. (1968), An Information Model of Urbanization", Urban Affairs Quarterly, 4:235-244 Galjart, B. F. (1968), Itaguai, Wagenineen: Center for Agricultural Publ-i shing and Documentation. Gould, J. and w. L. Volb (1964), A Dictionary of the Social Sciences, Glencoe: Free Press. Hagen, E. E. (1962), Onire Theory of Social Changg, Homewood: Dorsey. Hays, W. L. (1963), Statistics for Psycholocists, New York: N..." Holt, Rinehart and Winston. Herzog, W. A. (1967a), "Mass Media Credibility, Bxposure, and Modernization in Rural Brazil", Boulder: Association for Education in Journalism Meetings. (1967b), Literacy Trainins and Modernization, Ph.D. Dissertation, East Lansing: Department of Communication. Michigan State University, Technical Report 3. Hodgdon, L. L. and H. Singh (1963), "The Adontion of Agricul— tural Practices in Two Villages of Madbva Pradeshfi New Dehli: Ford Foundation. 128 Hovland, C. I. and I. L. Janis, and H. H. Kelley (1953), Communication and Perspasion, New Haven: Yale University Press. Hutchinson, B. (1966), "The Patron-Dependent Relationship in Brazil". Sociologia Ruralis 6:3-29. Jantsch, E. (1967), Technological Forecasting, Paris: ' Organization for Economic Cooperation and Development. Kahl, J. A. (1968), The Measurement of Modernism, Austin: The University of Texas Press. Kenny, M. (1960), "Patterns of Patronage in Spain", Anthropological Quarterly, 33:14-23. Kolenda, P. M. (1963), "Toward a Model of the Indian Jaimani System", Human Organizations, 22:11-31. Leeds, A. (1964), "Brazilian Careers and Social Structure", American Anthropologist, 66:1321-1347. Lerner, D. (1958), The Passing of Traditional Society, New YOrk: Free Press of Glencoe Lewin, K. (1936), Principles of_T0pologica1 Psychology, New York: McGraw-Hill. McClelland, D. C. (1961), The Achieving Society, New York: Van Nostrand. McIver, R. and C. Page (1957), Society, London: MacMillan, McNelly, J. T. (1964), "Mass Communication and the Climate for Modernization in Latin America", Austin: Association for Education in Journalism Meetings. Odaka, K. (1964), "Traditionalism and Democracy in Japanese Industry", London: International Sociological Association Transactions, Vol. III, 39-49. Pocock, D. F. (1962), "Notes on Jajmani Relations", antribu- tions to Indian Sociology, 6:78-95. Rogers, E. M. (1962), Diffysion of Ipppyations, New York: ... ... g._ -‘.._ _.-..—o- Free Press of Glencoe. 1”,- (1965) "Mass Media Exposure and Modernization Among Colombian Peasants", Public Opinion Quarterly, ~“-.--.—- 1...- ‘~.- g. -- 129 (1966) Flementos del _Camhio Socialan America Latina, Bogota: Tercer Mundo Rogers, F. M. with L. Svenning (1969), Diodernization Amqgg --w—c. .—...‘-——.--— Peasants, New York: Holt, Rinehart and Winston. Rokeach, M. (1960), The Open and Closed Mind, New York: Basic Books. (1968), Beliefs, Attitudes and Values, San Francisco: Jossey Bass. Roling, N. (1969), "The Communication of Self-Control", East Lansing: Department of Communication, Michigan State University (unpublished paper). Rosemberg, M. (1968), The Logic of Survey Analys11, New York: Basic Books. Rostow, W. W. (1960), The Stases of Fconomic Growth, Cam- bridge: Cambridge University. Salcedo, R. N. (1968), A Communication Model of Moderniza- tion, Ph. D. DiSsertation, East Lansing: Department of Communication, Michigan State University. Schuessler, K. (1966), "A Note on Scale Analysis and Factor Analysis", Sociometry, 29:461-467 Shills, D. L. (ed., 10 68), International- Encyclopedia of Social Sciences, New York: MacMillan and Free“ Press. Siegel, S. (1956), Nonparametric Statistics, New York: McGraw-Hill. Stanfield, D. J. (1968), Interpersonal Trust and Modern— ization in Rural Brazil, Ph. D. Dissertation, East Lansing: Department of Communication, Michigan State University, Technical Report 9. Stanfield, D. J. J. Madureira, and G. M. Quesada (1967), "Selecoes de Inovacoes a Ferem Usadas na Fase III", Belo Horizonte: Diffusion Project (unpublished paper). Stanfield, D. J., L. Fonseca, W. Herzoe, and G. Whitins (1968), Communicat1on 1n Brazi1, Fast Tansins: Department of Communication, M1ch1aan State University, Research Report 14. '130 Stewart, D. and T. Hoult (1959), "A Social-Psycholorical Theory of the Avthoritarian Personality", fmericag Journal of Socio1ogy, 65:274-279. Van Ms, J. C. and F. A. Wilkening (1969), "Response Stability in Survey Research", San Francisco: Rural Socio- logical Society Meetings. Waisanen, F. B. 0969), Actors, Social Systems, and the Modernization_P§g§gs§. Bloomington: The Carnegie Seminar on Political and Administrative Develop- ment, Indiana University. Waisanen, F. B. and H. Kumata (1969), "Education, Functional Literacy and Participation Develonment", Mexico City: Society for Applied AnthrOpology Meetings. Weiner, M. (ed., 1968), Modernization, New York: Basic Books. White, R. K. and R. Lirpit (1960), Aptocracy and Democracy New York: Harper. Whiting, G. C. (1967), Ejnatny, I.'Iass1';edi_a_,_and Moderniygiggl in Rural Bragil,"PH.D. Dissertation, East Lansing: Department offlCommunication, Michigan State Univer- sity, Technical Report 1. Wilkening, E. A. (1964), "Some Perspectives on Change in Rural Societies", Rural Sociology, 29:1-17. C Willems, E. (1955), "Protestantism as a Factor of Cultural Change in Brazil", Economic DevelOpment and Cultural Change, 3:320-333. APPFNDIX 131 APPENDIX A THE CCDEBCOK AND INSTRUMENT F09 THE PRESFNT STUDY Topic IBM Question Column # Code Question # Card 1 1,2,3 Project Number by Phase and Countgy 712 - Brazil, Phase 2.5 4.5 ACAR Local Cffice and Community 20 -- Soa Joao Neponucemo (Rochedo de Minas) 22 —- Santos Dumot (Sao Joao da Serra) 24 -- Rio Novo (Goiana) 30 —- Paraopeba (Picada) 32 -- Pedro LeOpoldo (Mates) 34 -- Corinto (Curralinho de Dentro) 42 -— Itauna (Pedra) 43 -- DivinOpolis (Quilombo) 51 —- Formiga (Albertos) 70 -- Uba (Corrego Alegre) 6,7,8 Respondent Identification by Questionnarie Type- 500 ::: -- Also interviewed in Phase I 519 520 ::: —- Interviewed only in Phase II 599 Questionnarie A, Poorer Community: 600 ::: -- Also interviewed in Phase I 619 620 °'° —— Interviewed only in Phase II ga- Topic IBM ' Question Column # Code Question # Card I 9 CpmmunityType — Phase 2.5 5 -- Animation 6 —- Literacy 7 -- Control 8 -- Simulation 10. Type of Respondents 0 -— Land-owner, male, lives in nuclear center 1 -- Land-owner, male, lives outside nuclear center 2 -— Land-owner, female, lives in nuclear center 3 -- Land-owner, female, lives outside nuclear center 4 -— Non-land-owner, male in nuclear center 5 -- Non-land—owner, male outside nuclear center 6 —- Non-land-owner, female, in nuclear center 7 -- Non-land-owner, female, outside nuclear center 11,12 IBM_gard_Number 01 :: -- Card number 05 13,14 53:110. .f..____R $111,121.19: How 01a are you? (INTERVIFWFB: IF THE RESPONDFNT DCFS NCT VNow HIS CWN ACE ASV THE YFAR OF FIFTH) 133 .- TOpic IBM Cuestion Column M Code ' Question # EEFE 1 00 -— Actual age ...—....— __ n n 99 -- 15 Years of schooling completed by respondent How many years did you attend school? 0 -- None 1 -- one year 2 -- two years 3 __ H 4 __ H 5 __ N 6 __ n '7 __ It 8 __ u 9 -— nine years or more 17,18 Number of visits to a large city in the past_year Did you visit a large city last year? (One with more than 40,000 inhabitants) 00 —- Did not visit a large city past year 01 —— Number of visits m.past year to a large city 0! U3 \0 I I 19,20 NLmber of contacts per year with relative living in a large citv Do you have any relative who lives in a large cit (More tha.n 40, 000 inhabitantsK? 134 TOpic IBM Question - Column # Code Question # Card 1 00 -- Does not have relative living in a large city 01 -- Number of contacts per year with relative living in large city __ n __ n __ n 99 __ fl 25,26 Number of newspapers or magazines read per month ' Have you read (or has somebody read for you) newspapers or magazines lately? (IF YES) How many times a month? 9,9b 00 -- No or a "0" code in column ' 24 01 -- Number of times per month exposed to newspaper or magazine __ n __ u __ n 99 -- " 29 Fregugnge of radio listening How often do you listed to the radio? 12 0 -- Never 1 -— Almost never, doesn't know, no answer 2 -- Sometimes 3 -- More or less an hour per day 31 ifiequence of TV viewing Do you watch TV? 14 135 TOpic IBM Question . & Column # Code Question ” @1721 0 -- Never 1 -- Sometimes sees it, in some other place, doesn't know, no response 2 —- More or less regularly, at home 34 Frequence of letter writing Can you write letters? 16 0 -- Not able to write a letter 1 -- Is able, but doesn't; doesn't know; no response 2 -- Writes about one a year 3 -- Writes about one a month 4 -- Writes about one a week 38 Mutpal aid given" In the last year, whiCh type of help did you give to your friends? 20a a. help on farming, with some construc- tion given 0 __ (bx) Guns-wroI—J l I ...-m M *New items not No; never; no response Once a year 2-6 times a year 7 - 11 times a year Once a month 2 - 3 timesa month Once a week 2 - 6 times a week Daily present in previous phases. Topic IBM Question . Column 4 Code Question ” Lard l 39' Mutual aid received* a. help on farming, with some construc— 20a tion - received 0 —- (Same code as col. 38) 1 -.. 2 _- 3 __ 4 __ 5 _- 5 __ 7 __ 8-— 40 Mutual aid g1ven* b. cash loan - given 20b 0 -- (Same code as column 38) 1 -_ 2 _- 3 -_ 4 _- 5 -_ 6 _- 7 __ 8-- 41 Mutual aid received* b. cash loan - received 20b _- (Same code as column 38) CDx'lmU‘l-bLMMHO I I *New items not present in previous phases 137 --.—.-...- .__H---_._ TOpic IBM Question Column # Code Question # Card_l Mutual aidxiven* 42 c. tools, machinery, or animal loan - 200 given 0 -- (Same code as column 38) 1 __ 2 -_ 3 __ 4 __ 5 __ 5 __ 7 __ 8-.. 43 Mutual aid received* c. tools, machinery, or animal loan - 20c received 0 -- (Same code as column 38) 1 __ 2 _- 3 __ 4 -_ 5 __ 5 -_ 7 __ 8-- 44 Mutpal aid_giyen* d. help is case of illness — given 209 0 -- (Same code as column 38) 1 __ 2 __ 3 _- 4 -_ 5 __ 5 __ 7 __ 8-- * New items not present in previous phases, 138 Tonic IPM Question Column 4 Code Question # Ca?fi"l Reciprocal aid received* d. help in caaaof illness - received 20d 0 -- (Same code as column 38) 1 -_ 2 -_ 3 __ 4 __ 5 -_ 5.... 7 __ 8-- 47 Counter-factual behavior as responsible for the community If you were the person responsible for some movement1m>improve the community, what would you do? 22 0 -- Doesn't know; foolish responses; responses that show respondent's inability to think of himself as another person's role. 1 -- Very general answers that show some understanding of the situation and thethings that he could do but are not specified. 2 -- Specific responses with indica- tion that the person was actually thinking of himself in another person's role. 48 Cgppter—factual behavior as mayor the municipig If you were the mayor of the municipio, what would ..-... A... ...-n. a. *New items not present in you do? 23 (Same code as column 47) previous phases. 139 TOpic IBM Question Column # Code Question # DEFE"I 51 Counter—factual behavior as director of a factogy _ If you were the person in charge of a factory, what would you do? 26 O -- (Same code as column 47) 1 __ 2 _- 58 Kngwledge about home—pharmacy* What is a home-pharmacy? 28 O —— Wrong knowledge; doesn't know 1 —- First aid kit 59 Do you need something at home to take care of the small accidents that may happen on your property? 28a 0 -- No l -- Yes 60 What is necessary? 28a- 0 -- Doesn't know 1 -- One right thing 2 —- Two right things 3 -- Three right things 4 —- Four right things 61 Where do you use to keep these things? 28b 0 -- Doesn't know 1 -- Some place 2 -- A particular place *New items not present in previous phases. 140 Topic IBM Question Column # Code Question # Card 1 66 Knowledge about erosion-controlling_folliagg* What is a erosion-controlling folliage? 29 0 -— Wrong knowledge; doesn't know 1 -- Right knowledge (conservation- ist practice) 67 What do you know about it? 29a 0 -- Doesn't know 1 -- Knows 68 What is the distance between strips? 29b 0 -— Doesn't know 1 -— Knows 69 What is the best time to do them? 290 O —- Doesn't know 1 -- Knows 70 Is it necessary to redo them every year? 29d 0 -— Doesn't know; yes 1 —- No Cardiz .13 Knpwledge about planting maching* 'What is a planting machine 30 O -— Wrong knowledge; doesn't know 1 —- Right knowledge *New items not present in previous phases. 141 Tonic IBM Question 'Column # Code Question # Card 2 14 How has the soil been prepared for . the use of a planting machine? 30a 0 -- Doesn't know 1 -- Knows 15 Which part of the machine has to be changed for different types of seeds? 30b 0 -— Doesn't know 1 -- The disc 16 Where do you adjust the seedling machine? - 30c 0 -- Doesn't know 1 -— In a flat piece of land 23 Knowledge about trench-silo* What is a trench-silo? 31 0 -- Wrong knowledge; doesn't know 1 -- Right knowledge 24 What do you use to fill the trench— silo? 31a 0 -- Doesn't know 1 -- Knows 25 For how long must the trunch-silo be closed before starting to use it? 31b 0 -- Doesn't know 1 -- One month, more or less *New items not present in previous phases. 142 Topic IBM Question Column # Code Question # Card 2 26 How wide should the "loaf" be that you take daily from the trench-silo? 310 O -- Doesn't know I l -- 15 cm., more or less 33, 34 gpntact with ACAR How many times have you talked to the ACAR agent in the past year? OO -- Never 01 -- Once O2 -— Twice 99 : 38 Qounterfactual behavior as ACARagent What would you do if you were the ACAR agent? 35 O -- Doesn't know; foolish responses responses that show respondents inability to think of himself as another person's role. 1 -- Very general answers that show some understanding of thesntua— tion and the things that he could do but are not specified. 2 -- Specific responses with indica— tion that the person was actual- ly thinking of himself in another person's role. AdOption of agriculturalpractiggg Did you ever use: 43 53 Reforestation? O -- No' l —- Doesn't know; no response 2 -- Yes 143 ' Topic IBM Question Column # Code Question # Card 2 54 Termite control? 0 -— No 1 -- Doesn't know; ro response 2 —- Yes 55 Ant-killer? O —- No l —- Doesn't know; no response 2 -— Yes 56 Controfled breeding? O -- No l —— Doesn't know; no response 2 —- Yes 57 Tick control? 0 ..- NO. 1 -— Doesn't know; no response 2 -- Yes Year of praptice adoption How many years since you used for the first time: 43 58,59 Reforestation? OO -— Never used :: -- Year of first use 67 -- 60,61 Termite control: 00 --.Never used :: —~ Year of first use 67 -- ’ 144 , Topic IBM Question Column # Code Question # Card 2 62,63 Ant-killer? OO —— Never used :: -— Year of first use 67 ~- 64,65 Controlled breeding? OO -- Never used :: -- Yearcf first use 67 -- 66, 67 Tick control: 00 -- Never used :: -- Year of first use 67 -- 68 Do you still use: 43 Reforestation? O -- No l -- Doesn't know 2 -- Yes 69 Termite control: 0 -- No l -- Doesn't know 2 —- Yes 70 Ant—killer? O -- No 1 -— Doesn't know 2 —- Yes 145 ......‘ TOpic IBM Question Column # Code Question fl 9e92- 71 Controlled breeding? O -- No l -— Doesn't know 2 -- Yes 72 Tick control? 0 -- No l -- Doesn't know 2 —- Yes Card 3 Patriarchalism* If it were possible . . . 19 Would you wish that your sons follow 47a an occupation: O —- Chosen by you! (1 —- Doesn't know) 2 —- Chosen by themselves? 20 Whensour wife needs to buy clothes or medicines: 47b 0 —- She has to ask you first? or (1 -- Doesn't know) 2 -— She can buy and then tell you? 21 One of your daughters dates some boy you 470 know: 0 -- She needs a chaperone? or (1 -- Doesn't know) 2 —- No * New items not present in previous phases. 146 Code IBM . Question Column # Code Question ¥ Card 3 22 Would you allow your married sons to smoke when you are present? 47d 0 -- No (1 -- Doesn't know) 2 -- Yes 23 On your property, do you think it is always better to hire: 47e O -- A relative? or (1 -- Doesn't know) 2 -- A stranger if he is a good worker In a general way . . . 24 What the priest says: 48 O —- Is right? (1 -- Doesn't know) 2 -- Is good to discuss with the others? 25 When we need a job, what do you think: 49 O -- It is always better to accept a position near the relatives? or (1 -- Doesn't know) 2 -- You should accept a better position, even if it is away from the relatives? Do you think . . . 26 A father has to express his Opinion about the way his daughters use their clothes? 50a 0 -— No (1 -- Doesn't know) 2 -— Yes 147 TOpic IBM Question Column # Code Question # sari; 27 . Technical help: 50b 0 -- Is a favor that the govern- ment does the farmers? or (1 -- Doesn't know) 2 -- Is an obligation that the government owes to the farmers? 28 A girl should marry: 50c 2 —- Whoever she wants? or (1 -- Doesn't know) 0 -- Should first seek the advise of her parents 29 Counterfactuality as President of Brazil What would you do if you were the President of Brazil? 0 -- Doesn't know; foolish answers without originality l -- A little bit specific responses but without originality 2 -- Original and specific answers. 42,43 Functional literacy (CODER: SUBTRACT FROM 50 THE NUMBER OF WRONG WORDS AND WRITE THE ANSWER) "He who cannot read is like a blind man who has to be guided according to other peOple's wishes; or then he will stumble On his way. Tmeilliterate man is not altogether free; he is a slave of his ignorance. Never stop reading something everyday and keep learning." 00 -— No right word 50 -— Everything correct 148 TOpic IBM Question Column # Code _ Question # Card 3 45,46,47 Sociometric choices for discussions, leader- ship and trust* Can you give me the names of the persons with whom you talk most often about agricultural matters, such as productivity, new methods, etc.? 598-1 Are there any other persons with whom you do talk about the same tOpics with less regularity? 59a—2 Who are the three individuals that are most followed by others in topics about agriculture or cattle raising? 59b Who are the three persons from this com- munity that you trust most on farming? 590 000 -- No nomination 500 -- Indentification number of theiirst nominated person (a column field) 999 -- Person outside Phase IT sample 50 First person's place of work* Where he works? 0 -— No response 1 -- Same community 2 -- Same municipio 3 -— Another municipio 51 General meetings with first person* How frequently do youget together? - . *These questions were repeated ten times for a total possibility of up to ten nominations. 14 KO TOpic IBM Question Column 5 Code Question # Card 3 O —- Never; no response 1 -- l - 3 times a year 2 -- 4 — 7 times a year 3 -- 8 - 11 times a year 4 -- Once monthly 5 -- 2 - 3 times per month 6 -- Once a week 7 -- 2 - 4 times per week 8 -- Daily Card 5 Home and farm equipment and improvements 6O 13 'Do you own . . . Water filter? 0 -— No 2 -- Yes 15 Plumbing for running water? 0 -- No 2 -- Yes 16 Inside bathroom? O -- No 2 -— Yes 17 Electric lighting? 0 -- No 2 —- Yes 18 Radio? 0 -— No 2 —— Yes 150 TOpic IBM Question Column # Code Question # Card 5‘ 19 Motorized vehicle? 0 -— No 2 -- Yes 20 House in town? 0 -- No 2 -- Yes 22,23,24 Size of prOperty What is the total area of your prOperty? 61 000 -- No prOperty; no response 999 -— 999 hectares or more 25,26 Number of cows How many cows do you own? 62 OO -- None 99 -- 99 or more "liliiill'iilllliI