document resume kivlin, joseph e.; and others ... · document resume. ac 008 385. kivlin, joseph...

ED 042 105

AUTHORTITLE

PUB DATENOTEAVAILABLE FROM

EDRS PRICEDESCRIPTORS

IDENTIFIERS

ABSTRACT

DOCUMENT RESUME

AC 008 385

Kivlin, Joseph E.; And OthersCommunication in India; Experiments in IntroducingChange.May 6861 p.

The Department of Communication, Michigan StateUniversity, East Lansing, Michigan 48823

EDRS Price MF-i0.50 HC-$3.15Adoption (Ideas), *Agriculture, Communication(Thought Transfer) , Comparative Analysis,*Developing Nations, *Diffusion, *Farmers,*Innovation, Investigations, Literacy Education,Program Effectiveness, Radio*India

A study undertaken in India on the diffusion ofinnovations compared 108 villages on factors which affect adoption ofagricultural innovations, and then made comparisons of the adoptionbehavior of 680 Indian farmers. The third phase of the projectcompared the effectiveness of radio farm forums and adult literacyclasses for communicating information. The two villages using radiofarm forums showed significantly more progress then either the twovillages with literacy classes or the two control villages. There wassome tendency for literacy villages to show more progress thancontrol villages, but differences were usually not statisticallysignificant. It was concluded that a well-run radio farm forum shouldresult in rather immediate accelerated development, and that literacyprograms should he considered either as a long-term investment indevelopment or as a social welfare measure. (MF)

C 0

0 el;

%.".°.`

0

IX%

CD

COMMUNICATION IN INDIAO Experiments in Introducing Change

111

COMMUNICATION IN INDIAEXPERIMENTS IN INTRODUCING CHANGE

JOSEPH E. KIVLINPRODIPTO ROY

FREDERICK C. FLIEGELLALIT K. SEN

Research Report 15Project on the Diffusion of Innovations

in Rural Societies

Co-operating Institutions

National Institute of Community DevelopmentDepartment of Communication, Michigan State UniversityMinistry of Food, Agriculture, Community Development

and Co-operation, Government of IndiaUnited States Agency for International Development

NATIONAL INSTITUTE OF COMMUNITY DEVELOPMENTHYDERABAD-30

First Published May, 1968

Technical Editor

V. R. K. PARAMAHAMSA

Export Price $.1.35

rRINTED BY B. B. /1ADKARNI, AT THACKER & CO. LTD., PRINTING PRESS, EVERGREENINDUSTRIAL ESTATE, SHAKTI MILL LANE, OFF HAINES ROAD, BOMBAY 11 BC, ANDPUBLISHED BY THE NATIONAL INSTITUTE Of COMMUNITY DEVELOPMENT, HYDERABAD-30INDIA.

PREFACE

THIS is the third Indian report of the Diffusion ofI Innovations project which was undertaken in

collaboration with Michigan State University. Thefirst report was based on an analysis of 108 villages andwas published by the National Institute of CommunityDevelopment in March, 1968, under the title, Agri-cultural Innovations in Indian Villages. The second report,entitled Agricultural Innovation Among Indian Farmers isbeing published simultaneously with this report by theNICD, and is based on an analysis of 680 Indianfarmers and also investigates patterns of diffusion ofagricultural innovations.

This third report takes advantage of a previousstudy, done in collaboration with UNESCO andMichigan State University, to measure the continuingresults of communication treatments designed to promoteadoption of innovations. The previous study was reportedin August, 1967, NICD preliminary report, The impactof Communications on Rural Development in India.

Data for the follow-up survey, undertaken by theDiffusion of Innovations project, were gathered inLucknow in September and October of 1967. Projectdirectors were F. C. Fliegel, P. Roy, L. K. Sen andJ. E. Kivlin. J. M. Kapoor supervised the field team,consisting of P. M. Shingi, B. R. Patil, V. K. Surkar,S. K. Shelar and G. S. Sekhon. Data for this reportwere processed at the Computer Centre of the Pro-gramme Evaluation Organization, Planning Com-mission, New Delhi.

National Institute of GEORGE JACOBCommunity Development, Hyderabad. DeanApril 4, 1968. .

CONTENTS

Preface

INTRODUCTION ... 1

BACKGROUND OF THE STUDY 2

The 1964 and 1966 surveys ... ... 2Radio farm forums as a communication

treatment ... ... ... 4Literacy classes as a communication treat-

ment OD. 6The 'no treatment' control villages 8The sample for the 1967 survey ... 9Selection of practices ... ... 11

Interviewing and the 1967 interviewschedule ... ... 14

ANALYSISKNOWLEDGE, TRIAL ANDADOPTION OF PRACTICES 15

Knowledge, trial and adoption of agricul-tural practices ... ... ... 16

Knowledge, trial and adoption of healthpractices ... ... ... 19

Knowledge and adoption of individualpractices ... ... ... 22

Knowledge and adoption of family planning 25

ANALYSIS VARIABLES ASSOCIATEDWITH ADOPTION 30

Extension knowledge and contact 31

Literacy and mass media contact 34Attitudes toward programs of change 39Other measures of social change 41Villager evaluation of the communication

treatments 46

SUMMARY AND CONCLUSIONS 48

COMMUNICATION IN INDIAExperiments In Introducing Change

INTRODUCTION

ALLattempts at communication are in a sense

experimental. Whether messages get through andimpart the intended meaning is always somewhatuncertain. This is particularly true of experiments inintroducing change. People often resist change, forit may clash with personal and societal values andinterests. The study which we report here is part of acommunication experiment which tried to induceadoption of modern agricultural, health and familyplanning practices. Our study evaluates the continuingeffects of two communication treatments, radio farmforums and literacy training classes. Our data willshow that knowledge about the practices spreads ratherquickly. Adoption, however, has been slow and at lowlevels. It seems obvious that only rarely is there aneasy or quick solution to problems of development.As Millikan and Hapgood point out in the title of theirbook on the dilemma of agriculture in developingnations, there is No Easy Harvest.' Nevertheless, wewill show that some progress was made and we willshow clear-cut differences in effectiveness Of the com-munication treatments which were used.

1. M. F. Millikan and I). Hapgood, .N Easy garrest, Boston: Little, Brownand Company. 1967.

2 COMMUNICATION IN INDIA

BACKGROUND OF THE STUDY

THE 1964 AND 1966 SURVEYS

The study on which our report is based is a follow-up of part of a two-nation comparative study initiatedin 1964 by UNESCO, Michigan State University, theNational Institute of Community Development (NICD)of India and the Programa Inter-americano de In-formacion Popular of Costa Rica. The objectives ofthis study in both Costa Rica and India were: (1) todetermine the impact of different means of communi-cation on the spread of agricultural, health and familyplanning practices; (2) to determine the factors affectingthe acceptance of innovations; and (3) to describe thetotal process of diffusion of innovations. Our reportdeals only with the Indian data and focuses on the firstobjective, that of evaluating the effectiveness of com-munication treatments.

The location of the site for the study in India wasnear the city of Lucknow, Uttar Pradesh, mainly becauseof the proximity of facilities for conducting the com-munication treatments. Literacy House, an adultliteracy training center, is located there and waswilling to co-operate with The project in the literacyclasses treatment. It was also possible to secure theco-operation of the All-India Radio outlet in Lucknowfor the treatment of the radio farm forums. Additionally,competent research investigators who knew the locallanguage were available as Lucknow University haslarge departments of Anthropology, Sociology andSocial Work.

Villages were selected from a pool of 24 similarvillages near Lucknow on the basis of available statisticaldata, so as to make them as alike as possible. Factors

COMMUNICATION IN INDIA

such as per cent of off-farm employment, proximity tothe city of Lucknow, and amount of cultivable land wereconsidered. Three villages were re-selected when itwas found that they would be inaccessible during therainy season but otherwise the statistically selectedsample was retained.

In 1964, a benchmark survey was made of thestudy villages to establish the level of knowledge, trialand adoption of selected agricultural, health and familyplanning practices. Information was also obtained onvariables usually associated with adoption. Then, in1965, an experiment was conducted in which the com-munication treatments of literacy classes and radiofarm forums were applied to two pairs of villages butnot to a third pair on which benchmark data were alsoobtained.2 In 1966, a re-survey was made of the samerespondents to examine the effects of the treatments onvarious factors, especially the knowledge, trial andadoption of the selected practices. The details of thissampling and selection arc given elsewhere and willonly briefly be summarized here.' The selection ofvillages and respondents from 1964 to 1966, of course,set the limits for our re-survey in 1967.

The communication treatments were started inFebruary, 1965, and continued for about a year. Wewill discuss each treatment briefly to indicate how itwas used to stimulate interest in using new practices.2. A third communication treatment, 'animation', was also applied to two

villages. Animation is a method of intensified oral communication onrural development designed to develop local leadership. Although thiscommunication training method has had considerable success in Indiaand elsewhere, differences in adoption of practices between the animationexperimental villages and the control villages were so slight as measuredin 1966 that the animation villages were dropped from the 1967re-survey.

3. This description of the 1964 and 1966 surveys is largely taken from P. Roy,The Impact of Communications on Rural Development in India, Hyderabad:National Institute of Community Development, 1967. (Mimeo).


RADIO FARM FORUMS AS A COMMUNICATIONTREATMENT

Radio farm forums were introduced by the All-India Radio network in 1949 as a means of enhancingthe informational and motivational values of radio-listening. A forum consists of from 10 to 20 villagerswho gather together to listen to programs about farmproblems. They elect a secretary who keeps minutesof the discussions of the meetings which follow thebroadcast. A rotating chairman who leads the dis-cussions is elected for each broadcast. Requests foradditional information or about adaptations of recom-mendations to local conditions may be sent to the radiostation. Hired action leaders are also used. Theseguide discussions, if necessary, and give advice onputting the information to use. For example, if afarmer had losses from rats he would be encouraged toobtain poison and told how to use it in his own storageshed.

There was one radio farm forum in each of thetwo radio-treatment villages. One had nine membersand the other had 14 members. Each forum wasprovided with a transistorized radio set and memberslistened to about 100 half-hour talks, which focussedon improved agricultural and health practices. Weatherforecasts and other farm information were usuallyincorporated into the programs. The broadcast schedulewas designed to try to match seasonal variations infarm work. Some talks were repeated for emphasis.For example, talks on the modern plow and on fertilizerusage were given before both the rabi (winter) andkhariff (rainy) planting seasons. Complicated practiceslike Japanese method of paddy cultivation were discussedin sub-topics to take up the separate practices whici

COMMUNICATION IN INDIA 5

were part of the whole recommendation of this methodof cultivation. In all, the talks provided a maximumof about 50 hours of exposure to radio and about 100opportunities to participate in or listen to radio farmforum discussions.4

Attendance varied considerably, from 30 to 90 percent, with some members being chronically absent andattendance generally dropping during busy workseasons. Many people who were not members of theradio forums attended various broadcasts and, ofcourse, this was one of the intents of the programs. Itwas intended to obtain a 'two-step flow of information'from radio to opinion leaders and innovators whoattended the forums, and then from these to the largerpopulation of cultivators in the village. Thus therewere 'primary' audiences of listeners and 'secondary'audiences of cultivators who had the information passedon to them. Analyses of these two audiences are partof the report on the 1964 and 1966 data.' It waslargely this passing on of new information and theconsequent changing of social and cultural valueswhich, it is inferred, resulted in the differences observedin 1967 between the experimental and control villages.These differences are reported in our present study.

An example of how social and cultural factors werealtered through the advocacy in a radio forum of anew practice is that of rat poison. Cultivators weregenerally familiar with the idea of rat poison but fewwould use it openly because its use involved the killingof a living being. Such killing was felt to be proscribedby caste and religion. Even though some cultivatorswere using rat poison themselves, they openly dis-4. Ibid.5. ibid., pp. 35-38.


couraged its use by others for social reasons. Thisproblem was especially acute in a village with a largenumber of chamars (a caste near the bottom of theHindu social ladder) who were trying to raise theirstatus by rigorous emulation of upper castes. In thisvillage, the rat problem became acute as a result offollowing the proscription against their killing. Thefirst radio program and discussion on rat poison was notvery effective because the matter was considered toodelicate to discuss publicly. A later talk gave additionalinformation and vividly called attention to the lossesincurred as a result of rats. Listeners became moreinterested then and, in subsequent discussions, it wasdecided that use of rat poison was not such a seriousviolation of caste or religious prescription and couldbe adopted. In fact, knowledge and adoption of ratpoison reached such high levels by 1966 that thepractice no longer discriminated satisfactorily amongrespondents and was dropped in the 1964 - 66 com-parisons.

LITERACY CLASSES AS A COMMUNICATIONTREATMENT

Like the radio farm forums, literacy classes wereused as a vehicle for diffusing information aboutimproved practices and for promoting their adoption.Unlike the radio forums, the literacy classes had a majorseparate goal, that of teaching people to read andwrite. The 1961 Census of India shows a literatepopulation of only 24 to 34 per cent for men and13 per cent for women. For several reasons, onlymale literacy classes were started. Women do notplay as important a role as men in agricultural decision-making; it is more difficult to recruit women teachersand organize a sustained women's literacy class; partici-

COMMUNICATION IN INDIA .7

pation in the radio farm forums, as expected, wasexclusively confined to males and there was onlyrarely a female listener of the programs.6 As our chiefpurpose was to compare communication treatments onas equal a basis as possible, only male literacy classeswere started.

The literacy experiment was organized in co-ordination with Literacy House, an adult literacytraining center in the city of Lucknow. The programinvolved both literate and illiterate men. For literates,reading forums were organized and village librarieswere established. For illiterates, literacy classes werestarted. At the end of these classes, when readingproficiency had been obtained, the groups were con-verted into reading forums. The literacy materialconsisted lr.,rgely of description and advocacy of agri-cultural, health and family planning practices. It wasexpected that constant reference to the practices in theliteracy classes and reading forums would promoteadoption of the practices. Pamphlets designed topromote improved practices were mostly written es-.

pecially for project use by Literacy House staff. In-all,15 pamphlets were published with such titles as`Increase wheat yields with improved seed' and 'Rat ---an enemy of crops' (English translations).

Two classes of 25 participants each, met in one ofthe literacy treatment villages and one class of '35members met in the other village. The classes met forabout a year, for two-and-a-half hours daily, six daysa week, from 8.00 to 10.30 p.m. Hindi Devnagari

G. Women do listen to the radio when it is available, but not as often as men.Jn a previous study we found that, of 680 male cultivators, two-thirdsreported that their family did no radio4istening while less than one-quarterof the cultivators themselves reported that they did not listen to the radio.See P. Roy, F. C. Fliegel, J. E. Kivlin and L. K. Sen, Agricultural Innovationamong Indian Farmers, Hyderabad: National Institute of CommunityDevelopment, May, 1968.


script was used. Participants were relatively young,ranging from 15 to 45, and were from widely differentcastes. Attendance varied from 30 to 100 per cent.

An important part of the literacy classes communi-cation treatment was the reading forum discussions.As in the radio farm forum, these consisted of discussionsof the material on improved practices which was beingused in the literacy classes or read in the reading forumby the more advanced readers. Each forum had arotating group-leader whose duty it was to guide thegroup oa the selected subject. As in the radio forums,an attempt was made to introduce the various inno-vations in a timely manner and an action worker wason hand at times to also guide discussions and to promoteadoption of the innovations. There were about 50reading forum discussions held in each literacy treat-ment village.

THE 'NO TREATMENT' CONTROL VILLAGES

The two control villages were not subjected toany special communication treatment. Like the experi-mental villages, however, they received routine attentionfrom community development block headquarters andfrom other government agencies. It is possible thatthe experimental villages received some additionalattention from government officials as a function ofbeing singled out for the communication treatments,but this was not noticeable. In any case, we couldnot hope to conduct our experiments in a social vacuum.Comparisons of relative positions on various factorsamong the experimental and control villages will showthat the three sets of villages did not differ very much in1964 when the benchmark survey was taken. Like theexperimental villages, the control villages were selected


from a pool of 24 similar villages near Lucknow on thebasis of available statistical data, so as to make them asalike as possible. The control villages, then, weresimilar to the experimental villages except that we didnot conduct communication treatments in them.

THE SAMPLE FOR THE 1967 SURVEY

The benchmark survey taken in 1964 was a nearlycomplete enumeration of all male heads of householdsin eight villages for a total of 702 interviews. Adetailed presentation of some social characteristics ofpeople living in these villages in 1964 is given in theearlier report.' In the 1966 re-survey, there was areduction in the size of the sample to 463 respondentsbecause non-cultivators were deleted from analysisand because of death, migration or non-availability ofrespondents. In the 1967 re-survey there was a furtherreduction in number of cases because of the deletionof two villages in which the 'animation' communicationtreatment was applied.8 Additionally, nine respondentshad died or moved away in the lapse of a year from1966 to 1967. In the six remaining villages (tworadio, two literacy, two control), there were 291respondents who had been interviewed both in 1964and in 1966. Of these, interviews were obtained in 1967from 279 cultivators, or 96 per cent of those available.There were no refusals. Most of the 12 cultivatorswho could not be located, worked in the city ofLucknow.

A few characteristics of the 279 respondents as of1966 are given as comparisons among the three sets ofvillages, radio, literacy and control (Table 1). The

7. P. Roy, The Impact of Communications . . . , op. cit., pp. 24-34 and Appendix.Et. See footnote No. 2.


Table 1: Selected Characteristics of Respondents, Com-munication Treatment and Control VillagesCompared, 1966

Education

Radio Literacy Control(N = 78) (N = 112) (N = 86)

00 0 ut 0 v.- c 0 col

E, z 414

g-09a2, 51.1-

g4'4 i:4 4:1 5

No education ...Up to 4 ...5 to 89 to 12

71 9924 1

5 00 0

per cent

55 9119 522 4

4 0

82 9911 1

6 01 0

100 100 100 100 100 100

Average age of Radio Literacy Controlrespondent (N = 79) (N = 112) (N = 87)

42 40 44

Average number of Radio Literacy Controlfamily members (N = 79) (N = 112) (N = 87)

5.28 6.17 5.49

Per cent of joint Radio Literacy Controlfamilies (N = 79) (N = 113) (N = 87)

29 44 33

Tenure status of Radio Literacy Controlrespondent (N = 78) (N = 108)

per cert

(N = 87)

Landless tenant 0 2

Part owner, parttenant 22 18 24

Full owner 78 80 76

100 100 100


three sets of villages were generally similar in 1966.The control villages had the highest percentage ofrespondents with no education. In all of the villages,there were very few wives with any education at all,the literacy villages having 9 per cent wives withsome education. Average age of respondents rangednarrowly from 40 to 44 with a similar small spread inaverage number of family members, from 5.28 to 6.17.The latter figure, of courze, speaks graphically of thepresent problem of population growth. At these levelsof family size the population would double in about30 years.

The widest difference among sets of villages inthis limited comparison was in per cent of joint families,which varied from 29 to 44 per cent. The criterionfor jointncss was two or more complete nuclear families(husband and wife and children, if any) who atetogether. We attach no particular importance to thedifferences among the sets of villages. In another studywe found that family composition, whether joint ornuclear, was not importantly related to agriculturaladoption or to most other variables.9

The sample for the present study then, was com-posed of 279 respondents, all of whom were males whohad cultivated some land during the time of the study.They had all been interviewed in 1964 and 1966whether or not they participated directly in the com-munication treatments.

SELECTION OF PRACTICES

Practices were selected in 1964 on the basis of fourmajor criteria: (1) applicability in all of the samplevillages; (2) salience to both villagers and. agencies of

9. P. Roy, et al., Agricultural Innovation . . , op. cit.


change; (3) suitability to be included in the communi-cation treatments; and (4) adoption level ranging from0 to 60 per cent with a potential for increasing by theend of the experiment. There were 30 practices whichseemed to meet these criteria in 1964. Sixteen agricul-tural, health and family planning practices wereselected from among these to make internally consistentindexes with which to measure change over time. Theabove criteria, Guttman scaling, correlation analysis,and comparability for 1964, 1966 and 1967 in thepresent study were used as the basis for deletion ofpractices. The practices are listed below.

AGRICULTURAL PRACTICES

Japanese method of paddy cultivationSuperphosphateLine-sowing of wheatImproved potato seedGreen manureAmmonium sulphateInsecticidesModern plowThree or five-tine cultivatorAnimal disease innoculation

HEALTH PRACTICES

PRAI latrine (Planning Research and ActionInstitute)

Smokeless chula (stove)Bed-bug killerModern child-birth practicesTABC (innoculation for one or more of the

diseases of cholera, typhoid, small-pox)


FAMILY-PLANNING

Birth control, any method

Most of the practices arc self-evident. For example,`superphosphate' or 'improved potato seed' simplyrefers to use of these two items. Line-sowing of wheatis an improved method introduced about 15 years agoto reduce the amount of seed required and to ensurethe optimum planting distance. However, consider-ably more labor is required for this method of plantingwhich contrasts with broadcasting. By modern plow isgenerally meant a steel or iron moldboard-type imple-ment, which contrasts with the wooden desi plow.Many cultivators now use both plows. The three orfive-tine cultivator is another 'modern' steel or ironimplement, and is of course dependent upon line-sowing of crops for successful use.

The Japanese method of paddy cultivation is apackage of practices, all of which must be taken upin order to achieve the optimum results. The mostcomi-non component practices of the package areimproved seed, seed treatment, adequate irrigationand drainage, special preparation of the seed bed, line-sowing, use of fertilizer and pesticides, and intercultureor cultivating to control weeds. The practices arcsometimes taken up individually, with line-sowing andinterculture often omitted.

Of the five health practices, only bed-bug killer,usually a product called Gammaxcne, is self-evident.The PRAI latrine is a locally constructed water-seal la-trine, the water being poured into the latrine with a bucketby the user. Wastes flow into an underground septictank where decomposition takes place. Originally thelatrines were subsidized but subsidies are no longer given.


The smokeless chula is a stove designed to use fueland remove smoke more efficiently than unimprovedmodels. It usually provides more cooking surfaces.The smokeless chula is unique among our 16 practices,in that there were no users in any village in any of thethree years for which we present data. The chiefreasons for non-adoption seem to be that some modelsin neighboring villages were anything but smokelessand that villagers like to move their stove about, bothindoors and outdoors. The smokeless chula has a fixedlocation and also has a higher initial cost.

Modern child-birth practices were taken to meanattendance at birth by a trained midwife or doctor,either at home or in a hospital. TABC usually refersto a single innoculation giving protection against two ormore diseases. In our study, any single or combinedinnoculation against cholera, typhoid or small-pox wastaken as adoption of TABC. Family planning referredto any method of birth control, including indigenousmethods such as withdrawal. The comparisons thatfollow, then, are for ten agricultural practices, five healthpractices, and family planning by itself as one practice.

INTERVIEWING AND THE 1967 INTERVIEW SCHEDULE

For the 1967 re-survey, which was conducted inSeptember and October, a single interview schedulewas prepared. It was translated into Devnagari (Hindi)script and was administered by experienced inter-viewers knowledgeable in the Avadhian-Hindi dialectof the Lucknow area. Question format and codingnecessarily followed that of the 1964 and 1966 surveysso as to permit comparisons across years. Coding wasdone on the schedules, especially prepared for thispurpose, se that IBM punching and verifying could be


done directly from the schedules. This was done andaverages and percentages were calculated from thesedata. Comparable averages and percentages werealso calculated from the 1.964 and 1966 data obtainedfrom the same 279 cultivators who were interviewed in1967. These comparisons provide the basis for analysisin the present study.

ANALYSIS KNOWLEDGE, TRIAL ANDADOPTION OF PRACTICES

Our chief comparisons are those between the twocommunication treatments of literacy classes and radiofarm forums and between the communication treatmentand the control villages. Presumed to have beenconstant for all three sets of villages were the normalblock development programs. These on-going programswere expected to result in increases in adoption andother factors in all villages over the time period whichwas studied. Statistical tests for the significance ofdifferences in grouped averages and percentages ( T-tests)were run only for the 1967 data, which are reportedhere for the first timc.'° Changes from 1964 to 1966to 1967 arc apparent fbr most variables in the Tableswe present. We will note these changes but have notmade statistical tests of the differences across years orof the differences across sets of villages for 1964 and1966. To repeat, tests of significance were used tocompare treatment effects in 1967 only. The 5 percent level of significance was used in all Tables.

First, we discuss knowledge, trial and adoption ofagricultural, health and family planning innovations.These provide a direct test of the efficacy of the corn-10. Comparisons for the 1961 and 1966 data are also given in P. Roy, The

Impact of Communications . . . , op. cit. These differ slightly from those WCpresent here because of the larger samplc used in 1966.


munication treatments. Then, we consider other,more indircct consequences such as an increase in thenumber of change agents known or in the amount ofradio-listening. Increases in such reactions arc likelyto result in higher knowledge and adoption, for thesevariables arc usually positively related with bothknowledge and adoption of practices.

In general, to anticipate our data somewhat, theradio farm forum treatment villages showed significantlymore progress than the literacy and control villages.This was true whether measured by adoption or byother variables. There was some tendency for theliteracy villages to score higher than the control villagesbut usually these differences were not statisticallysignificant. In only two of the 1967 comparisons didthe control villages score highest.

KNOWLEDGE, TRIAL AND ADOPTION OFAGRICULTURAL PRACTICES

The chief emphasis in the communication treat-ments was on agricultural practices. Over 80 per centof India's people live in rural areas and most of themare engaged in agriculture. In Table 2 we present theaverage number of agricultural practices known in eachof the three sets of villages for each of the years, 1964,1966 and 1967. This format will be used throughoutour analysis to compare communication treatments.We would like to stress once more that the samepractices, same respondents and identical questions toobtain data were used for all three years in theseTables. Thus, while perceptions of our questions byrespondents have undoubtedly altered over time insome respects, we have attempted to make a continuingstudy of comparable data to show changes which arefundamental to development.


Table 2: Average Number of Agricultural Practices Known,Communication Treatment and Control VillagesCompared, 1964, 1966 and 1967

Treatment 1964 1966 1967**

Radio farm forum* -. 6.38 9.51 8.98

Literacy classes* ... -. 5.89 7.92 8.81

Control (no treatment)* ._ 6.33 8.48 8.60

* For this and succeeding Tables in this report, there were two radio farmforum treatment villages, N = 79, two literacy classes treatment villages,N = 113, and two control villages in which there were no communicationtreatments, N = 87. There were only negligible missing data in any of theTables.

** For this and succeeding Tables, differences significant at the .05 level willbe indicated, for 1967 figures only. T-tests for significance of the differencesbetween means and between percentages were used. None of the differencesin the 1967 figures for this Table was significant. Non-significance ofdifferences should be assumed in succeeding Tables in the absence of state-ments of significance.

There were substantial increases in knowledge ofour ten agricultural practices in all three sets of villagesfrom 1964 to 1967 but averages for the sets of villagesdid not vary significantly in 1967 (Table 2). However,what proved to be the predominant pattern in ourcomparisons is shown in these 1967 averages in that theradio villages ,are highest, 8.98, followed by literacyvillages, 8.81, with control villages having the lowestaverage, 8.60. The total possible knowledge was of tenpractices, the first ten listed earlier. It is evident thatknowledge of these agricultural practices was widelydiffused by 1967 and that average knowledge had`peaked' close to the upper limit of ten, so that wecould not discriminate among communication treatmentand control villages on this variable. The small de-cline in average number of practices known in the radiovillages, from 9.51 in 1966 to 8.98 in 1967, is probablythe result of poor recall by some cultivators in 1967.


There were significant differences in averagenumber of agricultural practices tried between radioand literacy and between radio and control villages(Table 3). The average for radio villages was highest.Literacy villages were only slightly higher than controlvillages. All sets of villages showed increases inpractices tried from 1964 to 1967. These averages forpractices tried must be viewed with considerablecaution; although the figures probably reflect accuratelythe relative positions of radio, literacy and controlvillages in the matter of practice trial, they do notnecessarily reflect disadoptions. That is, the figuresfor trial (Table 3) are higher than those for adoption(Table 4). Similarly, the figures for trial (Table 3)are lower than those for knowledge (Table 2). It isnot necessarily so that cultivators were not interestedin trying practices after gaining knowledge of them orthat they tried and voluntarily disadopted somepractices. Drought and flood conditions, non-avail-ability of supplies and services, cropping changes andother factors making a practice non-relevant to aparCcular farm, could all result in fewer trials thanknowledge and more trials than adoptions. Voluntarydisadoptions of relevant. practices, of course, do occurand are a fruitfill area for study but our data do notpermit a careful examination of them.Table 3: Average Number of Agricultural Practices Tried,

Communication Treatment and Control VillagesCompared, 1964, 1966 and 1967

Treatment 1964 1966 1967*

Radio farm forum ... 2.75Literacy classes ... ... 2.4-2Control ... 2.68

5.44 6.683.46 4,573.38 4.47

* DifferCUCCS betWrell radio and literacy and between radio and controlsignificant at .05 level.


Table 4: Average Number of Agricultural PracticesAdopted, Communication Treatment and ControlVillages Compared, 1964, 1966 and 1967

Treatment 1964 1966 1967*

Radio farm forum 2.58 4.84 5.90Literacy classes ... 2.21 2.83 3.38Control 2.43 2.71 3.45

* Differences between radio and literacy and between radio and controlsignificant at .03 level.

In Table we present averages for number ofagricultural practices adopted. Again, the radio villageshave a significantly higher average than the literacyand control villages but on this factor, control villageshave a slightly higher average than literacy villages.These figures for average adoption are the most validevidence we have of the greater efficacy of the radiofarm forums in promoting adoption of innovations. Anaverage of almost six out of ten practices were adoptedin the radio villages as compared to less than four outof ten for literacy and control villages. This evidenceis hardly conclusive, for we have dealt with only oneexperiment, all aspects of which were in some respectsunique. Another experiment, in another place, con-ducted somewhat differently, might show quite differentresults. Nevertheless, combined with the other datawhich we present, the comparisons on average adoptionargue strongly for the promotional qualities of radiofarm forums.

KNOWLEDGE, TRIAL AND ADOPTION OF HEALTHPRACTICES

Figures for the average number of health practicesknown, tried and adopted follow the pattern whichemerged in our analysis of agricultural practices.


These averages arc based on our five health practiceslisted earlier. Radio villages had a significantly higheraverage than one or both of the other sets of villages,while the literacy villages scored higher than thecontrol villages, but not significantly so (Tables 5, 6and 7).

Table 5: Average Number of Health Practices Known,Communication Treatment and Control VillagesCompared, 1964, 1966 and 1967

Treatment 1964 1966 1967*

Radio farm forum ... 1.57 2.29 3.47

Literacy classes ... ... 1.19 2.79 3.25Control ... 1.26 1.88 2.91

Difference between radio and control significant at .05 level.

Table 6: Average Number of Health Practices Tried, Com-munication Treatment and Control VillagesCompared, 1964, 1966 and 1967

Treatment 1964 1966 l967*

Radio farm forum 0.66 (1.89 1.56

Literacy classes 0.68 1.08 1.13

Control 0.74 0.73 1.03

* Diflia.ence bctween radio and control significant at .cri hwo.

Table 7: Average Number of Health Practices Adopted,Communication Treatment and Control Villages

Compared, 1964, 1966 and 1967

Treatment.... 1964 1966 1967*

Radio farm forum 0.61 0.82 1.11Literacy classes ... 0.62 1.00 0.68Control 0.69 0.60 0.57

* Differences between radio and control and butc:cen radio and literacysignificant at .05 level,


Health practices have a different relationship tothe whole complex of development factors than doagricultural practices. There arc more social welfareand fewer direct economic implications to healthpractices. There is also more emotional involvementand a greater sense of urgency for adoption in a crisissituation. Furthermore, some practices like malariacontrol measures may be quite involuntary. However,only TABC of our five health practices had someaspects of involuntary adoption. School children weresometimes involuntarily innoculated. TABC andmodern child-birth practices also suffered somewhatbecause of non-availability of medical services in thevillages. This resulted in some seeming disadoptionsof health practices.

An average of about three out of the five healthpractices were known in 1967 (Table 5). There waslittle difference among the three sets of villages. Con-sidering that information about these practices has oftenbeen disseminated, this average level of knowledge canbe considered low. The reason for this relatively lowlevel of knowledge probably lies in part in the very lowlevels of trial and adoption (Tables 6 and 7). Wide-spread knowledge depends in part on trial and adoptionby a few innovators who then report their success toothers and spread information about a practice. Averagetrial and adoption figures ranged from a low of 0.57average adoption in 1967 in the control villages to ahigh of only 1.56 average trial in the radio villagesfor that year.

In all six villages combined, there were tenrespondents who disclaimed knowledge of any healthpractices whatsoever in 1967. This is unusual if onlybecause these people had been asked about these same

1..


practices in 1964 and in 1966. There were 53respondents who had not tried any practices and 126respondents who were not currently using any healthpractices in 1967. These low levels of knowledge, trialand adoption are probably indicative of the generalsituation in regard to health in many Indian villages.The health situation per se would not seem to be parti-cularly alarming by local definition. Despite a shortageof trained medical personnel and a shortage of economicmeans of villagers to afford medical services, therewere no crucial health problems reported in our studyvillages, although infant mortality was reported to behigh. Ong!. of he most important implications of ourhealth data may be that of a low level of concern forfamily planning. Past experience in attempts to reducebirth-rates has generally been that concern for familyplanning follows and does not precede concern forbetter health. People must first appreciate that smallernumbers of children will have a good chance of surviving.It is true, of course, that the first effect of health controlmeasures is to add to the population by decreasing thedeath-rate, but there seems to be no easy way aroundthis temporary result. We will take up family planningas an improved practice in a later section and willdemonstrate that while knowledge of at least one birthcontrol technique was high, adoption was at a verylow level.

KNOWLEDGE AND ADOPTION OF INDIVIDUALPRACTICES

To conclude our discussion on the 15 agriculturaland health practices, we present data on knowledgeand adoption of the individual practices (Tables 8and 9). These Tables follow for most practices the.pattern which has emerged in the grouped data. Radio

Table 8: Per cent of Respondents Having Knowledge of Practices, Communication Treatmentand Control Villages Compared, 1964, 1966 and 1967

1964 1966 1967

Radio Literacy Control Radio Literacy Control Radio Literacy Control

Agricultureper cent

JMRC ... 65 52 63 92 76 78 87 77 85Superphosphate ... 49 47 47 78 60 69 96 93 90Line-sowing of wheat 78 85 79 99 95 84 95 98 90Improved potato seed 16 16 16 94 85 68 73 66 77Green manure ... 67 74 61 90 88 69 85 81 87Ammonium sulphate 96 91 93 97 74 86 96 97 91Insecticides ... 82 77 76 95 87 79 99 94 93Modern plow ... 99 80 82 100 98 91 100 99 1003 or 5-tine cultivator 14 8 30 82 66 67 80 73 67Animal innoculation 73 67 74 91 75 79 90 89 82

HealthPRAI latrines 35 15 7 49 51 38 74 71 53Smokeless chula 1 5 1 9 37 7 42 38 28Bed-bug killer ... 22 28 20 44 58 29 73 55 57Modern child-birth

practices ... 20 8 14 38 50 30 72 72 62TABC ... 77 67 80 90 88 72 95 94 92

Table 9: Per cent of Respondents Having Adopted Practices, Communication Treatment and isDoPControl Villages Compared, 1964, 1966 and 1967

1964 1966 1967

Radio Literacy Control Radio Literacy Control Radio Literacy Control

Agriculture--per cent

JMRC ... 19 7 18 22 5 15 59 11 40Superphosphate ... 10 14 28 49 20 35 78 39 29Line-sowing of wheat 24 19 14 56 46 18 71 52 48Improved potato seed 16 II 13 42 29 22 46 24 22Green manure ... 10 21 14 15 31 11 34 18 11Ammonium sulphate 73 54 67 70 24 57 72 52 53Insccticides ... 42 29 45 41 43 28 80 35 54Modern plow ... 41 45 34 77 66 59 90 76 673 or 5-tine cultivator I 3 1 4 3 1 6 4 3Animal innoculation 23 21 10 24 27 27 52 28 14

II ealthPRAI latrines 1 2 1 4 2 6 3 3 0Smokeless (-hula 0 0 0 0 Cl 0 0 0 0Bed-bug killer ... I5 10 8 13 42 8 33 5 9Modern child-birth

practices ... 1 2 2 3 6 6 8 16 9TABC ... 44 52 55 62 51 24 70 44 38


Villages generally had a considerably higher percentageof adoption in 1967 than either literacy or controlvillages. Literacy villages were sometimes higher thancontrol villages but not markedly so. The contrasts inpercentages for knowledge were less striking.

The three sets of villages were roughly similar inknowledge of practices in 1964 (Table 8). Radiovillages had the highest per cent knowledge in thatyear on seven of the 15 practices while in 1967 theyhad the highest per cent knowledge on nine of the15 practices. Seven of the 15 practices had 'peaked'in knowledge at over 90 per cent in 1967 in one ormore of the three sets of villages.

With much less 'peaking' occurring to flattenpossible comparisons, the percentages for adoption ofthe individual practices show clearly the leadingposition of the radio villages (Table 9). Only formodern child-birth practices, at a low level of adoption,were the figures lowest 1hr the radio villages. Forthree or five-tine cultivator and PRAI latrines, thepercentages were also at low levels and the position ofthe radio villages was not clearly leading. As we havepointed Out earlier, there were no adoptions of thesmokeless chula in any villages in any year. For allother practices, the radio villages had the highest percent adoption.

KNOWLEDGE AND ADOPTION OF FAMILY PLANNING

India's population is conservatively estimated tobe about 520 million. Recognition of the mountingconcern with which this problem of population sizeand growth is viewed may be had ill the intensifiedaction programs of the Government of India, theUnited Nations and other interested agencies. Films,


slides, posters and the family planning red triangle symbolmay be seen everywhere in urban areas. Althoughsuch promotional efforts are much less in view in India's520,000 villages, we found a steady rise in our studyvillages in knowledge of family planning (Table 10).

Table 10: Per cent of Respondents Who Knew AboutFamily Planning, Communication Treatmentand Control Villages Compared, 1964, 1966 and1967

Treatment 1964 1966 1967

Radio farm forum ... 37.97 82.28 83.54

Literacy classes ... 45.13 84.96 86.73

Control ... ... 29.86 73.56 86.21

k -In 1964, knowledge of family planning rangedfrom 29.86 per cent in the control villages to 45.13per cent in the literacy villages. Reflecting the in-creased publicity given to family planning in 1965 and1966, figures approximately doubled for all three setsof villages in 1966. There were much smaller increasesfrom 1966 to 1967. Surprisingly, the control villagesmade the greatest gain in knowledge of family planningfrom 1964 to 1967, and the radio villages had thelowest percentages of knowledge of family planning in1967. However, differences among the three sets ofvillages in 1967 were not significant. It is apparentthat there was wide-spread knowledge of family planningin 1967 and that the communication treatments didnot produce lasting differences. Viewed another way,we may say that the regular block programs and otherchannels of communication had brought the informationto all of the villages, including the control villages.


Before we present additional family planning data,we would like to indicate that these data are probablyconservative. That is, it is likely that they under-estimate the amount of knowledge and adoption offamily planning. This is not only because familyplanning is still an essentially private and non-discussedtopic for many villagers but also because someone otherthan the respondent and the interviewer may havebeen present during the interviewing. It is likely thatthis stifled some affirmative responses. Althoughwe would have preferred an isolated interview situation,crowded village conditions and the constant presenceof others made this difficult to attain without risk oflosing rapport and arousing suspicion. We wished tomaintain the excellent rapport established in 1964 and1966 and followed the practice used in those years ofnot insisting on complete privacy for the interview.Also, the primary focus of the study, especially in thecommunication treatments, was on agricultural prac-tices. Therefore, we accepted less than ideal conditionsfor interviewing on family planning questions.

The format for determining whether a respondentknew about family planning was to first ask 'Have youever heard about family planning?' If the responsewas 'yes', which was the case for most respondents,then they were asked, 'What different methods of familyplanning have you heard about?' In this way we couldverify by a correct answer whether knowledge was interms of an actual method or if only the rubric offamily planning was familiar. We could not deter-mine, of course, whether the knowledge was functionalbut we could make some inference from the fact thatfew respondents had tried or were using any methodof family planning, as we will show below.


The radio villages knew about significantly moremethods of family planning than either the literacy orcontrol villages (Table 11). The control village average,1.62, was slightly higher than the literacy villageaverage, 1.49.

Table 11: Average Number of Family Planning MethodsKnown, Communication Treatment and ControlVillages Compared, 1964, 1966 and 1967

Treatment 1964 1966

1.95

1.90

1.36

1967*

2.04

1.49

1.62

Radio farm forumLiteracy classesControl

t Data were not obtained in 1964.* Differences between radio and literacy and between radio and control

significant at .05 level.

Respondents predominantly mentioned the methodsof vasectomy and IUCD (Intra-Uterine ContraceptiveDevice), (Table 12). These data are for all six villagescombined. It is likely that indigenous methods, in-cluding withdrawal and abstention, might have beenmentioned more often had the interviewing situationbeen focussed on family planning. The responsesshown in Table 12 clearly indicate, however, thatknowledge of currently advocated practices is widelydisseminated and that the problem for the future is notso much of information as of motivation. .

Both the IUCD and vasectomy arc methods inwhich user technique is not a problem. Unknown forthe present sample is how successful respondents wouldbe in practising methods dependent upon propertechnique for success. This is probably a crucial factorand further advances in technology arc needed todevelop methods less subject to failure but which alsopermit desired reproduction.


Table 12: Knowledge of Different Family PlanningMethods, Communication Treatment and ControlVillages Combined, 1967

Number whoMethod mentioned method*

Vasectomy 229

IUCD (Loop) 124

Sheath (Condom) 28

Vaginal foam pills 23

Female sterilization 27

Abstention

Oral pills 9

Rhythm 6

Diaphragm 2

Withdrawal

Other methods 2

* Respondents were first asked if they had heard about family planning. Ifthey responded 'yes', they were then asked 'What diarent methods offamily planning have you heard about?' Responses were recorded onlyif the respondent himself mentioned the method; respondents were not askedif they knew about each method. Some respondents mentioned morethan one method,

Table 13: Per cent of Respondents Who Had AdoptedFamily Planning, Communication Treatmentand Control Villages Compared, 1964, 1966 and1967

Treatment 1964 1966 1967*

Radio farm lbrom ... 6.33 8.86 11.39


Control 1.15 3.43 5.75

* Difference between radio and literary signi flea lit at. . 05 level.

Few respondents practised family planning(Table 13). The percentage fbr the radio villages was

COMMUNICATION IN INDIA

significantly higher than that of the literacy villageswhich declined from 1966 to 1967. The latter werelower than the control villages. However, the radiovillages were considerably higher than the other villagesbefore the communication treatments began, and wedo not attach much importance tO the differences amongvillages in 1967. Perhaps the most important con-clusion is that levels of adoption of family planning arclow but seem to be rising slowly. There does not seemto be much promise for rapid reduction of the birth-rate until death-rates, particularly infant mortality rates,decrease still further and there are economic incentivesto have smaller families.

ANALYSIS VARIABLES ASSOCIATED WITHADOPTION

So far, in our discussion of the continuing effects ofthe communication treatments conducted in 1965, wehave examined only knowledge, trial and adoption ofthe practices selected for study. Now, we will look at

ITsome variables commonly associated with practiceadoption to see if there were changes in them in 1967which will help us to evaluate the communicationtreatments. These variables in themselves can beconsidered as measures of social change. Most of them,like extension knowledge and contact, functionalliteracy, and exposure to mass media, arc consistentlyfound to be positively related to adoption of inno-vations and arc often used as indicators of modernization.It seems obvious that they would be related to techno-logical change and modernization. A principal meansof learning about new practices is through some exten-sion organization or via print and other mass media.Personal communication channels are also important,


but the inibrmation for the leaders and the innovatorscomes chiefly through imperson.1.1 contact with thelarger society. From the innovators, information andpersuasion flow outward to the bulk of the cultivatorsand reinforces the impact of the mass media.

As with knowledge, trial and adoption, we presentaverages and percentages to compare the radio,literacy and control villages. We will present data Ib1964, 1966 and 1967 but 'bens our attention on the1967 figures, for which we make statistical comparisons.Having had the advantage of the earlier study, com-paring the 1964 and 1966 data," it hardly seemed fairto formally hypothesize direction of differences. Itseemed likely that the trends shown then in favor ofthe radio villages would continue and this is in factwhat we th1.111(1. The results are not so conclusively infavor of the radio villages as with the knowledge, trialand adoption data, but then some of the variables aremuch more complex and would not be influenced bythe communication treatments as readily as wouldadoption behavior.

EXTENSION KNOWLEDGE AND CONTACT

We measured extension agent knowledge by makinga simple index of number of change agents known(Table 14). Radio village average scores were signi-ficantly higher than those ibr literacy and controlvillages.

For extension agent contact we also constructeda simple index, using the same agents on which theextension knowledge index was based (Table 15).Here, however, the results were opposite to those wehad expected. Control and literacy villages had signi-II. mu


ficandy higher average index scores than the radiovillages. This reversal of what had been a consistenttrend in the direction of higher averages for the radiovillages seems attributable to two factors. First, a verypopular village level worker (VLW), lived in one ofthe control villages and second, the control villages weremuch nearer to block headquarters. These facts arereflected in the 1964 figures (Tables 14 and 15). Controlvillages had a somewhat higher average score forextension agent knowledge (Table 14), and a substan-tially higher average score for extension agent contactin 1964 (Table 15). Average scores increased consider-ably for all three sets of villages from 1964 to 1967 onboth variables, however, and the increases from 1964to 1967 are highest for radio villages and lowest forcontrol villages.

Table 11: Average Score, Index of Change Agents Known,Communication Treatment and Control VillagesCompared, 1964, 1966 and 1967f

Treatment 1964 1966 1967*



Control ... 4.22 4.63 6.55

t Change agents were block development officer, midwife, block doctor,village level worker, agricultural development officer (ADO) for agriculture,ADO for co-operatives, ADO for panchayat (village council), veterinarydoctor, school-master, pradhan (president of village panchayat), block prarnukh(head of block council). A respondent received a score of 'I' if he knewthe role of the change agent and a score of '2' if he knew the name of thechange agent. Scores thus had a possible range of (1 to 22.Difference between radio and control significant at .03 level.

1;


Table 15: Average Score, Index of Number oc TimesTalked with a Change Agent, CommunicationTreatment and Control Villages Compared,1964, 1966 and 1967f

Treatment 1964 1966 1967*

Radio farm forum 6.58 7.71 10.75


Control 10.81 12.77 13.23

t The same 11 change agents listed in the footnote to Table 14 were usedfor this index of change agent contat.t. For the VLW and school-master,the number of times talked was divided by 10. For the other nine officialsthe actual number reported was given, with a maximum of nine. Thissimple weighting procedure simplified coding and prevented undue weightingby a very popular official or one who lived in the village. Scores thus hada possible range of 0 to 99.

* Differences between radio and literacy and between radio and controlsignificant at .05 level.

We have chosen not to present difference scoresshowing differences between 1964 and 1967, becauseof the difficulty of 'peaking' on some variables, as weobserved in our discussion of knowledge of agriculturalpractices. However, it seems clear that the differencesin extension agent contact reflect a condition whichexisted in the study villages when the benchmark surveywas taken. We conclude that the general tendency ofradio villages to show higher averages and to makegreater increases from 1964 to 1967 is not altered bythese findings on extension agent contact. A correlativeconclusion is that individual extension agents can maketheir presence felt and that contact with one impression-making agent can precipitate contact of cultivatorswith other agents. The VLW in the control villagesacted as a catalyst for additional extension agent contact.It remains true, of' course, that the radio villagesgenerally scored higher on adoption and other variables.


Proximity to block headquarters'= and the popular VLWin the control villages did not compensate for theapparent effect of the radio farm forum in the radiovillages.

LITERACY AND MASS MEDIA CONTACT

Literacy is one of the chief concerns of developingnations. Almost always a matter of national pride,the amount of scarce resources which should be allocatedto it remains a matter of some debate. Howeverviewed, literacy is an important indicator of develop-ment and we regard it so here. Promoting literacy,of course, was one of the programs by which we soughtto induce adoption of innovations. As might beexpected, the literacy villages had the highest averagenumber of words correct on our seven-word test offunctional literacy (Table 16 ). However, literacyvillages scored only slightly higher than the radiovillages. Both literacy and radio villages scored signi-ficantly higher than the control villages. We mightpoint out that the literacy villages started with thehighest average score in 1964, rose in 1966, but .thenshowed the same score in 1967. The radio and controlvillages, on the other hand, continued to make modestscore increases in each year.

We are inclined to view these findings as anindication that literacy training programs not only donot make a very good vehicle for promoting acceptanceof innovations, but that they require a reat deal ofinput for a payoff in increased literacy. Tie regramswere costly in time and money inputs. Time input12. Proximity to block headquarters was not significantly associated with

village adoption of agricultural practices in a study of 103 villages in threestates of India in 1966. Sec F. C. Fliegcl, I'. Roy, L. K. Sen and J. E.Kivlin, Agricultural Innovations in Indian Villages, Hyderabad: NationalInstitute of Community Development, 1963, p. 46.


for literacy classes were about 60,000 trainee man-hours and above 2,000 teacher or organizer man-hours.Money costs were about Rs. 14,000. By comparison,time inputs for the radio farm forums were about 2,300listener man-hours and about 200 trainer or organizerman-hours. Money costs were about Rs. 10,000.This is not to deny that there is often much interest inliteracy, or that increases in literacy are not in thelong run desirable and, in fact, are probably inevitable.However, it does indicate that literacy improvement isslow in product and return and should be weighedcarefully in the allocation of resources."

In all six villages combined, there were 186 respon-dents who had no words correct in 1967 on our seven-word test and hence, were functional illiterates by thisdefinition. There were 78 respondents who had sixor seven words correct and would be considered func-tionally literate. Percentages of literates and illiteratesamong the three sets of villages, show the same patternas the average scores in Table 16. That is, literacy

Table 16: Average Number of Words Correct (functionalliteracy test of seven words), CommunicationTreatment and Control Villages Compared,1964, 1966 and 1967

Treatment 1964 1966 1967*

Radio farm forum ... 1.25 2.01 2.33Literacy classes ... ... 1.92 2.69 2.69

Control ... 0 80 0.95 1.15

* Differences between radio and control and between literacy and controlsignificant, at .05 level. However, the distribution is badly skewed as therewere 186 out of 279 respondents who had no words correct.

13. For two discussions of this view, see J. K. Galbraith, Economic Development,Boston: Houghton Mifflin Company, 1964, pp. 77-86. and F. C. Fliegel,"Literacy and Exposure to Agricultural Information: A Comparison ofsome Indian and Brazilian Data", Behavioural Sciences and Community

. Development, 1 (September, 1967), pp. 89-99.


villages had the highest percentage of literates, followedby the radio villages and then by the control villages,which had the highest percentage of illiterates.

Mass media contact was measured by per centwho listen to radio, number of papers read last week,and number of times seen a commercial film last year(Tables 17, 18 and 19). As might be expected, theradio villages showed the highest percentage of radio-listening but differences among the three sets of villageswere not significant (Table 17). Radio-listening hadmore than doubled between 1964 and 1967 for allvillages. The radio farm forum villages rose from23 per cent who listened in 1964, to 86 per cent in 1966,reflecting the communication treatments in 1965.However, also reflecting the termination of the forums,the percentage dropped to 76 per cent in 1967. Thesehigh rates of radio listening probably reflect the broadquestion which was asked, 'Do you listen to the radio?'Listening may well be infrequent.

Table 17: Per cent of Respondents Who Listen to Radio,Communication Treatment and 'Control VillagesCompared, 1964, 1966 and 1967

Treatment 1964 1966 1967

Radio farm forum ... 22.67 86.18 75.70Literacy classes ... ... 30.22 58.75 66.39Control ... 35.46 59.69 71.00

The average number of times the respondentreported having read a paper in the previous week waslow for all villages, less than once a week (Table 18).The literacy villages showed a spurt in 1966 but fellback in 1967 to almost the 1964 level. However,literacy villages showed the highest average in 1967.


Figures for the control villages were practically static.As 186 or two-thirds of the respondents in all six villagescombined were functional illiterates according to ourseven-word test, these low levels of reading were to beexpected. To remove the effect of this large numberof illiterates we re-calculated the averages for papersread the previous week for literates only and showthese averages in the last column of Table 18. Theaverage for the literacy villages was again highest.The control villages showed a higher average than theradio villages but this is probably a function of smallsub-sample size. There were only 12 literate respon-dents in the control villages.

Table 18: Average Number of Times Read a Paper LastWeek, Communication Treatment and ControlVillages Compared, 1964, 1966 and 1967

1967 All 1967 literateTreatment 1964 1966 respondents* respondents

only

Radio farm forum... 0.11 0.20 0.43 1.36

Literacy classes 0.68 0.99 0.75 2.05Control ... 0.24 0.1.3 0.24 1.75

* Difference between literacy and control significant, at .05 level. However,the distribution is badly skewed as there were 208 out of 279 respondentswho had read no papers.

Reading levels were relatively low among theliterate respondents. Thirty per cent of these in allsix villages combined had read no paper in the weekprior to the time of the interview. It is apparent thatprint mass media have decided limitations. Althoughmany leaders and innovators can read and do pass oninformation to promote acceptance of improvedpractices, this communication channel cannot make

38 COMMUNIaATION IN INDIA

its impact directly upon the bulk of the cultivators.Here, of course, is a dilemma, for we pointed out abovethat literacy training is a costly input. It is likely-Thatenhanced use of print mass media will have to waitfor the larger percentage of children now attendingschool to move into adult roles. This circumstance,combined with a selection factor favoring the morerational, commercialized and literate cultivators willprovide an increased audience for print forums andpractice promotion.

The average number of fillies respondents ;ladseen a commercial film in the year previous to theinterview was also very low (Table 19). Differencesbetween radio and literacy villages and between controland literacy villages were significant but these differ-ences should not be considered important as thedistribution is badly skewed. There were 207 out ofthe 279 respondents (74 per cent) in all six villagescombined who had seen no films. The only reasonableconclusion is that levels of film-viewing are low andthat there is not much indication that they will increasein the near futut,.. The six villages vary somewhat inaccessibility to Lucknow but it is doubtful that proximitywill become very important in the future. Cultivatorsare more likely to be restricted by their farming opera-tions and by lack of disposable income than by trans-portation or proximity problems. This apparentelimination of commercial films as a direct suasive massmedia communication channel for this sample probablyhas greatest implications for family planning promotion.Information about family planning is much moreadaptable to mixed, rural-urban audiences than is thepromotion of agricultural practices. The most likelyuse of films as a promotional agent is via leaders and


innovators, who do tend to view films in larger numbersand whose film attendance is associated with villageadoption of agricultural innovations."

Table 19: Average Number of Times Seen a CommercialFilm Last Year, Communication Treatmentand Control Villages Compared, 1964, 1966 and1967

Treatment 1964 196G 1967*

Radio farm forum .35 .30 .90

'Literacy classes ... .51 .36 .45

Control .99 .61 1.05

* Differences between radio and literacy and between control and literacysignificant, at .05 level. However, the distribution is badly skewed as 207out of 279 respondents had seen no films.

ATTITUDES TOWARD PROGRAMS OF CHANGE

Most programs of change, almost by definition,can be expected to arouse some opposition. There areusually individuals and groups who feel that they willsuffer if change comes. For some, the advantages ofthe status quo are very real. For example, if a wife goesto literacy classes when most wives in the village arcilliterate, her husband may suffer much ridicule ordistrust by his peers. We were interested, then, inexamining attitudes toward programs of change. Theseattitudes can be taken as an indication of communitysupport for social. change programs or as an indicatorof village norms and social change itself. They areonly expressions of opinion however, and actual be-havioral support for change programs would be themore valid indicator.

We constructed a simple index of attitudes towardprograms of change and show average scores for the14-. F. C. Ilk -gel el al.. Agricultural Innovations . , p. 50.


index in Table 20. Figures are given for 1966 and1967 only, as these data were not obtained in 1964.The responses indicate widespread encouragement forthe three actions about which we asked the cultivators :trying a new method of farming, bringing up childrenin a modern way and a housewife getting more schooling.The last question, about the housewife, received thelargest number of unfavorable responses but even here72 per cent of the responses were favorable, in all sixvillages combined. There were 174 or 62 per centrespondents who answered 'would encourage' to allthree questions. Radio villages had the highest averagescore, followed by literacy and then by control, butdifferences were not significant. All three sets of villagesshowed increases in average scores from 1966 to 1967.

Tat e 20: Average Score, Index of Attitudes Toward Pro-grams of Change, Communication Treatmentand Control Villages Compared, 1964, 1966 and1967*

atment 1964 1966 1967

1 lio farm :orum

acy classes

Control

1 4.43

4.04

4.63

5.30

5.16

4.90

* The index was composed for the following three items, scored '0' for`discouraged'; 'I' for indifferent% and '2' for 'encourage'. In your com-munity would most people encourage or discourage . . (1) a man whotried a new method of farming . . . (2) a woman Who brought up herchildren not in the traditional way but in a modern way . . (3) a house-wife getting more schooling. Scores could thus range from 0 6.

t These data were not obtained in 1964.

Taken at face value, these highly favorable res-ponses indicate widespread social acceptance of adoptionof new practices. As we have shown, however, actual


adoption behavior has been far short of this seeminglyuniversal acceptance. We have shown adoption ofagricultural, health and family planning practices tobe at quite low levels. Even if allowances arc madefor non - availability; of supplies and services, for highinfant mortality rates, and for other relevant andrational reasons for non-adoption, actual adoptionbehavior lags far behind opinions of encouragement forprograms or actions which involve change. Theseexpressions of favorable opinions do represent anadvantage, however. Cultivators do consider it sociallyacceptable to give such favorable opinions and thisundoubtedly makes the task of the change agent easier.

OTHER MEASURES OF SOCIAL CHANGE

In this section, we present data on four additionalvariables which are usually associated positively withadoption of agricultural innovations and which we regardhere as indicators of social change in our evaluation ofcommunication treatments. The first variable is politicalknowledgeability (Table 21).

Table 21: Average Score, Political Knowledgeability Index,Communication Treatment and Contrcl VillagesCompared, 1964, 1966 and 1967t

Treatment 1964 1966 1967



Control ... 1.77 2.46 2.21

t The political knowledgeability index was composed of the following fiveequally weighted items: Who was Gandhiji; who was Jawaharlal Nehru;who is the chief minister of Uttar Pradesh; who is the prime minister ofIndia; who is the MLA of your area (member of Legislative Assembly).


We constructed an index of five equally weighteditems on knowledge of national, state and area politicalfigures. Such knowledge is presumed to reflect aware-ness and interest in national and regional affairs inparticular and in the world beyond the village ingeneral. Radio villages again had the highest score,2.73, followed by literacy villages at 2.62 and controlat 2.21. These differences were not significant. Allthree sets of villages showed increases in scores from1964 to 1967. Respondents more often knew aboutthe two deceased national leaders, Gandhi and Nehru,yet it is surprising to note that in all six villages com-bined there were 40 per cent who did not know ofthem. These two leaders are the most prominent inmodern Indian history and their pictures arc commonlyhung in public and commercial buildings as well as inprivate homes. A larger percentage did not knowwho was their state chief minister or member of theLegislative Assembly.

Formal social participation is the second of theother measures of social change for which we obtained-data (Table 22). Social participation is likely to resultin wider horizons, new information and perhaps increasedopportunities to obtain the supplies and services neces-sary to adopt an improved practice. We constructedan index by scoring '1' for membership and '2' forholding office in eight organizations in the village andoutside the village. Radio villages scored significantlyhigher than literacy and control villages, with literacyvillages dropping below the control villages in 1967.However, formal social participation was at low levelsand there were 101 respondents in all six villagescombined who had no membership and held nooffice.


Table 22: Average Score, Formal Social ParticipationIndex, Communication Treatment and ControlVillages Compared, 1964, 1966 and 1967'

Treatment 1964. 1966 1967*

Radio farm forum ... 1.19 1.12 2.09Literacy classes ... ... 0.72 1.13 0.81Control ... 0.69 {.11 1.15

't The formal social participation index was constructed by scoring '1' formembership and '2' for holding office in the following eight organizations,plus a seldom mentioned 'others': co-operative society, panCliktyat, youthclub, night school, women's club, balwadi (children's school), defenceforce, school committee.

* Differences between radio and literacy and between radio and controlsignificant at .05 level. The distribution was skewed in that 101 respondents inall six villages combined had zero scores, with no memberships or offices held.

Our next measure is of secularism, the orientationtoward secular rather than traditional values (Table 23).Like the other variables we discuss in this section, ahigh score on secularism bespeaks a generally rationalset of values, orientation to the larger society ratherthan to the village, and predisposition to accept somesocial change. Radio villages were again found tohave the highest score, followed by the literacy villagesand then by the control villages. These differences in1967 Nyere not significant, however. There were onlyslight increases in all three sets of villages from 1961 to1967. The distribution of secularism scores in 1967was reasonably normal with only 17 respondents scoringzero, indicating a traditional response to all questions.No respondent had a secular response to all questions.The item which stood out as inducing a traditionalresponse was 'If your son wanted to marry a low castegirl, would you allow it?' Only 33 or 12 per cent ofthe respondents answered in the affirmative to thisquestion. In general, all villages scored about midwaybetween the extremes of secular and traditional scores.


Table 23: Average Score, Secularism Index, Communica-tion Treatment and Control Villages Compared1964, 1966 and 1967f

Treatment 1964 1966 1967



Control 2.88 2.52 3.39

f The secularism index was composed of the following seven equally weighteditems in which each 'secular' response was scored '2' for a range of 0 to 14:Can evil spirits cause disease; Have you made sacrifice to prevent sickness;Should Harijans be allowed to draw water from all wells in the village;Should Harijans and other children take meals together in school; If yourson wanted to marry a lower caste girl, would you allow it; Do you thinkHarijans should be allowed to worship in all temples of the village; Ia youropinion is an illiterate village Brahmin superior to a lower caste B.A. orM.A.? 'Traditional' responses were scored '0'. Thcrc were a very few`no answer' responses which were scored '1'.

The final measure of social change which wepresent is an index of educational and occupationalaspirations for children (Table 24). Again, differenceswere not significant and all three sets of villages scoredsomewhat below the midpoint of the index. Literacyvillages scored highest, with the control villages scoringslightly higher than the radio villages. All three setsof villages showed somewhat increased scores from 1964to 1967 indicating a gradually rising level of aspirations.This increase in aspiration is taken as an indicationthat present views of cultivators are somewhat optimisticand that this in turn will contribute to a favorableattitude toward change and development programs.Family and children rank high in the hierarchy ofvalues in India and the questions we asked focussed oneducational and occupational aspirations which lookedbeyond the village. If these aspirations are to beachieved, of course, heavy investments must be madeto provide schools and jobs.


Table 24: Average Score, Index of Educational and Occu-pational Aspirations, Communication Trettr-meat and Control Villages1966 and 19671

Compared, 1964,

Treatment i964 1966 1967



Control 1.58 2.14 2.01.1011,

t The aspiration index was constructed by using the following five equallyweighted items, for a range of from 0 to 5:

(1) Education desired for sonScore: 0 to 8 years = 0; 9 years and above = 1

(2) Is it possible to educate your son (up to year indicated)Score: No = 0; Yes = I

(3) Occupation desired for first sonScore: Agriculture, fate, can't say = 0;

Other occupations = 1

(4) Occupation desired for second sonScore: Same as for (3) above

(5) Education desired for daughterScore: 0 to 5 years = 0; 6 years and above = 1

Cutting-points for scoring were determined in 1964 and 1966 surveys butproved also to be appropriate for index construction in 1967.

Our conclusions in regard to these four measuresof social change are that they indicate considerableprogress in development programs and augur well forcontinued progress in the future. However, they donot importantly distinguish between the communicationtreatments or between the communication treatmentvillages and the control villages. When consideredwith the highly favorable attitudes toward programs ofchange discussed in the last section, the findings weobtained suggest that change agents who have well-conceived programs can expect reasonable co-operation


and success. Programs will have to be relevant tovillage conditions; promises of supplies and servicemust be kept up unless mitigating circumstances likeflood or disaster arc clearly evident; technical advicemust be competent and timely. If these conditionsarc met, then, it would seem that progress can bemade. There is some political knowledgeability amongthe villagers, especially the leaders. Villagers arereasonably secular in their orientations and have risingoccupational and- educational aspirations for theirchildren. Social participation is low but is also rising.We interpret these findings favorably and urge continuedefforts where possible which will result in higher village'levels for these variables.

VILLAGER EVALUATION OF THE COMMUNICATIONTREATMENTS

Our final section considers villager evaluation ofthe communication treatments. We wanted to deter-mine if villagers could accurately recall the programsof radio farm forums and literacy classes. We alsowanted to know, some year-and-a-half after the ter-mination of the .programs, whether villagers thought thatthe programs were useful and, if so, in what way.

Most respondents could accurately recall theprograms (Table 25). Eighty-four per cent of therespondents in the radio villages gave a correct descriptionof the programs as compared to 91 per cent whocorrectly recalled the literacy classes. The percentagesof respondents who thought that the program was notuseful were low and also quite similar. Eight percent of respondents in the radio villages gave thisresponse as compared to 12 per cent in the literacyvillages.


Table 25: 'Were the Programs in 1964 Useful? If so, inwhat way?' Radio Farm Forum and LiteracyClasses Compared, 1967

Radio Literacy

per cent*Not useful 8 12

Information about new agricultural pfactices 63 7

Market news, weather forecasts 17 0Education, learn to read and write 12 81

100 100

No answer or no recall (number)t 19 22

* Percentages arc of the first specific response only. There were few multipleresponses spanning the three categories of usefulness.

t Eighty-four per cent of the respondents in the radio villages gave a correctdescription of the programs as compared to 91 per cent who correctly recalledthe literacy classes.

Responses as to how the programs were usefulapparently followed closely the perceived principalfocus of the programs. Most of the radio farm forumswere devoted to broadcasts and discussions of agriculturalpractices and 63 per cent of the responses in the radiovillages indicated that information about new practiceswas the chief usefulness. Only 7 per cent of culti-vators in the literacy villages gave this response althoughthe literacy classes reading materials also stressed newpractices. Learning to read and write was one mainfocus of the literacy classes and there were 81 per centin the literacy villages who reflected this by responding`learn to read and write', 'get educated' or 'learn'.There were only 12 per cent in the radio villages whogave the general response of 'learn' without mentionof specific information. Seventeen per cent of respon-dents in the radio villages mentioned market news orweather forecasts, while no respondents in the literacyvillages did so.


These findings indicate a generally favorableresponse to the communication treatments and areasonably lasting impression. They also indicate thatthe intent of the treatments, to disseminate informationand to promote acceptance of improved practices, wasmuch better served by the radio farm forums than bythe literacy classes. This villager evaluation, ofcourse, confirms what we have shown previously. Theradio farm forums were more efficacious in practicepromotion and in bringing about higher levels ofvarious social change variables. As we pointed outearlier, they were also more efficient in that time,labor and money inputs were considerably less. Werecognize that literacy training, even for adults, is along-term investment. However, we have checkedspecific results of literacy training over time and theywere not striking. Advantages in allocation of scarceresources would seem to be with the radio farm forums.

SUMMARY AND CONCLUSIONS

We have reported on a continuing study to deter-mine the effects of two communication treatments uponadoption behavior and other measures of social change.We found that one of the communication treatments,radio farm forums, was considerably more effectivethan the other treatment, adult literacy classes. Bothsets of communication treatment villages tended toshow more change than control villages, in which onlythe usual community development programs wereundertaken.

The study on which our report is based is a follow-up of part of a two-nation comparative study initiatedin 1964 in India and Costa Rica. Our report dealsonly with Indian data. In 1964, a benchmark survey


was made of selected villages in India to establish thelevel of knowledge, trial and adoption of certain agri-cultural, health and family planning practices. In-formation was also obtained on variables often associatedwith adoption. Then, in 1965, an experiment wasconducted in which the communication treatments ofliteracy classes and radio farm forums were applied totwo pairs of villages but not to a third pair on whichbenchmark data were also obtained. In 1966, are-survey was made of the same respondents to examinethe initial effects of the treatments. In 1967, anotherre-survey was made, again of the same respondents, toexamine the continuing effects. Attrition due tovarious factors resulted, in 1967, in a sample for thepresent study of 279 respondents. All were males whohad cultivated some land during the time of the studyand who had been interviewed in 1964, 1966 and 1967whether or not they participated directly in the com-munication treatments. Data which we report wereobtained by asking identical questions in all three years,so as to yield comparable results.

We were mainly interested in examining our datafor possible individual and village changes of a con-tinuing nature. Consequently, we have stressedsimilarities and differences as we found them in thethree sets of villages in 1967. The villages variedsomewhat in 1964 but were reasonably at the same levelthen on almost all of the variables which we investi-gated. In 1967, the radio farm forum treatmentvillages showed significantly more progress than theliteracy and control villages. This was true whethermeasured by adoption of practices or by other variables.There was some tendency for the literacy villages toscore higher than the control villages but usually these


differences were not statistically significant. In onlytwo of the 1967 comparisons did the control villagesscore highest. We present these comparisons visuallyin the form of rankings (Tables 26 and 27), and discussthe variables briefly below.

Table 26: Ranking of Communication Treatment andControl Villages, Knowledge, Trial and AdoptionVariables, 1967?

VariablesRadio farm

forumvillages

Literacyclassesvillages

Control (notreatment)

villages

Average number ofagricultural practicesknown

Average number ofagricultural practicestried

Average number ofagricultural practicesadopted

I-

Average number ofhealth practicesknown

Average number ofhealth practicestried

Average number ofhealth practicesadopted

1--

Per cent who knewabout familyplanning 3''

..01

Per cent who hadadopted familyplanning

2Average number of

family planningmethods known

2t 'Ranking' here refers to the position of the three sets of villages in terms of

averages or percentages. Thus in the first item, the radio farm forum villageshad the highest average number of agricultural practices known, followedby literacy classes villages with the second highest and by control villageswith the third highest. For ^eual figures and statistical comparisons consultthe appropriate Table in the text.


Table 27: Ranking of Communication Treatment andControl Villages, Selected Variables of SocialChange, 1967

VariablesRadio farm Literacy Control (no

forum classes treatment)villages villages villages

Average score, indexof change agentsknown

Average score, indexof change agentcontact 3-'

Average score, testof functional 2 --literacy

Average numberof papersread

Per cent who 1,listened toradio

Average numberof films 2-seen

Average score, indexof attitudes towardprograms of change

Average score, indexof politicalknowledgeability

Average score, indexof socialparticipation

Average score,index ofaspirations

Average score, index oforientation towardsecularism

1-_______

1--

-2

- 2

-2

-3

)


in 1967 were not significant but radio villages knewsignificantly about more methods of family planningthan either the literacy or control villages. Respondentspredominantly mentioned the methods of vasectomyand IUCD. (Intra-Uterine Contraceptive. Device). Fewrespondents practised family planning and differencesamong the villages are probably not reliable. Levelsof adoption of family planning seem to be rising slowly.

In addition to knowledge, trial and adoption ofimproved practices, we examined eleven variablescommonly associated with practice adoption to see ifthere were changes in them in 1967 which would helpub to evaluate the communication treatments. Theresults generally showed averages and percentages ofradio villages to be higher than those of the literacyand control villages. However, they were not as con-clusively so as they were in the knowledge, trial andadoption data. Some of these variables are consider-ably more complex and are more difficult to measure.

Index scores for extension agent knowledge weresignificantly higher for radio villagei than those forliteracy and control villages. For extension agentcontact, however, these positions were reversed. Controlvillages scored highest. We attributed this reversal ofwhat had been a consistent trend in the direction ofhigher averages for the radio villages to greater proximityof control villages to block headquarters and to theeffect of a very popular VLW living in one of thecontrol villages.

Literacy villages scored highest on a seven-wordtest of functional literacy, but not significantly highertitan the radio villages. Two-thirds of all respondentswere functionally illiterate. We concluded that literacyclasses not only do not make a very good vehicle' for pro-

f.


moting acceptance of innovations, but that they requirea great deal of input for a payoff in increased literacy.

Mass media contact was measured by per centwho listened to radio, number :of newspapers read lastweek, and number of times seen a commercial filmlast rear. Radio villages showed the highest per-centage of radio-listening but differences among thethree sets of villages were not significant. Levels ofnewspaper reading were quite low and there were notsignificant differences among the three sets of villages.Three-fourths of all 279 responderits had not seen acommercial film in the previous year and thus differ-ences among the sets of villages cannot be consideredvery valid.

There was widespread favorable response to threequestions about encouragement or discouragement ofprograms of change, with radio villages again scoringhighest. As actual adoption behavior was shown tolag far behind this near-universal social acceptance, weconcluded that the chief result of such favorable opinionswas to perhaps make the task of the change agenteasier.

Radio villages scored highest on an index ofpolitical knowledgeability, testing awareness of state andnational leaders, but differences were not significant.State leaders were less well-known than national leaders.

Radio villages scored significantly higher thanliteracy and control villages on an index of formalsocial participation but there were 101 respondents inall six villages combined who had no membership andheld no office.

An index attempting to measure secularism alsodisclosed radio villages to have the highest scores butdifferences were not significant and there were only


slight increases in all three sets of villages from 1964 to1967.

The final measure of social change which wepiesented was an index of educational and occupationalaspirations for children. Literacy villages scored highestbut differences were not significant.

It is apparent that in our study the communicationtreatment of radio farm forum was more effective thanthe treatment of literacy classes. Literacy villagestended to score higher than the control villages but thedifferences were often not significant.

In another time and place, with differently-runprograms, different results may be obtained. As socialexperiments go, however, subject as they are to amultitude of intrusive and uncontrollable factors, thisone is reasonably defensible. We believe that ourresults are as conclusive as one social experiment canbe. We would advise increased use of radio farmforums where they can be organized so as to elicitvillager participation. Forums should take up onlyhighly relevant and technologically sound practices.Practices like the smokeless chula should be avoided.Whatever the merits or demerits of the smokeless chulamay be, zero adoption in all three years in which wesurveyed seems sufficient evidence to regard it as aquestionable practice for promotional efforts.

Additionally, because we have shown substantialeffects of more indirect measures of social change, radiofarm forums should be integrated into an on-goingprogram that will capitalize on initial social gains ofthe radio forums. It may not be possible for everyvillage to attain its own individual, self-sustaining`take-off', but wherever possible, an upward spiralshould be encouraged. Our results indicate that


progress can be expected and that this predictabilityof success of a well-run radio forum should permitaccelerated progress.

We do not, of course, condemn or slight literacytraining programs. As we have stated, and the villagersthemselves indicated, literacy programs have a highsocial value. Ultimately, any country that desires ahigh level of development must attain a similar levelof literacy. For developing nations, however, resourcesare almost always scarce. Literacy programs, as ourexperience has shown, are expensive. Therefore, theyshould be considered either as a long-term investmentin development or as social welfare. It may be thatadditional discussion could be incorporated into theliteracy classes, or that the classes could be combinedwith another program, perhaps even with a radioforum. However, it is likely that the importanceattached to learning to read and write will remainparamount with villagers and that it will be difficultto transfer their motivation from literacy to adoptionof specific innovations.

ERIC Clearinghouse

SEP 1 2 1970

on Adult Education

document resume kivlin, joseph e.; and others ... · document resume. ac 008 385. kivlin, joseph...

Documents