III ’ , I II I I III II I I I II —— AN EXPLORATORY STUDY ON THE RELATIONSHIP BETWEEN MEDIA PARTICIPATION AND POLITICAL PARTICIPATION IN NATIONAL ELECTIONS ACROSS NONnCONIMUNIST COUNTRIES Thesis for the Degree of M. A. MICHIGAN STATE UNIVERSITY SHAH - FANG YIEN 1968 [.an ”3:19.394: State University MSU LIBRARIES 4—:— RETURNING MATERIALS: PIace in book drop to remove this checkout from your record. FINES wiTT be charged if book is returned after the date stamped beTow. ABSTRACT AN EXPLORATORY STUDY ON THE RELATIONSHIP BETWEEN MEDIA PARTICIPATION AND POLITICAL PARTICIPATION IN NATIONAL ELECTIONS ACROSS NON-COMMUNIST COUNTRIES By Shah-pang Yien This is an attempt to explore the relationship between media participation and political participation in national elections across 33 non-communist countries. Since some researchers have pointed out that there is a direct relationship between communications networks and awareness of res- ponsibilities of citizenship, and that media function most effectively only in the rapidly-modernizing nations, the purposes of the study were to provide the answers with evidence. Non-communist countries were first studied without breaking down into sub-groups and various correlational analyses were done in order to answer the first question in mind. Then the same non—communist countries were broken down into three groups on the basis of urbanization speed; namely, high, mo- derate and low urbanization speed countries. The analysis of the data shows that the two statements mentioned above do not hold for the non—communist countries included in this study. However, the findings are in accord with the directions pre— dicted. Another point which deserves mentioning here is that urbaniza- tion speed is a good indicator for studying, cross-nationally, the relationship between newspaper participation and political partici- pation, with radio participation held constant. AN EXPLORATORY STUDY ON THE RELATIONSHIP BETWEEN MEDIA PARTICIPATION AND POLITICAL PARTICIPATION IN NATIONAL ELECTIONS ACROSS NON-COMMUNIST COUNTRIES By Shan—pang Yien A THESIS Submitted to MICHIGAN STATE UNIVERSITY in partial fulfillment of the requirements for the degree of MASTER OF ARTS Department of Communication 1968 Accepted by the faculty of the Department of Communication, College of Communication Arts, Michigan State University, in par- tial fulfillment of the requirements for Master of Arts degree. Director of Thesis ACKNOWLEDGEMENTS I am deeply indebted to my academic and major advisor, Professor Hideya Kumata, Director of International Communica— tion Institute, who has provided, with great patience, conti- nuous advice and suggestion. Also, I have benefited from thoughtful and critical read- ings of the manuscript by Dr. Verling C. Troldahl and Dr. Bradley Greenberg. From both of them I have gained a direct and immeasurably important intellectual source for my research. Acknowledgement goes also to Miss Marilyn Zigmont who spent a whole weekend to help me prepare for the final draft. Finally, my gratitude to my wife, who proofread and typed the thesis for me. TABLE OF CONTENTS page ACKNOWLEDGEMENTS ... ............................ ..... i TABLE OF CONTENTS .................................... ii LIST OF TABLES ..... ..... ........... ........ . ..... . iii CHAPTER I: THE PROBLEM AND THE HYPOTHESIS ........... 1 Introduction .. ............... ............ 1 The Hurdles of Research ...... ......... ... 4 Problem and Rationale .................. 6 Problem and Assumption .................. 9 Hypotheses ....... ..... ...... ..... ... 12 CHAPTER II: THE METHOD . ..... ..... ..... ... ........ ... 14 Sampling Procedures .... ........ . ..... .... 15 Data Analysis and Statistics ............. 20 CHAPTER III: FINDINGS ....... ........ ................. 22 CHAPTER IV: CONCLUSIONS ............... ......... ...... 31 BIBLIOGRAPIIY 0.0.0.0... ....... ......OOOOOOOOOOOOOOOO. 37 TABLE II (a) II (b) II (0) III IV VI VII VIII LIST OF TABLES Media Participation and Political Participation,by Nation .... ............ . Media Participation and Political Participation in High Urbanization Speed Countries ........... ............. Media Participation and Political Participation in Moderate Urbaniza— tion Speed Countries ........... ........ .. Media Participation and Political Participation in Low Urbanization Speed Countries .............. .............. Simple—Correlation Coefficient, Across 33 Non-Communist Countries . ...... ... Multiple-Correlation Coefficient, Across 33 Non-Communist Countries .... ..... Partial-Correlation Coefficient, Across 33 Non-Communist Countries . ........ Simple-Correlation Coefficient, by Urbanization Speed .......... ........ Multiple-Correlation Coefficient, by Urbanization Speed .. .............. . Partial—Correlation Coefficient, by Urbanization Speed .......... ........ .. - iii - page 15 18 19 20 22 23 24 25 27 28 CHAPTER _I_ THE PROBLEM AND THE HYPOTHESES Introduction Daniel Lerner and others have indicated that deveIOping aware- ness of the responsibilities of citizenship is deeply imbedded in the whole matrix of increasing literacy, greater use of the public media and movement toward industrialization and urbanization. Following this line, research studies concerning the role of the media in deveIOping citizenship awareness often raise such ques- tions as : ( a ) Does mass—media use facilitate an awareness of responsibili- ties of citizenship ? For instance, will increase use of the media be related to increased voting in a country with low previous political participation ? ( b ) What are the correlations between the use of public media and level of knowledge of public affairs; i.e., local government, national politics, and international issues ? Studies about mass communication and voting behavior have been very fruitful, but sad to say, none has answered the questions men- tioned above. With some regret, most of the studies in this field have been conducted in the Western world, especially in the United States. The current generalization is that the mass media in the Western world are relatively ineffective in producing massive changes in ' ... Opinions but at the same time may have subtle effects in molding political behavior. Paul Lazarsfeld and Robert Merton applied to this dysfunctional aSpect of mass communications the colorful label of narcotization.( 1 ) By narcotization they mean increasing dosages of mass communications may be inadvertently transforming the energies of men from active participation into passive knowledge. Man spends so much time absorbing news that he takes little direct action; he may believe that to be an informed citizen is equivalent to being an active citizen. This generalization may hold for Western societies, but may not apply to societies recently introduced to mass media. In Western world, the presence of the media may be taken for granted such that some of the potential effect is attenuated sheerly by the pervasiveness of the media. Both the cultural stage of Western publics in their experience of mass media and the abundant supply of mass communication with possible satiation effects may not be found in most of the deveIOping countries. The evidence from technical assistance programs, conducted for some of the Latin American countries, indicates that political floud- ering and upheaval is often one of the consequences of technological develOpment. ( 2 ) Technological and economic deveIOpment is not equal- ly distributed in all the countries of the world. Some countries are at the beginning stage, and are, as a result, importing the assistance from highly deveIOped countries. Political awareness is a by-product of this importation. Since media development directly hinges on tech- nological and economic develOpment with the exception of radio, it is generally agreed that media Open to the large masses of man- kind the infinite vicarious universe. ( 3 ) Thus, it seems that political awareness in the less develOped countries comes into ex- istence by means of technological importation should be natural and beyond dispute. Considering communication and political articulation, Lucian Pye has indicated that a direct relationship exists in all socie- ties between the structure and organization of communication and the character, tone, and even to a degree, content of political ex- pression. The politician's role both as articulator of the collective identity and as champion of specific interests is invariably condi- tioned and limited by the media of communications available to him. No leader can rise above the restrictions of the specific communi- cations networks to which he has access, and at the same time none can escape the consequences of being surrounded by a communications system. In a broader sense, the political process as a whole is in- fluenced by access to the means of communications. ( 4 ) Communication researchers have examined the numerous ways in which the communication process can facilitate citizenship educa- tion and influence the process of political socialization in transi— tional societies. Their findings tended to suggest that in societies not saturated by the mass media these forms of communication may have a greater influence on people's conscious, cognitive learning about politics than they appear to have in the advanced countries. 1 5 ) ‘ , n. I) l1»! . l- The Hurdles of Research Any attempt to study the effects of media use on some type of human behavior will inevitably confront the problem of circularity-- a dilemma in which one will have difficulty in telling whether the use of media leads to a certain kind of human behavior or the other way around. For instance, on the one hand, some peeple, based on the assumption of causality, claim that violence in the media produces delinquency; thus, the more exposure to the violent content of the media the higher the probability of becoming a delinquent. On the other hand, some people, based on the assumption of catharsis, claim that it is the violent disposition that leads to the exposure to vio- lent content; thus, the higher the violent diSposition a person has, the more the frequency for that person to use violent content of the media as ways of relieving his aggressiveness. The same kind of controversy may exist in studying the relation- ship between political behavior and media consumption of peeple. There has been lack of empirical evidence to support whether media consumption initiates political interest or political interest eli— cits participation in politics via media consumption. The only sugges- tion in this aSpect is that in the Western world, especially in the United States, where participation in politics by means of media often results in inactive political behavior--- the notion of narcotization. This narcotizing effect of mass media is not very serious because we are sure that media deveIOpment in major proportion of the world is not as high as it is found in this country. Nevertheless, the controversy about the deveIOpment process is still a sensitive one. William Stephenson has indicated that urbanization, industriali- zation, higher standards of education, greater political awareness, and increasing use of mass media are all interwoven. 1 6 ) To il- lustrate this interdependent relationship, Daniel Lerner also hinted at the possible circularity by asking such questions as: What is a necessarycondition of modernization ? ..... I and ) What is nece- ssary to start modernization ? ( 7 ) Nevertheless, Daniel Lerner has also indicated that the Western model of modernization exhibits cer- tain components and sequences whose relevance is global. Everywhere increasing urbanization has tended to raise literacy; rising literacy has tended to increase media participation; increasing media exposure has gone with wider economic participation and political participa- tion.( 8 ) LMeanwhile, Daniel Lerner has also suggested that cities produce the machine tools of modernization. Accordingly, increases of urbani- zation tend in every society to multiply national increases in literacy and media participation. By drawing people from their rural communities, cities create the demand for impersonal communication. By promoting literacy and media, cities supply this demand. ( 9 ) Thus, following this notion, we can easily see that Daniel Lerner has subsumed industrialization under the index of urbanization, and his viewpoint on the process of modernization can be best conceptua- lized as the following paradigm. Other Types of Urbanization _g Media Partlgipation ) Participation ( Industrialization ) ( Literacy Growth ) Problem and Rationalg Such communication researchers as Daniel Lerner and William Stephenson have also suggested that mass media function most effec- tively only in rapidly-modernizing countries. Therefore, along with the two questions raised in the previous section, the problem this thesis intends to attack can be stated in two parts: 1 1 ) Do mass media facilitate an awareness of responsibilities of citizenship ? ( 2 ) Do mass media function most effectively only in the rapid- ly-modernizing countries ? A careful look at the problem itself would suggest that only when we find such a relationship as the higher the media use the more the economic and/or political participation can we say that mass media facilitate an awareness of responsibilities of citizenship. To put it another way, if the use of public media does not contribute anything at all to other types of participation, then no matter what stage a society is at in terms of the develOpment process, the rela- tionship between media use and other participations will tend to re- main unchanged. Also, the problem itself would suggest that only when we find that the function of the media in countries with faster rate of mo~ dernization is significantly different from its function in other countries with slower rate of modernization can we have the right to say that mass media function most effectively only in rapidly— modernizing countries. Miller Taylor and Author Jones have pointed out that societal configurations in Western societies are not the result of surrepti- tious planning on the part of those in power, that mass societies are neither imposed by force nor affirmed by self-corruption. Quite the contrary, mass urbanized society is the ordering of large numbers of peOple, most of them are in secondary communication with one ano- ther, by requiring a common loyalty to government, a support of com— mon ideas concerning the production and distribution of goods and services, and a belief in the common ideas of the meaning of life.( 10 Daniel Lerner also sees urbanization as the transfer of pepula- tion from scattered hinterlands to urban centers that stimulates the needs and provides the conditions needed for " take-off " toward wide— spread participation. So, it is quite apparant that urbanization is the first step toward modernization. Definitions of what is urban differ greatly. For instance, in some countries the definition of urban may include places having a population of a few hundred, while for some others they may require a place to have several thousand people before it is designed as ur- ban. As a result, an arbitrary standard of a population of 20,000, the figure suggested by G. Breeses and accepted by most of the re- searchers when conducting cross-national studies, is used for desig- nating a place as urban in this thesis. Following the arbitrary standard, countries all over the world are found to differ from one another in two aspects. Some countries such as the United States, the United Kingdom, Israel, and New Zealand show higher percentages in terms of urbanization of the population in cities over 20,000 than other countries do in cities of comparable size. Some countries such as Venezuela, West Germany,and Japan show higher rates of increase of urbanization of the population in cities over 20,000. It is the second notion, that of rate of increase, that draws the attention and will be discussed in this thesis. Urbanization movement, however, is not a form of human behavior which will grow continuously at the same speed. Geographical limita- tions as well as somessccto-economic conditions, such as transporta— tion facilities, may quite often reduce this movement after a nation has reached some sort of level. This is to say that as soon as a na— tion has reached a certain point of urbanization the speed for her pepulation in moving into the cities over 20,000 inhabitants will tend to slow down. For instance, statiStics indicate that the annual increase of population in cities over 20,000 for the countries such as the United States, the United kingdom, France, and Canada is much slower than that for the ones like Venezuela, New Zealand, Japan, and Israel. Nevertheless, it is also very important to pOint out that even if the percentage of annual increase of population in cities over 20,000 for two or more countries is exactly the same, such an equal increase does not necessarily infer that these countries share the same urbanization speed. The above is due to the fact that countries are found different in terms of their urbanization level; for the countries with a high urbanization level, a further increase of their population in cities of over 20,000 might be more difficult, if not impossible, than that in the countries with a low urbanization level. Therefore, the concept of urbanization speed of a country should take her existing urbanization level into consideration. The higher the existing urbanization level a nation indicates, the lower the possi- bility for that nation to have further urbanization. Whenever, two countries have the same percentage of annual increase of pOpulation in cities over 20,000 the one with higher urbanization level would have higher urbanization speed than the other, for higher urbaniza- tion level would simultaneously mean lower possibility of further urbanization for that nation. Thus, urbanization speed in this study is Operationally defined as a ratio between the annual increase of p0pulation in cities over 20,000 and the probability of further ur- banization in that particular country. Namely, Percentage of annual increase of Urbanization Speed = pOpulation in cities over 20,000 Probability of further urbanization where probability of further urbanization for a particular country wasccomputed as the result of total probability for urbaniza- tion minus existing urbanization level -—- 100 % — present urbanization level. Problem and Assumption With the above discussion concerning the justification of employ- ing urbanization Speed as an index of deveIOpment process, and toge- ther with the understanding of the fact that there is some relation- ship between the use of public media and political awareness, the consideration of the problem in this thesis becomes more specific. Generally speaking, this thesis was intended to find the re- lation between media participation and political participation in the non—communist countries under the assumptions that media par- ticipation represents the use of public media and that political participation reflects political awareness of the peOple in these countries. Before we go into the establishment of the hypothesis for test- ing two notions should be clarified first. From the beginning of the thesis the term public media has been mentioned again and again. By definition, public media use refers to the media used by the mass in societies. Therefore, the use of public media implies the participa- tion of the population in such media as newspapers, radios, televi- sions, etc. When nations are treated as units for comparison media participation means the consumption of the pOpulation in terms of newspapers, radios, and televisions. However, for a major propor- tion of the world television is still a new device and is not being so popularly used as either newspapers or radios. Consequently, media participation in this thesis particularly excludes the consideration of television as one of the public media. Operationally defining, newspaper participation is the percent- age of daily newspaper circulation per 100 population, and respective- ly, radio participation means radios per 100 pOpulation. Media parti- cipation as a whole is the combination of both newspaper and radio participation. Another aspect in which we are interested is the level of know- ledge of public affairs. Several methods are available to detect hu- man behavior in this dimension. For instance, we can obtain the re~ sult by asking the peOple what are the most important issues, locally or nationally; we can obtain the data by voting participation records. For the sake of national comparisons the latter method would seem more convenient and more objective. Limited to availability of data for cross-national study this paper would only use voting participation in national elections as an index of the level of knowledge of public affairs. It is assumed that the more the pOpulation knows about public affairs the more they will approach : to politics; hence, participation in national elections should reflect the level of knowledge of public affairs to a great extent. As a result, this assumption excludes the consideration of communist countries, for the voting behavior in those countries would be nothing but voluntary ones and empirical evidence suggests that voting participation in those countries are often times higher than they are in non-communist countries. To sum up, political participation in national elections was Ope- rationally defined as the percentage of actual votes over the percent- age of a papulation beyond the voting age. To put it another way, if the percentage of a population beyond the voting age is 65% , and the political participation of that nation as a whole is 40%, then the ' new ' political participation which would be used here is about 62%; e.g., political participation in national elections for that particular nation should be 40% out of 65% rather than 40% out of 100% because there is only 65% of the pepulation qualified to attend the election. Hypotheses If there is a direct relationship in all societies between the structure and organization of communications and the character, tone, and even to a degree, content of political expression, then we expect the correlation coefficients between media participation and political participation to differ from a chance fluctuation. Now, since the media participation here is composed of both newspaper participation and radio participation this hypothesis should be re-stated as follows. Hypothesis 1 : The multiple-correlation coefficient between the combination of newspaper participation and radio participation, and political participation in na- tional elections should be significantlytdifferent from a chance fluctuation for all non—communist countries. It has been generally recognized that in societies not saturated by the mass media these forms of communication may have a greater in- fluence on peOple's conscious, cognitive learning about politics than they appear to have in advanced countries. Specifically, Daniel Lerner and William Stephenson have indicated that mass media function most effectively only in rapidly modernizing nations. Also, the rise of cities has been accepted as an important index of modernization. Thus, a rapid modernization for a nation would mean that in that very nation the Speed of her pOpulation to move into cities over 20,000 is rapid, too. Based on the above discussion, if urbanization Speed is taken into consideration we would then expect that mass media function most effectively only in rapidly—urbanizing nations. In terms of correla- tion, the correlation coefficient between media participation and po- litical participation should be highest in the countries where urbani- zation Speed is most rapid. In a word, the second hypothesis should be stated as follows. Hypothesis g : The multiple—correlation coefficient between the combination of newSpaper and radio participation, and political participation should be highest for the countries which have the most rapid urbaniza— tion speed. This relationship will become weaker as the urbanization Speed decreases. CHAPTER ;; THE METHOD As the previous chapter has indicated the pOpulation for this study should ideally include all the countries in the world, for we are interested in the relationships between two types of human be- havior—-— media participation and political participation---in all societies. However, sociologists as well as political scientists have come to the conclusion that in sOcieties with strong central governments an unregulated press tended to grow only very slowly. ( 11 In general, the greater the extent to which a form of government is actually dependent upon favorable public Opinion, the more likely it is to support a free press. That is, in societies where the com- mon man plays significant roles in the determination of his own poli- tical destiny, the participation in media and formation of public opinions are highly related and strongly maintained. ( 12 ) Statistics in terms of voting participation in national elections also indicate that the percentage for the tap five nations are Russia (99.6 per 100 population ), Bulgaria ( 99.2 ), Czechoslovakia ( 98.1 ), Romania ( 97.9 ) and East Germany ( 97.3 ). Communist countries, as they are commonly recognized as societies with other forms of highly centralized power, do not require active public discussion of issues about which every citizen must reach an informed decision. Even if they do, the political expression by means of voting in national elec- tions will be nothing but a voluntary behavior. Consequently, the - SOOpe of pOpulation should be limited to non-communist countries only. SamplingAProcedures Since an individual nation is our unit of measurement and due to the fact that media participation and political participation are the two phenomena in which we are interested, the collection of data should be able to provide us with adequate information for each of the non-communist countries in the aspects of newspaper consumption, radio consumption, and voting attendance in national elections. The satisfaction for the requirements cited above, unfortunately, reduced the number of the countries to 33. In the order of urbaniza- tion level, i.e., the percentage of total population in places of 20,000 or more inhabitants, these non—communist countries are listed as follows. * Table I Media Participatign and Pglitical Participation, by Nation ( up to 1960 ) 15 Newspapers Radios Voting Attendance Nation Per 100 Per 100 Per 100 United Kingdom 50.6 28.9 78.0 Israel 21.0 19.4 88.0 New Zealand 38.9 24.2 86.4 West Germany 30.7 31.9 86.9 United States 32.6 94.8 64.4 Denmark 34.5 36.5 84.0. Venezuela 9.6 18.6 83.8 Chile 13.4 13.1 37.4 Table I ( Continued ) 16 Newspapers Radios Voting Attendance Natigg Per 100 Per 100 Per 100 Japan 41.6 10.7 71.2 Sweden 47.7 37.8 83.1 Austria 20.8 28.8 90.4 Canada 23.1 45.1 74.2 Greece 12.5 8.9 73.3 Ireland 22.5 17.6 71.3 Panama 10.4 15.9 56.2 South Africa 6.1 6.6 10.4 Finland 35.8 28.9 72.8 Luxembourg 44.5 31.9 71.1 Italy 10.1 17.0 92.9 Belgium 28.5 29.8 87.6 France 25.2 28.2 89.4 Mexico 8.3 9.7 34.6 Switzerland 37.4 27.2 28.0 Brazil 5.4 6.4 34.4 Nicaragua 6.6 6.7 92.7 Malaysia 6.7 3.7 54.8 Turkey 4.5 5.3 72.5 British Guiana 7.9 7.2 52.1 Portugal 8.1 9.8 18.5 India 1.1 .5 52.6 Honduras 2.5 6.6 36.5 Table I ( Continued ) Newspapers Radios Voting Attendance Nation Per 100 Per 100 Per 100 Dominican Republic 2.7 3.3 63.6 South West Africa 1.4 3.7 11.0 * Table I is mainly after Tables 9, 24, 31, and 35 in World Handbook Of Political and Social Indicators. Other sources are : ( a ) U.N. Compendium of Social Statistics, 1963 1 b ) U.N. Statistical Yearbook, 1962 ( c ) Review of Elections, 1962 AS far as the purpose of this thesis is concerned this kind of categorization will only answer part of the problem indicated pre- viously. In other words, a correlational analysis on the basis of this type of categorization will provide empirical evidence for the part of the problem which asks the relationships between use of pub- lic media and level of knowledge of public affairs across 33 non- communist countries. A statistical significance for such a relation- ship would then indirectly indicate that media use does facilitate an awareness of the reSponsibilities of citizenship, assuming that voting turn-outs in national elections reflect the awareness of the responsibilities to some extent. However, the second part of the problem --- media function most effectively only in the rapidly—modernizing nations --— remains unsolved. This inadequacy leads to the consideration and data col— lection for the concept of urbanization speed. As explained in the previous chapter urbanization speed, by definition, is synonymous with modernization. Thus, rapidly-modernizing nations refer to the nations where peOple are rapidly urbanizing. Also, urbanization speed should take the present urbanization level of a nation into consideration. To be brief, countries indicating high urbanization speed do not conside with countries indicating high urbanization level. Computational procedure and the rationale behind it will not be repeated here. The following three tables will indicate media participation and political participation according to the order of urbanization speed. With the purpose to test the second hypothesis, 33 countries have'been arbitrarily categorized into three groups; namely, high Speed, moderate Speed, and low speed. Table I ( a ) Media Participation and Political Participation ip High Urbanization Speed Countries 18 Urbanization Voting Newspapers Radios Nation Speed 1 ratio ) Attendance Per 100 Per 100 Venezuela 2.44 83.8 9.6 18.6 New Zealand 1.62 86.4 38.3 24.2 Panama 1.59 56.2 10.4 15.9 West Germany 1.49 86.9 30.7 31.9 Japan 1.18 71.2 41.6 10.6 Ireland 1.16 71.6 22.5 17.6 Table _I ( a ) --— Continued Urbanization Voting Newspapers Radios Nation Speed I ratio 1 Attendance Per 100 Per 100 Greece .99 73.3 12.5 8.9 Israel .99 88.0 21.0 19.4 Nicaragua .92 92.7 6.6 6.6 Chile .88 37.4 13.4 13.0 Mexico .85 34.6 8.3 9.7 Finland .85 72.8 35.8 28.9 Table _l ( b ) Media Participation and Pplitica; Participation ip Moderate Urbanization Speed Countries Urbanization Voting Newspapers Radios Nation Speed I ratio ) Attendance Per 100 Per 100 Brazil .65 34.4 5.4 6.4 Sweden .62 83.1 47.7 37.8 Denmark .62 84.0 34.5 36.5 South Africa .58 10.4 6.1 6.6 Honduras .44 36.5 2.5 6.6 Malysia .44 54.8 6.7 3.6 United States .38 64.4 32.6 94.8 Dominican Republic .37 63.6 2.7 3.3 Canada .36 74.2 23.1 45.1 France .35 89.4 25.2 28.2 Table ll.( b ) --- Continued Urbanization Voting NewSpapers Radios Nation Speed I Ratio ) Attendance Per 100 Per 100 Italy .33 92.9 10.1 17.0 Turkey .33 72.5 4.5 5.3 United Kingdom .30 78.0 50.6 28.9 Table II ( c ) Media Participation and Political Participation in Low Urbanization Speed Countries Urbanization Voting Newspapers Radios Nation Speed ( ratio ) Attendance Per 100 Per 100 India .20 52.6 1.1 .5 Switzerland .15 28.0 37.4 27.5 Luxembourg .11 71.1 44.5 31.9 South West Africa .10 11.1 1.4 3.7 Portugal .06 18.5 8.1 9.8 Bristish Guiana .003 52.1 7.9 7.2 Austria -.13 90.4 20.8 28.8 Data Analysis and Statistics The measurement instruments employed in this thesis are: ( 1 ) Pearson product—moment correlation coefficient; ( 2 ) Multiple-cor- relation coefficient, and ( 3 ) Partial-correlation coefficient. The simple-correlation coefficient technique was employed to compute the relationship between newspapers circulation per 100 population and radios per 100 population, neWSpapers circulation per 100 pOpulation and voting attendance in national elections per 100 pOpulation, and between radios per 100 pOpulation and voting attendance per 100 pOpulation. The multiple-correlation coefficient method was employed to compute the combined predictability of media participation toward voting participation. In other words, this technique was employed. to obtain the variability of voting participation explainable by media participation. Finally, the partial—correlation coefficient technique was em- ployed because we wanted to know which one of the media was more closely related to voting participation if one of them did not have a chance to boost, either positively or negatively, the relationship. CHAPTER III FINDINGS Correlationa; Analyses and Tests for Significance The data were analyzed in two ways. Tables III, IV, and V give correlational analyses without breaking the non-communist countries into sub-groups. Tables VI, VII, and VIII give the cor- reSponding analyses after the non-communist countries were cate- gorized into high, moderate, and low urbanization Speed groups. Chi square, Student's t, and F tests were used to test the signi— ficance of obtained correlation coefficients. Correlates of Political Participation Table III Sipple Correlation Coefficient, Across 33 Countrigg Correlation Between Correlation Between Correlation Between NewSpaper and Radio Radio and Political NewSpaper and Poli- Participation Participation tical Participation .63 .29 .33 * t: 4.5 t: 1.71 t: 1.94 * t) 2.04 p< .001 df: 31 s. The use of t—test for significance of simple correlations in- dicated that only newspaper participation and radio participation were significantly related with each other. Thus, an increase in newspaper participation appeared to parallel an increase in radio participation in the countries studied. However, neither radio nor newSpaper participation was significantly related to political par- ticipation. Table 11 Multiple Correlation Coefficient: Correlation between.Media Partipi— pation and Political Partici ation, across 33 Countries 33 Non-Communist Countries Media Partici- pation & .35 Political Par- ticipation F: 2.09 df1= 2 df2= 30 p) .05 The interpretation of the multiple-correlation coefficient and F value computed as a test of significance led us to conclude that there was no significant relationship between media and political participation across 33 non-communist countries. In other words, the media participation Of the people in these countries would explain little of the turn-out in their national elections. It should be noted that the correlation obtained in the sample was moderate in size, and in the direction predicted. Thus, the lack _ 23 _ of a statistically significant relationship between media partici- pation and political participation may have been due to an inade- quate sample size rather than an inherent absence of such a corre- lation. It would require another study, with a larger sample of countries to verify this possibility, of course. anew Partial—Correlation Coefficient, across 33pCountries Partial-Correlation Coefficient Partial—Correlation Coefficient between Newspaper and Political between Radio and Political Participation with Radio Parti- Participation with Newspaper cipation Held Constant Participation Held Constant .20 .11 t=1.18 df=31 p) .10 t: .64 df= 31 p>.10 As was mentioned at the end of the previous chapter, the pur- pose of the partial—correlation technique was to examine whether the inclusion of one type of medium participation would make any contribution to the relationship between another type of medium par- ticipation and political participation. A comparison between Table III and Table V Shows that when radio participation is held constant the relationship between newspaper participation and political partici- pation across the 33 non-communist countries decreases from .33 to .20. Further, when newspaper participation is held constant the re- lationship between radio participation and political participation for these countries decreases from .29 to .11. The t-test for significance here indicated that neither Of the two partial correlation coefficients were statistically sig— nificant. That is, for the 33 non—communist countries, the rela- tionship between one type of medium participation and political participation, without the possible contribution from another type of medium participation, was attributed to a chance fluctuation. Correlation between : Newspaper & Radio Newspaper & Voting Radio_ & Voting Urbanization Speed and Political Participation Table V; Simple Correlation Coefficient, py’Urbanization Speed Moderate Urbaniza- Low Urbanization High Urbanization tion Speed Countries Speed Countries Speed Countries N: 12 N213 N: s .57 .64 .92 x2: 2.98 df: 2 p) .05 .62 054 043 x2: .2; df: 2 101.0 .35 .34 ‘ .62 Table VI gives us the simple correlation coefficients between newspaper and radio participation, between newspaper and political participation , and between radio and political participation for three groups of different urbanization Speeds. Here, the relation- ship between newspaper and radio participation appears to increase as urbanization speed decreases; that is, the higher the urbaniza- tion speed, the less the relationship between these two participa- 2 test did not verify this interpreta- tions. Unfortunately, the X tion. The obtained X2 value suggested that these three correlations were homogeneous and could be considered as estimates of the same pOpulation value. Similar findings and interpretation may apply to the three simple correlations between radio and political participation. How- ever, the relationships between radio and political participation failed to show a progressive increase or decrease as a function of urbanization speed. Such a progressive change was noticeable in the relationships between newSpaper and radio participation and between newSpaper and political participation. Briefly Speaking, the nonsignificant X2's in Table VI indicate that different urbanization Speeds did not significantlycbhapge.the simple correlation coefficients for the three variables. Thus, the contribution resulting from the break down of the non—communist countries, on the basis of urbanization speed, was not useful as far as the simple-correlation coefficients were concerned. Table VII Multiple Correlation Coefficient, p1 Urbanization Speed High Urbanization Moderate Urbanization Low Urbanizatio: Speed Countries Speed Countries Speed Countries: N = 12 N = 13 N = 8 Multiple Correlation between Media .62 .54 .72 and Political Participation 2 X = .27 df= 2 p) .05 Table VII indicates that, individually speaking, for both high and low urbanization speed countries the combined predictability of newspaper and radio participation on political participation is sta- tistically significant from a chance fluctuation ( p<<.05 ). How- ever, when compared to one another, the three multiple-correlation coefficients are Similar regardless of differences in urbanization speed. Therefore, the nonsignificant X2 indicates that the function of the media in countries with high urbanization speed is p93 sig- nificantly different from its function in other countries with slow- er speeds of urbanization. Namely, the second hypothesis that media function most effectively pply in rapidly—urbanizing nations was not supported by the study. Finally, the question of whether the inclusion of newspaper participation would change the relationship between radio and poli- itical participation, and whether the inclusion of radio participation would change the relationship between newspaper and political par- ticipation if urbanization Speed of the countries was taken into consideration. In other words, would one type of medium partici- pation make any contribution to the relationship between political participation and another type of medium participation if we held urbanization speed constant ? Since newspaper participation requires certain degree of literacy while radio participation does not, this question is of importance. Table VIII Partial Correlation Coefficient, plerbanization Speed High Urbanization Moderate Urbanization Low Urbanization Speed Countries Speed Countries Speed Countries Newspaper N = 12 N = 13 N = 8 and Political Participation _ with Radio '55 .45 .46 Held Constant x2: 2.41 df: 1 p 4 .05 Radio and Political Par- -.01 -.01 .63 ticipation with Newspaper Held Constant x2: 1.93 df=1 'p) .05 In Table VIII, the partial-correlation coefficients for three groups of countries show that when radio participation was held con- stant the relationship between newSpaper and political participation was not homogeneous. Thus, the three partial correlations could not be considered as estimates of the same pOpulation value. In other words, classification of the countries on the basis of urbaniza- tion speed resulted in a significant finding indicating that radio participation contributed unequally to the relationship between news- paper and political participation when such a relationship was com- pared on levels of urbanization speed. Further, Table VIII shows that in countries with a low rate of urbanization, newspaper and political participation 1 with radio use held constant ) are far more negatively related than in countries with a faster Speed of urbanization. Comparing this correlation ( -.46 ) against the ra- ther large positive correlation between radio consumption and politi- cal participation ( with newSpaper use held constant ) of +.63 in the same low speed urbanization countries, one sees a far greater discrepancy than is present in similar comparison in countries with faster urbanization. This point will be discussed in detail later. Finally, Table VIII Shows that in countries with high and/or moderate urbanization speed, newspaper and political participation ( with radio use held constant ) are ppp too much different from radio and political participation ( with newspaper use held con- stant ), though radio are found unrelated to political participa— tion. This points out that literacy does play an important role in the relationship between media and political participation, for li- teracy tends to be higher in high and moderate urbanization speed countries than in countries with a low rate of urbanization. CHAPTER ;1 CONCLUSIONS The effects of media use have been widely discussed and the conclusion reached are controversial. When the researchers appro- ach the problem within the societies where public media have already been deveIOped and the peOple there have been habitually supplied with abundant media, the findings tend to underestimate the effects of these informative devices. However, when the researchers see the problem within the societies not saturated with mass media, the find- ings tend to overestimate their effects. The problem becomes more difficult to deal with when the pur- poses of a study are to find some generalization which will then be applicable to further studies in other places of the world. The dif- ficulty exists because of the fact that, for any society in the world, the growth Of media is more or less accompanied by the growth of other human behaviors. For instance, the question that whether political interest leads to an increase of media use or media use results in increasing political interest is the question which will never be satisfactorily answered. However, the difficulty does not necessarily mean that the problem concerning the possible influence of media use is unapproachable. The point should be that p23 we look at the media when we are looking for such an influence. Media are part of the products of the modern world, and media use, which occupies a fairly short time span in man's history, is one Of human behaviors. Taking a glance at the countries included in this study, one sees that the development of one type of medium participation is closely related to the deveIOpment of another type of medium participation. That is, for 33 non-communist countries, the higher the participation in one medium, the more the partici- pation in another medium. However, this evidence did not satisfy and answer the other questions in our mind. We wanted to see whether a single medium participation could tell us anything about voting attendance in national elections. We wanted to see whether the combined information about newspaper and radio participation could improve this predictability to some extent. Also, we wanted to see whether the control of one medium participation would make any difference for the relationship bet- ween political participation and another medium participation. Two approaches were employed to see the problem. The first approach dealt with the problem by viewing the non- communist countries without breaking them down into sub-groups. The findings thus obtained showed that neither radio nor newspaper par- ticipation was found significantly related to political participa- tion ( the criterion variable ). Further, the combined predictability resulted from newspaper and radio participation as.a prediCtOrctoward political participation was found not significantly different from a chance fluctuation,either. Besides, when radio participation was eliminated from the relationship between newspaper and political participation , and when newspaper participation was eliminated from radio and political participation, the partial—correlation coeffi- cients were found not statistically significant. The above findings led us to the following interpretations. First, the sheer information about one type of medium parti- cipation failed to explain political participation. In other words, either newspaper circulation per 100 population or radios per 100 pOpulation ( as two predictive independent variables ) was only re- lated to voting participation in national elections ( as the crite- rion variable ) by chance. Secondly, the combined information of radio consumption and newspaper circulation also failed to explain the political partici- pation in these 33 non-communist countries.In a word, the first hypothesis was not supported by the study. However, as mentioned before, the failure to support this hypothesis might be attributed to an inadequacy of the sample Size. Should we have a larger sample, we might have had the finding which is in accord with the rationale behind the establishment of the hypothesis. Thirdly, the nonsignificant partial-correlation coefficients for the relationships between newspaper and political participa- tion without the contamination of radio participation and between radio and political participation without the contamination of news- paper participation showed that neither one of the media aippp was a good index to explain political participation. To sum it up, the above discussion seems to suggest the follow- ing two phenomena: the growth of media participation is not related to the growth of political participation in these non-communist countries, ppp the growth of one medium participation is accom- panied by the growth of another medium participation. The second approach dealt with the problem by viewing the re-- lationship between media and political participation in the non- communist countries after they were broken down into three groups. Urbanization Speed was used as a basis to classify the countries, for the evidence suggested that urbanization level 1 percentage of pOpulation in cities of 20,000 or more inhabitants ) and urbaniza- tion speed 1 rate of a country to be further urbanized ) was not related 1 r=.20 ; p)..05 ). Besides, the test for the second hypo— thesis required the classification of the non-communist countries on the basis of urbanization speed. The results of simple-correlation coefficients indicated that as the urbanization speed decreased the relationship between news- paper and radio participation increased though the test for homo- geneity of the simple correlations led us to conclude that these correlation coefficients were not significantly different. The re— sults of simple correlation coefficients between radio and politi— cal participation, and between newspaper and political participa- tion, however, did not provide such an implication. Thus, one may say that urbanization speed seems to provide a noticeable trend for the relationship between newspaper and radio participation, but it fails to do so for the relationship between media participation and political participation. When the relationship between political and media partici- pation was compared for three groups, the findings indicated that the function of the media in countries with high urbanization speed was not significantly different from its function in other countries with Slower Speeds of urbanization. Thus, as far as the data in this study were concerned, the second hypothesis that media function most effectively only in rapdily-urbanizing nations was not supported. The involvement of literacy in newspaper participation was mentioned. This question is important especially when countries were considered on the basis of urbanization speed. Following Daniel Lerner's notion about modernization process, the growth of cities promotes the demand for literacy and media. Should this be the fact, the faster the urbanization speed a nation has, the more the demand for literacy and media will be. Comparing the two media mentioned in this thesis, an increase of newspaper participation is a direct function of literacy, but an increase of radio participation is not. Thus, if low urbanizatiOn speed implies low increase in literacy growth the elimination of radio participa— tion from the relationship between newspaper and political partici- pation should affect countries with low urbanization Speed the most. It is because the relationship between newspaper and political par- ticipation in these countries may likely be due to the contamination of radio participation, which does not require literacy. The findings in Table VIII indicated that if radio participa- tion was held constant, the relationship between newSpaper and poli- tical participation varied significantly different from a chance fluctuation for three groups of countries. To put it another way, the partial correlation coefficients computed for the three groups were no longer the estimates of the same pOpulation value. In a word, urbanization Speed is a good indicator for studying the rela- tionship between newspaper and political participation, with radio participation held constant. Although the findings failed to support both of the two hypo- theses about the relationship between media use and such human be— havior as political participation in national elections, the des— criptive data provide evidence that is in accord with the rationale behind the hypotheses. The biggest problem this thesis seems to have confronted was data-collection. If all the data required for this study were avai- lable for all the non-communist countries, the sample size for this thesis should have exceeded 50 instead of 33 countries. The find- ings, as a result of larger sample, may likely have supported the / hypotheses raised here. 10. 11. 12. BIBLIOGRAPHY Lazarsfeld, Paul and Merton, Robert:"Mass Communication, POpu- lar Taste, and Organized Social Action,"printed in The Communi- cation of Ideas, the Institute for Religious and Social Studies, New York, 1948, pp. 95-118 Department of Communication, Michigan State University: Communi- cation and Change in Latin America -- A Research and Graduate Education Program Proposal, 1963, pp. 23-26 Lerner, Daniel: The Passing of Tragitional Society:Modernizipg the Middle East, New York, Free Press of Glencoe, 1958, pp.43—75 Pye, Lucian: Communication and PoliticalpDevelppment, Princeton, New Jersey, Princeton University Press, 1963, p.58 : Communication and Political Development, 1963, p.149 Stephenson, William: The Play Theory of Mass Communication, the University of Chicago Press, Chicago and London, 1967, p. 96 Lerner, Daniel: The Passing of Traditional Sociepy: Modernizing the Middle East, 1958,pp. 43-75 : The Passing of Traditional Society: Modernizing the Midgle East, 1958, pp. 43-75 : The Passing of Traditional Society: Modernizing the Middle East, 1958, pp. 43-75 Taylor, Miller and Jones, Author R. Jr.,Rural Life and.Urbanized Society, New York, Oxford University Press, 1964, pp.38-41 De Fleur, Melvin L.: Theories of Mass Communication, David Mckay Company, Inc., New York, 1966, p. 10 : Theories of Mass Communication, 1966, p. 11 Data Sources : 1. 2. 3. 4. Russett, Bruce M. et al : World Handbook Of Political and Social Indicators, New Haven and London, Yale University Press, 1964 U.N.,Compendium of Social Statistics, 1963 U.N., Demographic Yearbook, 1960 & 1962 Review of Elections, Institute of Electoral Research, London, 1962 Sources of Statistical Tests for Significance: 1. 3. Student's t-test : McNemar, 0., Psychological Statistics, John Wiley and Sons, Inc., New York, London, Sidney, 1962 F-test : Edwards, Allen L., Experimental Design in Psychological Research, Holt, Rinehart and Winston, 1950 Chi-Square test : Edwards, Allen L., Experimental Design in Psy- chological Research, Holt, Rinehart and Winston, 1950 II III IIIIII 1293