Political learning and political culture: A comparative inquiry

Transcription

1 Political learning and political culture: A comparative inquiry Thomas Denk Department of Political Science Åbo Akademi University Finland tdenk@abo.fi Sarah Lehtinen Department of Political Science Åbo Akademi University Finland salehtin@abo.fi Paper to be presented at ECPR General Conference, University of Glasgow, 3-6 September 2014, Section: Citizenship: New Trends and Developments, Panel: Citizenship Education, Competence and Participation.

2 Introduction Traditional theories on political socialisation have assumed that the political orientations of individuals are developed by communication with other actors. Actors are expected to send messages through communication channels that affect the political orientations of the individuals (Easton & Dennis 1969; Greenstein 1965; Hyman 1959; Langton 1965). Harold Lasswell (1948) summarised this conventional perspective of political socialisation in his classic formulation: a) who b) learns what c) from whom d) under what circumstances e) with what effects? 1 Later studies have also provided empirical support to the assumptions about the importance of political communication for political culture. Political communication, particularly mass media, frames and organises information that affects how citizens understand politics. It influences what citizens regard as important and how they evaluate politics by activating some memories before others. Additionally, political communication may provide persuasive arguments that motivate citizens to change their opinions and engagement. Political communication can therefore result in citizen mobilisation as well as demobilisation (Inglehart & Norris 2009; Kinder 2003; Norris 2002; Nye & Zelikow 1997; Putnam 2000; Zaller 1992). This paper tests an alternative perspective on political communication and political culture. We analyse if the use of political communication is connected with political orientations. Instead of examining the effects of political communication on political orientations, we investigate whether individuals with different political orientations tend to use different channels for political information. However, our ambition is not to explain the use of communication channels for political information but to investigate if there are differences in the use of communication channels for political information between political subcultures. Previous studies have mostly analysed the differences in political communication between social groups (e.g., Inglehart & Norris 2009; Putnam 2000), while the main question for this paper is whether there are differences between political subcultures, which consist of individuals with similar political orientations, concerning the use of communication channels 1 This formulation is also a summary of the differences between the theories about political socialisation as they present diverse expectations concerning the five aspects (Greenstein 1968; Merelman 1986; Wasburn 1994). 1

3 If there are differences between political subcultures concerning the use of communication channels, it indicates that the processes of political learning and how they receive information are different between political subcultures. It could also indicate that the opportunities to reach individuals with political information are different between political subcultures. For example, we need to use one communication channel to reach individuals in one subculture, while communication with another subculture requires the use of other channels. This provides motives to examine the connection between political communication and political subcultures. This paper is organised in five sections. After this introduction, the second section presents a theoretical explanation of why individuals, based on their political orientations, may be expected to use different communication channels to learn about politics. The third section presents how the use of communication channels for political information and political culture is measured. The fourth section consists of empirical analyses that test the expected connection based on material from cross-national surveys in six states (World Value Survey, wave 6: ). Conclusions from these analyses are formulated and discussed in the last section, which also discusses the main contribution from the analyses, which is that there are differences between political subcultures concerning the use of communication channels for political information, but that the differences are moderate and vary between communication channels as well as between countries. Analytical framework The expectation that there are differences in the use of communication channels for political information between political subcultures is based on three theoretical reasons. First, the use of communication channels requires different amounts of resources. For example, the amount of time to select and read an article is supposed to be higher than to watch a news report. Individuals with a political interest are expected to prioritise political communication to a higher degree than individuals with less interest in politics (Downs 1957; Zaller 1992). We therefore expect differences in the use of communication channels for political information based on political interests. Second, political agents use different communication channels to communicate their messages. Citizens have different degrees of trust toward political agents and are expected to be 2

4 more open to using channels where the trusted agents tend to send their messages (Coleman 1990; Tilly 2005; Zaller 1992). We therefore expect differences in the use of communication channels for political information based on political trust. Third, the use of communication channels requires different degrees of engagement by the individuals. Watching news reports on TV requires less engagement than discussing politics with friends or seeking information on Internet. Individuals with political interest and political trust are more likely to be politically active than individuals without political interest or political trust (Dahlgren 2005; Norris 2002). As a consequence, individuals who are politically interested and have political trust are expected to be more willing to use communication channels that require that an active receiver. We therefore expect differences in the use of communication channels for political information based on political interests and political trust. In sum, to be used as a source of political information, the channels for political communication require different conditions and the occurrence of these conditions are expected to be connected to political orientations, which creates expectations about differences in the use of communication channels for political information between political subcultures. Based on the theoretical reasons for our expectations, two aspects of political orientations are regarded as the most relevant: political interest and political trust. These two aspects are, according to the seminal works on political culture, central aspects of political cultures (Almond & Verba 1965; Inglehart 1977; 1990). Furthermore, according to their definitions of political culture, political orientations are combined and aggregated into political subcultures, which together constitute the political cultures of states. Based on this conceptual framework, we combined political interest and political trust into the four categories of political subcultures. As presented in Table 1, these four categories are termed with inspiration from later studies on political cultures as civic citizens (Almond & Verba 1965; Inglehart 1977), critical citizens (Norris 1999), stealth citizens (Hibbing & Theiss-Morse 2002) and disenchanted citizens (Hay & Stoker 2009; Stoker 2006; 2010). 2 2 We only use the labels from previous studies. The concepts used in these studies have more complex structures than the structure with two dimensions presented in Table 1. 3

5 Table 1 Political subcultures based on political interests and political trust Political interest Political trust Low High High Stealth citizens Civic citizens Low Disenchanted citizens Critical citizens Materials To measure the use of communication channels for political information and political cultures, we use material from the World Value Survey (WVS wave 6: ). The uses of communication channels for political information are indicated by questions where the respondents state if they use sources to obtain political information daily, weekly, monthly, less than monthly or never. Based on this information, we calculate the probability that an individual uses the communication channel for political information on a random day (se Appendix for more information). If all individuals use the channel to obtain political information daily, then the probability is one (1), while a probability of zero (0) indicates that no individual ever uses the channel for political information. This probability is calculated for five communication channels: daily newspapers, television news, radio news, Internet and talks with friends or colleagues. 3 Based on the typology presented in Table 1, political orientations are measured with survey items on political interest and political trust. Individuals are classified as politically interested if they state that they are very or somewhat interested in politics, while they are classified as politically uninterested if they answer that they are not very or not at all interested in politics. Political trust is measured with an additive index that consists of three survey items about confidence in government, parliament and political parties (Cronbach s alpha = 0.81). Individuals are regarded as having a high level of 3 The survey also includes printed magazines, mobile phones and s as sources of political information. These sources are for various reasons excluded from the analyses. Printed magazines are periodically published (weekly or monthly), which makes it difficult to compare the use of printed magazines with sources that are not periodised in the same way. The use of mobile phones as a political source is quite diffuse as mobile phones are multifunctional; for example, you may use the Internet or discuss politics with friends by mobile phones. This creates an overlapping with other sources for political information. is also diffuse as a source of political information as it, not by itself, contains political information, but it can be used by different senders to communicate political information. For example, can be used for discussions with friends or to obtain information from journalists. To avoid overlapping between channels, we initially excluded mobile and as sources of political information. 4

6 political trust if they respond that they have a great deal or quite a lot of confidence in two or three institutions, while not very much or no confidence at all in two or three institutions, indicating a low level of political trust. This classification of political trust is then combined with the classification of political interest to provide the classification of political subcultures in Table 1. The sample includes 12,495 respondents in six countries: Australia, Germany, Japan, Netherlands, Sweden and United States. We selected these countries based on three reasons. Firstly, the selection of cases is a way to ensure a reasonable comparative equivalence in the analyses (van Deth 2013). More countries than the six selected are included in the round of World Value Survey (WVS). However, it is questionable whether the meaning of the survey items is the same in Azerbaijan or Uzbekistan as in Japan or the Netherlands as the cultural, economic and political contexts are diverse. Secondly, the theories about political communication and political culture concern democratic systems in modern societies. The selected cases are consolidated democracies with a high level of modernisation. 4 Thirdly, the selection of countries provides opportunities to empirically investigate the connection between political communication and political culture in different institutional contexts. The selected countries represent different forms of democratic systems. Australia, Germany and the United States are federations, while Japan, Netherlands and Sweden are unitary states. The United States and Australia are also regarded as majoritarian democracies, while Germany, Japan, the Netherlands and Sweden are viewed as consensus democracies (Lijphart 1999; Powell 2000). Additionally, the countries also have different media systems, where, for example, the relations between media and political institutions differ (Hallin & Mancini 2004). Table 2 presents the composition of political subcultures in the six selected countries. As the contingency coefficient indicates, there are differences in the composition between the countries. Australia, together with the United States, has the highest share of disenchanted citizens, but also the lowest share of civic citizens. The highest share of civic citizens and the lowest share of critical citizens are in Sweden. Japan combines the highest share of critical citizens with the lowest share of stealth citizens as does the 4 New Zealand is also included in the sixth wave of World Value Survey. However, the sample for New Zealand was considerably smaller than for the selected cases, which may indicate problems with nonresponse that affects the significance of the cross-national analyses. 5

7 Netherlands, while Sweden has the highest share of stealth citizens. The next section analyses whether these differences in political subcultures among the citizens and countries are connected with the use of communication channels for political information. Table 2 Political subcultures within selected countries Political subcultures Countries Disenchanted Critical Stealth Civic Total citizens citizens citizens citizens Australia 38.2% 39.0% 8.3% 14.5% 100.0% (n = 1428) Germany 26.5% 34.4% 10.5% 28.7% 100.0% (n = 1960) Japan 26.2% 52.8% 4.4% 16.6% 100.0% (n =2032) Netherlands 26.2% 43.7% 6.2% 23.9% 100.0% (n = 1750) Sweden 23.6% 19.8% 18.3% 38.3% 100.0% (n = 1152) USA 32.8% 47.5% 7.1% 12.5% 100.0% (n = 2142) Total 29.0% 41.2% 8.5% 21.3% 100.0% (n = 10464) Contingency coefficient = (sign. = 0.000) Empirical analysis Table 3 presents the probability that individuals within political subcultures use a channel of communication for political information on aa random day. According to the analyses presented in Table 3, there are three general patterns in the connection between the use of communication channels for political information and political subcultures. Firstly, there are significant, but modest, differences in the use of communication channels between political subcultures. As indicated by the etacoefficients, the largest differences are found in the use of newspapers and television, while the differences in the use of radio, Internet, and talks with friends or colleagues for political information are minor. Secondly, television is the channel used most frequently for political information in all political subcultures. In a similar way, news on the radio is the least frequently used channel for political information, with the exception of civic citizens and disenchanted citizens who use conversation and newspapers, respectively, as channels with the least frequency for political information. Thirdly, civic citizens and 6

8 critical citizens use communication channels more frequently than stealth citizens and disenchanted citizens. The least frequently used are the communication channels used by the disenchanted citizens. However, the use of talks with friends and colleagues is an exception to this pattern. Stealth citizens and civic citizens are using friends and colleagues as sources for political information more frequently than disenchanted citizens and critical citizens. According to the analyses, critical citizens use this channel least frequently among the four subcultures. 5 Table 3 Political subcultures and use of communication channels for political information Channels of Communication Political Newspapers TV Radio Internet Talks subculture Disenchanted (n=3204) (0.4601) (0.4355) (0.4699) (0.4728) (0.4547) Critical (n=4553) (0.4594) (0.3501) (0.4711) (0.4696) (0.4488) Stealth (n=944) (0.4685) (0.4233) (0.4705) (0.4723) (0.4450) Civic (n=2414) (0.4380) (0.3386) (0.4615) (0.4682) (0.4460) Total (n=11115) (0.4659) (0.3876) (0.4723) (0.4714) (0.4505) Eta (Significance) The table presents the mean and standard deviation (in parenthesis) of the probability that individuals within political subcultures use a channel of communication for political information on a random day. However, there are differences between countries in the use of communication channels for political information. Table 4 presents the probability that individuals within countries use a channel of communication for political information on a random day, which indicates how the communication channels are used as sources for political information in different countries. According to the eta-analyses, there are significant differences in the use of all channels between the countries. The differences are the 5 The five items about the uses of communication channels as sources for political information have been analysed with factor analysis (Varimax as a rotation method). According to the analysis, there are two dimensions in the use of political channels. One dimension is channels with one-way communication (newspapers, TV and radio), while the other dimension consists of interactive channels (Internet and conversations) that provide opportunities for two-way communications. However, the factor solution is weak in several aspects: the variance explained is quite low (57.2%), the sampling adequacy is weak (KMO = 0.55) and the degree of extraction for the use of radio is low (0.390). Additionally, reliability analyses indicate a low degree of internal consistency (Cronbach s alpha < 0.5) for the different indexes. Based on these results, the factor solution is not used in the analyses. 7

9 largest in the use of newspapers and news on radio as sources for political information. In Japan and Sweden, newspapers are frequently used for political information, while newspapers are more seldom used in Australia and United States as a source for political information. In Australia, news on the radio is used more frequently, as is also the situation in the Netherlands and Germany. News on radio is least used for political information in Japan and the United States. There are also differences concerning news on television, Internet and talks with friends or colleagues. News on television is the channel of communication that is most frequently used for political information in all countries, except Japan, the Netherlands and Germany where the use of this channel is prominently high. Internet and talks with friends or colleagues are used most frequently in Sweden for political information. Neither of these channels is used regularly in Japan as sources for political information. Talks with friends or colleagues are also used more seldom in the Netherlands and the United States, but citizens in the Netherlands use the Internet more frequently as a source of political information. Table 4 Cross-national comparison of the use of communication channels Channel of communication Country Newspapers TV Radio Internet Talks Australia (n=1456) (0.4409) (0.4083) (0.4422) (0.4695) (0.4429) Germany (n=2046) (0.4637) (0.3833) (0.4531) (0.4745) (0.4268) Japan (n=2424) (0.4128) (0.2084) (0.4403) (0.4628) (0.4507) Netherlands (n=1886) (0.4681) (0.3761) (0.4627) (0.4462) (0.4290) Sweden (n=1200) (0.4205) (0.3967) (0.4693) (0.4292) (0.4047) United States (n=2193) (0.4392) (0.4559) (0.4579) (0.4691) (0.4508) Total (n=12020) (0.4660) (0.3855) (0.4721) (0.4731) (0.4513) Eta (Significance) The table presents the mean and standard deviation (in parenthesis) of the probability that individuals within countries use a channel of communication for political information on a random day. There are also cross-national differences in correlations between political subcultures and the use of communication channels for political information. Table 5 presents the correlation coefficients between political subcultures and use of communication 8

10 channels within the six countries. When the coefficients for political subcultures and newspaper are compared, we notice that there are significant differences within all countries, but that the strengths of the differences are different. The strongest coefficient is found in Sweden, while Australia has the weakest. This indicates that the differences between political subcultures are larger in Sweden than in Australia. In a similar way, the correlation coefficients concerning news reports on TV are the highest in Sweden, the Netherlands and Germany, while Japan and Australia have the weakest correlation coefficients. Sweden and the United States have the strongest correlation between political subcultures and news reports on radio, whereas the weakest coefficient is found in the Netherlands. However, concerning the use of the Internet as a communication channel for political information, there are no significant differences between political subcultures in Sweden. The strongest correlation for the use of Internet is instead found in United States. Also concerning the correlation between political subcultures and talks with friends or colleagues, the United States has the strongest correlation. However, in Australia, Germany, the Netherlands and Sweden, the correlations are not significant. We found significant correlations between political subcultures and talks with friends or colleagues only in the United States and Japan. Table 5 Political subcultures and use of communication channels within countries Channel of communication Countries Newspapers TV Radio Internet Talks Australia (0.001) (0.095) Germany (0.478) Japan (0.034) (0.001) Netherlands (0.004) (0.194) Sweden (0.193) (0.311) USA Total The table presents the eta-coefficients and the level of significance (in parenthesis). More detailed information is presented in Table A1 (appendix). 9

11 Are there differences between political subcultures in the use of communication channels even when cross-national effects are included in the analyses? Table 6 presents analyses that include political subcultures (Table 3), countries (Table 4), and interactions between these two factors (Table 5). 6 According to these analyses, there are significant differences between the political subcultures when countries and interaction effects are included in the analyses. However, there are no differences between political subcultures in the use of talks with friends and colleagues. Instead, the largest differences between political subcultures are found in the use of newspapers, news reports on TV and on radio as the sources of political information, while the differences between political subcultures are minor in the use of the Internet. Compared to the cross-national differences, the differences between political subcultures are quite moderate. According to the analyses, the differences between the political subcultures are lower than four per cent in the use of all communication channels. This indicates that there are considerable differences within the political subcultures. Table 6 Use of communication channels: differences between political subcultures and countries Channel of communication Factor Newspapers TV Radio Internet Talks Political subcultures (0.305) Countries Interactions (0.001) (0.096) Total model (Eta 2 adj) N The table presents the partial eta squared coefficients (Eta 2 ) and level of significance (in parenthesis). 6 General linear models are used as method to analyze the material. 10

12 Conclusion According to these analyses, there is no general connection between the use of communication channels and political orientations. Firstly, the strength of the connection to political subcultures varies among the different channels of communication. There are moderate connections among political subcultures and the use of newspapers, news reports on TV and radio as sources for political information, while there are weak or no connections among the political subcultures and the use of the Internet and talks with friends or colleagues. As Figure 1 illustrates, there is a pattern in the differences among political subcultures. Civic citizens and critical citizens use the channels of communication more frequently as sources for political information than stealth citizens and disenchanted citizens. Secondly, the strength of the connection between the use of communication channels and political subcultures are different in the investigated countries. These differences do not follow any specific pattern, but there are significant differences in the use of communication channels for obtaining political information among the countries. This paper presents initial analyses about the connections between political communication and political orientations. The conclusions are therefore only preliminary and there are several questions for future studies. One question is if there are other aspects of political orientations than political interest and political trust that are relevant. For example, internal and external efficacy has been regarded as important aspects of political orientations (Almond & Verba 1965). Another question is if there are other measurements of the use of communication channels for the political information. For example, the categories of the communication channels used in the analyses include sub-channels, such as different TV channels or homepages. The differences among political subcultures may be on the sub-channels level instead of the channel level. A third question for future studies is if there is a theoretical framework that stipulates alternative connections between political communications and political orientations. This question is critical in the development of a model that explains how the use of political communications for political information is connected to political orientations. Related to this is the question about an alternative research design. We have used a onelevel design. An alternative design is a multilevel design. However, the use of a 11

13 multilevel design requires a theoretical model that is an alternative to the model used in this paper. Figure 1 The use of communication channels for political information and political subcultures Talks Newspapers TV Disenchanted Stealth Critical Civic Internet Radio 12

14 References Almond, G. A. & Verba, S. (1965): The Civic Culture Political Attitudes and Democracy in Five Nations. Boston: Little, Brown and Company. Coleman, J. S. (1990): Foundations of Social Theory. Harvard, MA: The Belknap Press of Harvard University Press. Dahlgren, P. (2005): The Internet, Public Spheres, and Political Communication: Dispersion and Deliberation, Political Communication 22: , van Deth, J. W. (2013): Equivalence in Comparative Political Research in van Deth, J. W. (ed.): Comparative Politics: The Problem of Equivalence. Colchester: ECPR Press. Downs, A. (1957): An Economic Theory of Democracy. New York: Harper & Row Publishers. Easton, D. & Dennis, J. (1969): Children in the Political System: Origins of Political Legitimacy. New Work: McGraw Hill. Greenstein, F. I. (1965): Children and Politics. New Haven: Yale University Press. Greenstein, F. I. (1968): Socialization: Political Socialization in Sills, D. L. (ed.): International Encyclopedia of the Social Science (vol. 14). New York: MacMillan & Free Publishing Press Company. Hallin, D. C. & Mancini, P. (2004): Comparing Media Systems: Three Models of Media and Politics. Cambridge: Cambridge University Press. Hay, C. & Stoker, G. (2009): Revitalising Politics: Have we lost the Plot?, Representation 45: Hyman, H. (1959): Political Socialization. New York: Free Press of Glencoe. Inglehart, R. (1977): The Silent Revolution: Changing Values and Political Styles Among Western Publics. Princeton: Princeton University Press. Inglehart, R. (1990): Culture Shift in Advanced Industrial Society. Princeton: Princeton University Press. Inglehart, R. & Norris, P. (2009): Cosmopolitan Communications: Cultural Diversity in a Globalized World. New York: Cambridge University Press. Kinder, D. R. (2003): Communication and Politics in the Age of Information in Sears, D. O., Huddy, L. & Jervis, R. (Eds.): Oxford Handbook of Political Psychology. Oxford: Oxford University Press. Lasswell, H. D. (1948): The Structure and Function of Communication in Society. New York: Lyman Bryson. 13

15 Lijphart, A. (1999): Patterns of Democracy: Government Forms and Performance in Thirty- Six Countries. New Haven: Yale University Press. Langton, K. P. (1969): Political Socialization. Oxford: Oxford University Press. Merelman, R. M. (1986): Revitalizing Political Socialization in Hermann, M. G. (red.): Political Psychology: Contemporary Problems and Issues. San Francisco: Jossey-Bass Publishers. Norris, P. (ed.) (1999): Critical Citizens: Global Support for Democratic Government. Oxford: Oxford University Press. Norris, P. (2002): Democratic Phoenix: Reinventing Political Activism. Cambridge: Cambridge University Press. Nye, J. S. & Zelikow, P. D. (1997): Conclusion: Reflections, Conjectures, and Puzzles in Nye, J. S., Zelikow, P. D. & King, D. C. (Eds.): Why People Don t Trust People. Cambridge, MA: Harvard University Press. Powell, B. G. (2000): Elections as Instruments of Democracy: Majoritarian and Proportional Visions. New Haven: Yale University Press. Putnam, R. D. (2000): Bowling Alone: The Collapse and Revival of American Community. New York: Simon & Schuster. Stoker, G. (2006): Why Politics Matters Making Democracy Work. Basingstoke: Palgrave Macmillan. Stoker, G. (2010): The rise of Political Disenchantment in Hay, C. (ed.): New Directions in Political Science: Responding to the Challenges of an Interdependent world. Basingstoke: Palgrave Macmillan. Tilly, C. (2005): Trust and Rule. Cambridge: Cambridge University Press. Wasburn, P. C. (1994): A Life Course Model of Political Socializations, Politics and the Individual 4: Zaller, J. R. (1992): The Nature and Origins of Mass Opinion. Cambridge: Cambridge University Press. 14

16 Appendix Calculation of the probability that an individual uses the communication channel i (CCi) for political information on one random day: P(CCi) = (((1/1)*SD) + ((1/7)*SW) +((1/30)*SM)+((1/60)*SL)+(0*SN))/100 where S D = share that state that they use the channel i for obtain political information daily S W = share that state that they use the channel i for obtain political information weekly S M = share that state that they use the channel i for obtain political information monthly S L = share that state that they use the channel i for obtain political information less than monthly S N = share that state that they never use the channel i for obtain political information 15

17 Table A1: Political subcultures and use of communication channels within countries II Political subculture Channel of communication Newspaper TV Radio Internet Conversation Australia (n= 1419) Disenchanted Critical Stealth Civic Total Eta (Significance) (0.001) (0.095) Germany (n=1961) Disenchanted Critical Stealth Civic Total Eta (Significance) (0.478) Japan (n=2020) Disenchanted Critical Stealth Civic Total Eta (Significance) (0.034) (0.001) Netherlands (n =1744) Disenchanted Critical Stealth Civic Total Eta (Significance) (0.004) (0.194) Sweden (n = 1147) Disenchanted Critical Stealth Civic Total Eta (Significance) (0.193) (0.311) United States (n = 2125) Disenchanted Critical Stealth Civic Total Eta (Significance)