Patterns of Individual Political Communication Among German Youth. Prof. Dr. Gerhard Vowe (HHU Düsseldorf, Department of Communication Science)

Transcription

1 Jan-Erik Wiederholz, M.Sc. Graduate School Linkage in Democracy (LinkDe) Institut für Sozialwissenschaften Heinrich-Heine-Universität Düsseldorf Universitätsstr. 1 D Düsseldorf ECPR Graduate Student Conference July 2014, University of Innsbruck Patterns of Individual Political Communication Among German Youth Supervisors: Prof. Dr. Gerhard Vowe (HHU Düsseldorf, Department of Communication Science) Prof. Dr. Peter H. Hartmann (HHU Düsseldorf, Department of Sociology) Prof. Dr. Martin Emmer (FU Berlin, Department of Communication Science) Abstract: This study investigates patterns of individual political communication (reception of political information, interpersonal political communication and political participation) among German youth, with a special focus on their use of political online communication. This research deals with the following questions: To what extent do German youth use online media for individual political communication? Which patterns of individual political communication can be identified? Based on a standardized web survey with young adults from Germany (N=1003) in August 2013, five types of individual political communication are distinguished. 1

2 Introduction Today s adolescents and young adults are often referred to as Digital Natives because of their (presumed) affinity towards all kinds of digital media. The basic idea brought up by Prensky (2001) states that the generation born after 1980 is fundamentally different from previous cohorts because they were the first ones to grow up with digital technology. According to Prensky, the Digital Natives therefore possess a sort of intuitive understanding of all things related to the digital world they are thought to be native speakers of the digital language. Although the Digital Immigrants, i.e. the ones who weren t socialized with digital media, may be able to learn this language as well, they will always remain an accent which distinguishes them from the real native speakers. As becomes apparent from this short summary, the term Digital Natives should be considered more of a metaphor than an actual theoretical concept, which has been one of the main criticisms by many scholars throughout the years. In public discourse, though, the term has become a popular phrase which underlines the significance of digital media for today s young generation. Indeed, studies consistently show that young people below the age of thirty use the Internet and other digital media far more often than any other cohort. On the other hand, however, there is also a great deal of heterogeneity within the group of Digital Natives itself regarding their patterns of Internet use. Obviously, the type of Internet usage is not determined by age alone, but also influenced by education, individual life styles and many other factors. Today, most scholars agree that the term Digital Natives is a very simplistic generalization which overstates both heterogeneity between age cohorts and homogeneity within one particular age group. However, despite all the justified criticisms, there is no doubt that the Internet plays indeed a central role in the everyday life of most adolescents and young adults in Germany. According to the German JIM study [Youth, Information, (Multi-)Media], children and 2

3 adolescents in Germany between 12 and 19 years spend most of their time online for communication purposes (e.g. chatting, writing s), followed by entertainment (e.g. watching videos, listening to music), playing online games, and searching for information (e.g. using search engines, reading online news). But it has also been shown that adolescents and young adults increasingly use online media instead of mass media as a source for political information (Emmer, Vowe, Wolling & Seifert 2011: 87 ff.; Hasebrink & Schmidt 2012: 54). Therefore, I want to investigate the following research questions: Research Questions 1. To what extent is the individual political communication of adolescents and young adults in Germany digitalized? Which patterns of individual political communication can be distinguished? 2. To what extent can the type of individual political communication be explained by resources, political orientation and network integration of adolescents and young adults? Theoretical background and hypotheses Although the research questions mentioned above deal with individual political communication, they are strongly related to theoretical concepts from political sociology and participation research. According to the classic definition by Verba et al. (1972: 2) "[p]olitical participation refers to those activities by private citizens that are more or less directly aimed at influencing the selection of governmental personnel and/or the actions they take." Traditionally, two main dimensions of political participation are distinguished: conventional and unconventional political participation. The main criteria to differentiate between the two forms is the aspect of (perceived) legitimacy: While conventional political 3

4 participation refers to activities which are perceived as legitimate, i.e. those which correspond to the norms of law and custom that regulate political partiticipation under a partiticular regime (Barnes & Kaase 1979: 41), unconventional political participation refers to the opposite, i.e. illegitimate forms of political participation. There are several problems with this definition, though. First, the perception of legitimacy may change over time: various forms of political participation which have been considered as unconventional back in the 1970 s (e.g. attending a rally or signing a petition) are now perceived by a majority as a legitimate way to influence political decisions. Second, the digitalization has led to new forms of political participation which didn t exist in the past. But does this mean that all forms of online participation should be considered per se as unconventional? This leads us to a more general question, the role of media in participation research. Traditionally, the reception of political information through mass media was treated as a form of political communication, not participation. Concerning online media, this distinction between political participation and political communication is much more difficult to make. Based on the theoretical works by Emmer et al. (Emmer, Füting & Vowe, 2006a; Emmer et al., 2006b; Emmer & Vowe, 2004; Emmer & Wolling, 2010; Füting, 2011; Vowe, Emmer & Seifert, 2007), I therefore propose an integrated model of political participation and political communication at the individual level. Emmer et al. differentiate between three dimensions of individual political communication: (1) Receptive Political Communication, i.e. the reception of political information via the media; (2) Interpersonal Political Communication, i.e. political discussions and contact with politicians; (3) Participatory Political Communication, i.e. different forms of political participation. For each dimension, a further distinction is made between an online and an offline form of political communication. Each (sub-)dimension is operationalized by a range of indicator variables, e.g. Frequency of 4

5 watching TV news as an indicator for the dimension Receptive Political Communication (for a list of all indicator variables see: Rössler, 2011, p. 324 f.): Table 1: Basic theoretical model Individual Political Communication Receptive Political Communication Interpersonal Political Communication Participatory Political Communication Offline e.g. TV news, daily newspaper e.g. political discussion with other person, contact with politician e.g. attending rally, signing petition Online e.g. searching political information online e.g. political discussion with other person online, contact with politician online e.g. signing online petition Emmer, Vowe & Wolling (2011: 27) One of the most popular theoretical explanations for political participation is the Civic Voluntarism Model by Brady, Verba and Schlozman (1995). To understand why citizens participate in political processes, Brady et al. (1995) inverted the question and investigated the reasons why people don t take part in politics. The researchers came up with the following answers: because they can t, because they don t want to, or because nobody asked. (Brady et al. 1995: 271) "They can't" suggests a paucity of necessary resources: time to take part in political activity, money to make contributions, and civic skills (i.e., the communications and organizational skills that facilitate effective participation). "They don't want to" focuses on the absence of psychological engagement with politics a lack of interest in politics, minimal concern with public issues, a sense that activity makes no difference, and no consciousness of membership in a group with shared political interests. "Nobody asked" implies isolation from the recruitment networks through which citizens are mobilized to politics. (ibid) Resources, engagement with politics and network integration are thought to be the three basic determinants of political participation. Each of them can be operationalized in further detail. First, we can distinguish between different types of resources: available time 5

6 (e.g. occupational status), economic resources (e.g. income), and intellectual resources (e.g. educational level). Second, we can distinguish between different motives for political participation: political interest (e.g. internal efficacy), political orientation (e.g. party affiliation) and the support of the political system (e.g. political trust). Third, we can distinguish between different indicators for the integration in social networks: social participation (e.g. membership in volunteer associations), social trust (e.g. general trust in other people) and the frequency of meetings with friends. Based on the Civic Voluntarism Model by Brady et al. (1995), the following hypotheses are tested: H1: The more resources an individual has (time, economic resources, intellectual skills), the more intense the individual s political communication. H2: The stronger an individual s involvement with politics (political interest, political orientation, political efficacy), the more intense the individual s political communication. H3: The stronger an individual s network integration (organization membership, meeting with friends, contact with neighbors), the more intense the individual s political communication. Data analysis My data analysis is based on standardized web survey of young adults from Germany who have been born between 1980 and The data collection took place in August The candidates were recruited via an online access panel and selected using quotas based on the average demographic composition (age, gender, educational background) of the German population born between 1980 and invitations were sent out, which resulted in 1075 completed questionnaires (21% response rate). The median duration for completion was 12 minutes. 72 respondents with a completion time of less than 6 minutes were removed due to speeding, which leads to a final sample size of N=

7 In the dataset, there are 20 indicator variables of the dimension Receptive Political Communication ( Political Information ), 5 indicators of the dimension Interpersonal Political Communication ( Political Discussion ) and 10 indicators of the dimension Participatory Political Communication ( Political Participation ). All variables have been recoded as dummy variables with a value of either 0 or 1. See the annex for an overview of all variables and how they are operationalized. Table 2: Empirical model The indicators are being used as active variables in cluster analysis to identify more general patterns of individual political communication. Wiedenbeck and Züll (2010: 537) recommend using standardized variables in cluster analysis because otherwise variables with higher variance would have a stronger impact on the result than variables with lower variance. One of the most commonly used methods of standardization is the z-transformation. Z-values always have a mean of 0 and a standard deviation of 1 (see Table 3). Using the standardized variables, I first run a hierarchical cluster analysis to determine the ideal number of clusters (Wiedenbeck & Züll, 2010: 537). The hierarchical cluster analysis is run with the City-Block-Metric as a dissimilarity measure, which is identical to 7

8 the squared Euclidean distance for dichotomous variables (Bacher, 2010: 196 ff.). As fusion algorithm, the method Average Linkage Within Groups is chosen (Bacher, 2010: 264 ff.). The fusion values reveal that there is a gap between the fourth and the fifth cluster, so the ideal number of clusters is assumed to be four (Bacher, 2010: 241 ff.). Table 4 shows the results of the hierarchical cluster analysis. Unfortunately, the cluster sizes are very heterogeneous: while Cluster 1 (N=501) and Cluster 2 (N=470) are very large, Cluster 3 (N=17) and Cluster 4 (N=15) are extremely small, which makes an interpretation very difficult. To refine the result of the hierarchical cluster analysis and to get more evenly distributed cluster sizes, the means of the standardized variables across the four groups are used as starting values for a k-means cluster analysis. The results are reported in Table 5. Cluster 1 (N=444) and Cluster 2 (N=424) are slightly reduced in size, but still much larger than Cluster 3 (N=44) and Cluster 4 (N=91). Since the mean scores of Cluster 3 are much higher than those of the other clusters, we would assume that Cluster 3 is by far the most politically active type. But can we be sure that the results actually valid? The median duration for completing the questionnaire in Cluster 3 (10 minutes) is more than two minutes lower than the overall average (12 minutes). Furthermore, we can see from Table 6 that the respondents in Cluster 3 are not very highly educated, which is unusual given their high political activity. Since I have reason to believe that the respondents in Cluster 3 may be speeders, I excluded them from the dataset and then repeated the steps mentioned before. Since the sample size and therefore the means have changed, I have computed new standardized variables (see Table 7). Afterwards, I have run the hierarchical cluster analysis again with the same settings as before. This time, the fusion values suggest that the ideal number of clusters should be five (see Table 8). Once again, the cluster solution is optimized by using the means of the active variables across the five clusters as starting values for a k- means cluster analysis. The final results are reported in Table 9. 8

9 Results The five clusters can be ordered according to their degree of political activity, from Cluster 1 (least active) to Cluster 5 (most active). In the following, I will describe each cluster in more detail. The respondents in Cluster 1 (N=266) are named Refuseniks because they have the lowest rates of political activity across all categories. They can be characterized as people who refuse to participate even in the most basic political activities: they don t search for political information, they don t talk about political issues with others and their reported voter turnout in the last two national elections is the lowest of all clusters. In particular, they don t use online media for political communication. Two thirds of the Refuseniks are female (66%). They have the lowest mean age of all clusters (25.7 years) and also the lowest educational level: 41% have no more than lower secondary education, only 12% have college education. 52% currently do not have an occupation and 21% have at least one parent who has not been born in Germany. They spend about 3.2 hours per day online, which is the second-highest amount of all groups. They have little or no interest in politics and are the least satisfied with the way democracy works in Germany. All in all, this cluster mainly consists of socially deprived young adults who have no connection to public affairs. Because of their lack of education, they feel overwhelmed and confused by the complexity of political processes. The respondents in Cluster 2 (N=332) are labeled Indifferent because they do not avoid political information on purpose as the Refuseniks, but they are not actively seeking it, either they are indifferent. Their only major source for political information is offline TV news (private, public or news channels). In contrast to the Refuseniks, they also participate in national elections. But apart from these very basic political activities, the Indifferent prefer not to engage in politics. 9

10 More than half of the Indifferent are female (54%), their mean age is 26.9 years. Their educational level is close to the total average: 46% have a higher secondary education, 30% have college education. About half of them (47%) are currently working full-time. They have a medium political interest, moderate political views and are rather satisfied with the way democracy works in Germany. All in all, the Indifferent can be considered as average in many aspects. The respondents in Cluster 3 (N=155) are named Slacktivists because they only seem to participate in politics if it doesn t involve too much trouble for them. In addition to TV news, they especially use local or regional newspapers as a source for political information, which could suggest that they are particularly interested in local politics. In contrast to both Refuseniks and Indifferents, the Slacktivists also make use of online media for political communication. They discuss political issues with others both online and offline. In addition to voting, they mainly participate in politics by signing petitions. This has become a very popular form of political participation because it doesn t require much effort or involvement. 54% of the Slacktivists are female, their mean age is 27.1 years. Their educational level is significantly above average: 63% have a higher secondary education, 43% have college education. Their political views are similar to the Indifferent, but their political interest is higher and their democracy satisfaction is lower. The respondents in Cluster 4 (N=132) are named Digital Citizens. They use a variety of both online and offline media for political communication. They can be characterized as heavy users of political information. They also like to discuss political issues with others, both online and offline. Given their apparently high political interest, they are surprisingly reluctant as far as political action is concerned. They don t engage in any form of political participation except voting, therefore their interest in politics can be considered rather intellectual. 10

11 Two thirds of the Digital Citizens are male, their mean age is 27 years. Their educational level is the highest of all groups: 64% have a higher secondary education, 52% have college education. 54% are working full-time. They have a high political interest, moderate political views and are rather satisfied with the way democracy works in Germany. The profile of the respondents in Cluster 5 (N=74) is complementary to the Digital Citizens : they are labeled Missionaries because their focus is less on intellectual discourse, but rather on political action. Consequently, they consume less political information than the Digital Citizens, but they discuss political issues more frequently, especially in public. They seem to be on a mission to win as many people as possible for their political cause. Correspondingly, they also engage very actively in various forms of political participation. 72% of the Missionaries are male, the mean age is 26.1 years. Their educational level is similar to the Indifferent: 47% have a higher secondary education, 32% have college education. 47% are working full-time. They have the highest political interest of all groups, but their democracy satisfaction is rather low. They have more radical political views, 47% consider themselves as left or rather left. Conclusion Based on a standardized web survey with N=1003 young adults from Germany in August 2013, five types of individual political communication are distinguished using cluster analysis: Refuseniks (N=266), Indifferent (N=332), Slacktivists (N=155), Digital Citizens (N=132) and Missionaries (N=74). The five clusters can be ordered according to their degree of political activity, from Refuseniks (least active) to Missionaries (most active). This typology will serve as a dependent variable for further analysis. 11

12 Annex 1. Indicators of the dimension Political Information : - Information sources offline: has a value of 1 if at least one of the following offline sources for political information is very important or important to the respondent: a) newspaper, b) TV news, c) radio news, d) family and friends. - Information sources online: has a value of 1 if at least one of the following online sources for political information is very important or important to the respondent: a) news sites, b) news blogs, c) social networking sites, d) websites of political organizations. - Tabloid press offline/online: has a value of 1 if the respondent reads BILD (offline/online) at least once every week. - National press offline/online: has a value of 1 if the respondent reads one of the following national newspapers (offline/online) at least once every week: a) die tageszeitung, b) Die Welt, c) Frankfurter Allgemeine Zeitung, d) Frankfurter Rundschau, e) Handelsblatt, f) Süddeutsche Zeitung. - Regional press offline/online: has a value of 1 if the respondent reads a regional newspaper (offline/online) at least once every week. - News magazine offline/online: has a value of 1 if the respondent reads one of the following news magazines (offline/online) at least once every two weeks: a) Der Spiegel, b) Die Zeit, c) Focus, d) Stern. - Public TV news offline/online: has a value of 1 if the respondent watches one of the following public broadcast TV news (offline/online) at least once every week : a) heute/heutejournal, b) Tagesschau/Tagesthemen. 12

13 - Private TV news offline/online: has a value of 1 if the respondent watches one or more of the following private broadcast TV news (offline/online) at least once every week : a) ProSieben Newstime, b) RTL Aktuell/RTL Nachtjournal, c) SAT.1 Nachrichten. - News channel offline/online: has a value of 1 if the respondent watches one or more of the following TV news channels (offline/online) at least once every week : a) N24, b) n-tv. - Informational program offline/online: has a value of 1 if the respondent watches one or more of the following weekly informational programs (offline/online) at least once every two weeks: a) Auslandsjournal, b) Bericht aus Berlin, c) Berlin direkt, d) Weltspiegel. - Politainment program offline/online: has a value of 1 if the respondent watches one or more of the following weekly politainment programs (offline/online) at least once every two weeks: a) Günther Jauch, b) heute-show, c) Hart aber fair, d) Maybrit Illner. 2. Indicators of the dimension Political discussion : - Political activities on social networking site: has a value of 1 if the respondent has performed one or more of the following activities on a social networking site during the last month: a) posted political status update, b) posted link to political website, c) posted political picture, d) posted political video, e) liked other s political posting, f) commented on other s political posting, g) shared other s political posting. - Private political discussion offline/online: has a value of 1 if the respondent discusses political issues (offline/online) very frequently or frequently with at least one of the following persons: a) family members, b) friends, c) acquaintances (e.g. colleagues), d) strangers. - Public political discussions offline: has a value of 1 if the respondent has performed one or more of the following activities site during the last 12 months: a) wrote a letter to the editor, b) spoke out during an assembly meeting, c) wore political badge. 13

14 - Public political discussions online: has a value of 1 if the respondent has performed one or more of the following activities during the last 12 months: a) commented political article on online news site, b) commented blog entry on political issue, c) posted in online forum on political issue. 3. Indicators of the dimension Political participation : - Conventional participation offline/online: has a value of 1 if the respondent has performed one or more of the following activities (offline/online) during the last 12 months: a) contacted politician, b) donated money for political cause, c) supported election campaign. - Unconventional participation offline/online: has a value of 1 if the respondent has performed one or more of the following activities (offline/online) during the last 12 months: a) attended rally/flash mob, b) attended sit-in/ddos-attack. - Signed petition offline/online: has a value of 1 if the respondent has signed a petition (offline/online) during the last 12 months. - (Active) Membership in political organization: has a value of 1 if the respondent has been (an active) member in one or more of the following political organizations during the last 12 months: a) political party, b) trade union, c) environmentalist group, d) human rights group. - Voted in national election 2009/2013: has a value of 1 if the respondent has voted in the German national elections 2009/

15 Table 3: Descriptive statistics of active variables (N=1003) N Unstandardized Standardized (z-values) Political Information Min. Max. Mean SD Min. Max. Mean SD Importance: Information offline ,79 0,41-0,41 2, Importance: Information online ,62 0,49-0,45 2, National newspaper offline ,16 0,37-0,40 2, National newspaper online ,20 0,40-0,58 1, Regional newspaper offline ,16 0,37-0,79 1, Boulevard newspaper offline ,09 0,28-0,38 2, Boulevard newspaper online ,14 0,35-0,91 1, Regional newspaper online ,17 0,37-0,32 3, News magazine offline ,14 0,34-0,60 1, News magazine online ,25 0,43-0,31 3, Public TV news offline ,38 0,49-0,31 3, Public TV news online ,13 0,33-0,20 4, Private TV news offline ,45 0,50-0,43 2, Private TV news online ,09 0,29-0,27 3, News channel offline ,26 0,44-0,78 1, News channel online ,09 0,28-0,90 1, Informational program offline ,09 0,28-0,46 2, Informational program online ,04 0,19-0,39 2, Politainment program offline ,16 0,37-0,57 1, Politainment program online ,07 0,25-0,39 2, Political discussion Discussion on social networking site ,38 0,49-0,35 2, Private discussion offline ,45 0,50-0,58 1, Private discussion online ,18 0,38-0,53 1, Public discussion offline ,13 0,34-0,35 2, Public discussion online ,25 0,43-0,27 3, Political participation Conventional participation offline ,13 0,34-0,46 2, Conventional participation online ,11 0,31-0,30 3, Unconventional participation offline ,11 0,32-0,35 2, Unconventional participation online ,07 0,25-0,27 3, Signed petition offline ,25 0,44-1,25 0, Signed petition online ,22 0,41-1,78 0, Organization membership ,17 0,38-0,46 2, Active member in organization ,08 0,28-0,30 3, Voted in national election ,61 0,49-1,25 0, Voted in national election ,76 0,43-1,78 0,

16 Table 4: Results of hierarchical cluster analysis (N=1003) Cluster 1 Cluster 2 Cluster 3 Cluster 4 N=501 N=470 N=17 N=15 Political Information Importance: Information offline -0,50 0,50 0,52 0,36 Importance: Information online -0,42 0,40 0,79 0,79 National newspaper offline -0,09-0,01 2,28 0,65 National newspaper online -0,18 0,10 1,97 0,65 Regional newspaper offline -0,23 0,26-0,44 0,11 Regional newspaper offline -0,17 0,18-0,31 0,40 Boulevard newspaper online -0,03-0,02 1,43-0,22 Boulevard newspaper online -0,07 0,02 1,44 0,09 News magazine offline -0,04-0,05 2,52 0,19 News magazine online -0,19 0,13 1,73 0,35 Public TV news offline -0,38 0,37 0,91 0,04 Public TV news online -0,12 0,06 2,08-0,38 Private TV news offline -0,10 0,10 0,28-0,10 Private TV news online 0,15-0,23 1,94-0,08 News channel offline -0,17 0,12 1,28 0,31 News channel online -0,12 0,04 2,62-0,07 Informational program offline -0,14 0,05 2,83-0,31 Informational program online -0,06-0,09 4,36-0,20 Politainment program offline -0,16 0,08 2,14 0,30 Politainment program online -0,02-0,11 3,42-0,01 Political discussion Discussion on social networking site -0,10 0,03 1,27 1,14 Private discussion offline -0,32 0,28 1,11 0,71 Private discussion online -0,01-0,11 2,15 1,28 Public discussion offline 0,02-0,18 2,40 2,38 Public discussion online -0,11 0,00 1,75 1,75 Political participation Conventional participation offline -0,06-0,11 2,56 2,56 Conventional participation online -0,02-0,16 2,86 2,65 Unconventional participation offline -0,02-0,11 1,70 2,18 Unconventional participation online 0,00-0,15 2,52 1,83 Signed petition offline -0,20 0,12 1,17 1,71 Signed petition online -0,18 0,09 1,31 1,88 Organization membership -0,02-0,08 0,94 2,19 Active member in organization 0,03-0,18 1,19 3,33 Voted in national election ,42 0,43 0,08 0,66 Voted in national election ,48 0,49 0,15 0,56 16

17 Table 5: Results of K-means cluster analysis (N=1003) Cluster 1 Cluster 2 Cluster 3 Cluster 4 N=444 N=424 N=44 N=91 Political Information Importance: Information offline -0,57 0,49 0,41 0,31 Importance: Information online -0,69 0,52 0,70 0,56 National newspaper offline -0,36 0,06 2,03 0,49 National newspaper online -0,45 0,14 1,75 0,69 Regional newspaper offline -0,21 0,25-0,25-0,02 Regional newspaper offline -0,23 0,24-0,07 0,04 Boulevard newspaper online -0,18 0,04 1,14 0,12 Boulevard newspaper online -0,24 0,13 1,01 0,05 News magazine offline -0,31-0,04 2,13 0,69 News magazine online -0,48 0,23 1,37 0,59 Public TV news offline -0,51 0,45 0,57 0,14 Public TV news online -0,32 0,14 1,25 0,31 Private TV news offline -0,07 0,03 0,46-0,04 Private TV news online -0,13-0,14 1,82 0,41 News channel offline -0,31 0,24 0,64 0,08 News channel online -0,26 0,09 1,55 0,08 Informational program offline -0,30 0,02 2,76 0,04 Informational program online -0,19-0,16 3,91-0,20 Politainment program offline -0,38 0,13 1,81 0,41 Politainment program online -0,25-0,11 2,94 0,29 Political discussion Discussion on social networking site -0,46 0,16 0,99 1,03 Private discussion offline -0,69 0,43 1,11 0,80 Private discussion online -0,38-0,01 1,62 1,12 Public discussion offline -0,33-0,21 1,70 1,80 Public discussion online -0,45 0,04 1,22 1,44 Political participation Conventional participation offline -0,35-0,20 1,89 1,75 Conventional participation online -0,31-0,24 1,84 1,70 Unconventional participation offline -0,25-0,12 1,30 1,18 Unconventional participation online -0,21-0,18 1,79 0,99 Signed petition offline -0,43 0,14 0,82 1,03 Signed petition online -0,37 0,12 0,73 0,93 Organization membership -0,22-0,06 0,81 0,97 Active member in organization -0,21-0,18 1,10 1,33 Voted in national election ,41 0,35 0,38 0,17 Voted in national election ,50 0,39 0,30 0,46 17

18 Table 6: Cross-validation of clusters with passive variables (N=1003) 1. Resources Cluster 1 Cluster 2 Cluster 3 Cluster 4 Total N=444 N=424 N=44 N=91 N=1003 % N % N % N % N % N a) Secondary education Low 31% % 55 14% 6 16% 15 21% 214 Middle 39% % % 23 33% 30 34% 344 High 30% % % 15 51% 46 44% 445 b) Tertiary education No education (yet) 23% % 46 9% 4 19% 17 17% 167 Job training 60% % % 28 49% 45 53% 526 College education 17% 76 43% % 11 32% 29 30% 297 c) Occupational status Not working (yet) 49% % % 12 41% 37 44% 436 Working part-time 12% 54 12% 49 5% 2 12% 11 12% 116 Working full-time 39% % % 30 47% 43 45% Attitudes a) Political interest Low 64% % % 3 33% 327 Middle 31% % % 14 24% 22 40% 397 High 5% 20 35% % 30 73% 66 27% 263 b) Political orientation Left-wing 21% 95 31% % 14 44% 40 28% 280 Center 34% % % 13 33% 30 36% 366 Right-wing 16% 70 18% 78 34% 15 20% 18 18% 181 Refused/Don't know 28% % 45 5% 2 3% 3 18% 176 c) Democracy satisfaction Low 23% % 55 11% 5 27% 25 18% 185 Middle 41% % % 17 37% 34 42% 422 High 19% 86 39% % 22 35% 32 31% 306 Refused/Don't know 18% 78 3% % Social integration a) Organization membership No member 71% % % 13 51% 46 59% 594 Passive member 8% 37 10% 42 14% 6 12% 11 10% 96 Active member 21% 92 38% % 25 37% 34 31% 313 b) Meeting with friends Rarely 9% 39 3% 13 5% 2 4% 4 6% 58 Sometimes 20% 86 17% 70 14% 6 11% 10 18% 172 Often 70% % % 36 85% 77 77% 749 c) Contact with neighbors Rarely 31% % 95 16% 7 17% 15 25% 249 Sometimes 11% 45 11% 48 2% 1 12% 11 11% 105 Often 59% % % 35 71% 64 64%

19 Table 7: Descriptive statistics of active variables (N=959) Descriptive statistics N Unstandardized z-standardization Political Information Min. Max. Mean SD Min. Max. Mean SD Importance: Information offline ,78 0,42-1,87 0, Importance: Information online ,60 0,49-1,22 0, National newspaper offline ,13 0,33-0,38 2, National newspaper online ,17 0,38-0,46 2, Regional newspaper offline ,16 0,37-0,44 2, Regional newspaper offline ,09 0,28-0,31 3, Boulevard newspaper online ,13 0,33-0,38 2, Boulevard newspaper online ,15 0,36-0,42 2, News magazine offline ,10 0,30-0,34 2, News magazine online ,22 0,42-0,53 1, Public TV news offline ,37 0,48-0,77 1, Public TV news online ,11 0,31-0,35 2, Private TV news offline ,44 0,50-0,89 1, Private TV news online ,07 0,25-0,27 3, News channel offline ,25 0,43-0,58 1, News channel online ,07 0,25-0,27 3, Informational program offline ,05 0,22-0,23 4, Informational program online ,00 0,06-0,06 15, Politainment program offline ,13 0,33-0,38 2, Politainment program online ,04 0,19-0,19 5, Political discussion Discussion on social networking site ,36 0,48-0,75 1, Private discussion offline ,42 0,49-0,85 1, Private discussion online ,15 0,36-0,42 2, Public discussion offline ,10 0,31-0,34 2, Public discussion online ,22 0,42-0,53 1, Political participation Conventional participation offline ,10 0,30-0,34 2, Conventional participation online ,08 0,28-0,30 3, Unconventional participation offline ,09 0,29-0,32 3, Unconventional participation online ,05 0,21-0,22 4, Signed petition offline ,24 0,43-0,56 1, Signed petition online ,21 0,40-0,51 1, Organization membership ,16 0,37-0,43 2, Active member in organization ,07 0,25-0,27 3, Voted in national election ,60 0,49-1,23 0, Voted in national election ,75 0,43-1,75 0,

20 Table 8: Results of hierarchical cluster analysis (N=959) Cluster 1 Cluster 2 Cluster 3 Cluster 4 Cluster 5 N=414 N=370 N=118 N=47 N=10 Political Information Importance: Information offline -0,35 0,15 0,51 0,53 0,53 Importance: Information online -0,30 0,15 0,19 0,77 0,82 National newspaper offline -0,09 0,08-0,31 0,64 1,42 National newspaper online -0,23 0,07-0,07 1,46 0,87 Regional newspaper offline -0,16-0,12 1,02-0,33 0,37 Regional newspaper offline -0,17 0,00 0,62-0,08 0,40 Boulevard newspaper online -0,05 0,18-0,25-0,38-0,08 Boulevard newspaper online -0,06 0,14-0,28 0,00 0,42 News magazine offline -0,01-0,06 0,03 0,44 0,32 News magazine online -0,17-0,02-0,02 1,67 0,43 Public TV news offline -0,37 0,10 0,60 0,91 0,27 Public TV news online -0,12-0,11 0,31 1,22-0,35 Private TV news offline 0,07 0,14-0,44-0,63 0,32 Private TV news online 0,18-0,14-0,20-0,01 0,13 News channel offline -0,10 0,17-0,16-0,18 0,58 News channel online -0,04-0,04 0,14 0,41-0,27 Informational program offline -0,09 0,14-0,15 0,15-0,23 Informational program online 0,05-0,06-0,06 0,27-0,06 Politainment program offline -0,07 0,07-0,03 0,06 0,21 Politainment program online 0,10-0,07-0,15 0,15-0,19 Political discussion Discussion on social networking site -0,03-0,07-0,01 0,54 1,34 Private discussion offline -0,28 0,00 0,67 0,61 0,97 Private discussion online 0,01-0,08 0,01 0,06 2,11 Public discussion offline 0,03-0,01-0,15-0,34 2,60 Public discussion online -0,13 0,02 0,20 0,08 1,87 Political participation Conventional participation offline -0,05 0,02-0,06-0,20 2,95 Conventional participation online -0,02-0,04 0,01-0,14 2,97 Unconventional participation offline 0,01-0,05-0,06-0,17 2,78 Unconventional participation online 0,08-0,07-0,11-0,22 1,65 Signed petition offline -0,19-0,08 0,81-0,16 1,79 Signed petition online -0,16-0,14 0,83 0,02 1,96 Organization membership -0,02-0,01-0,04-0,14 2,30 Active member in organization 0,03-0,07-0,10-0,27 3,68 Voted in national election ,81 0,66 0,54 0,47 0,61 Voted in national election ,63 0,46 0,55 0,37 0,57 20

21 Table 9: Results of K-means cluster analysis (N=959) Cluster 1 Cluster 2 Cluster 3 Cluster 4 Cluster 5 N=266 N=332 N=155 N=132 N=74 Political Information Importance: Information offline -1,19 0,51 0,41 0,48 0,31 Importance: Information online -0,95 0,21 0,33 0,68 0,60 National newspaper offline -0,31-0,21-0,17 1,12 0,43 National newspaper online -0,40-0,32-0,27 1,49 0,76 Regional newspaper offline -0,32 0,01 0,80-0,30-0,04 Regional newspaper offline -0,26-0,24 1,08-0,20 0,07 Boulevard newspaper online -0,19 0,02-0,03 0,28 0,15 Boulevard newspaper online -0,17-0,08 0,07 0,41 0,11 News magazine offline -0,26-0,17-0,10 0,71 0,64 News magazine online -0,45-0,30 0,15 1,20 0,54 Public TV news offline -0,60 0,08 0,30 0,55 0,18 Public TV news online -0,28-0,15-0,02 0,72 0,43 Private TV news offline -0,18 0,19 0,04-0,08-0,15 Private TV news online -0,04-0,13-0,14 0,34 0,44 News channel offline -0,40 0,19 0,11 0,18 0,05 News channel online -0,21-0,12 0,09 0,49 0,22 Informational program offline -0,21-0,12-0,06 0,70 0,20 Informational program online -0,01-0,06-0,06 0,29-0,06 Politainment program offline -0,33-0,19 0,08 0,75 0,51 Politainment program online -0,13-0,16-0,16 0,50 0,61 Political discussion Discussion on social networking site -0,44-0,21 0,30 0,48 1,03 Private discussion offline -0,77-0,04 0,34 0,79 0,82 Private discussion online -0,37-0,16-0,06 0,58 1,14 Public discussion offline -0,28-0,22-0,24 0,18 2,18 Public discussion online -0,42-0,30 0,24 0,41 1,58 Political participation Conventional participation offline -0,30-0,20-0,13-0,07 2,37 Conventional participation online -0,29-0,27 0,03-0,11 2,35 Unconventional participation offline -0,24-0,24 0,17 0,07 1,45 Unconventional participation online -0,19-0,11-0,04-0,05 1,36 Signed petition offline -0,47-0,40 1,00 0,08 1,22 Signed petition online -0,45-0,44 1,21-0,06 1,19 Organization membership -0,27-0,12 0,10 0,02 1,27 Active member in organization -0,21-0,11-0,12-0,21 1,86 Voted in national election ,65 0,24 0,22 0,30 0,26 Voted in national election ,83 0,24 0,36 0,40 0,48 21

22 Literature Barnes, S. H. & Kaase, M. (1979). Political action. Mass participation in five western democracies. Beverly Hills: Sage Publ. Brady, H. E., Verba, S. & Schlozman, K. L. (1998). Beyond SES: A Resource Model of Political Participation. In P. Norris (Hrsg.), Elections and voting behaviour. New challenges, new perspectives (S ). Aldershot ;, Brookfield, Vt: Ashgate, Dartmouth. Buse, M., Nelles, W., Oppermann, R. & Hahn, R. (1978). Determinanten politischer Partizipation. Theorieansatz und empirische Überprüfung am Beispiel der Stadtsanierung Andernach (Studien zum politischen System der Bundesrepublik Deutschland, Bd. 20). Meisenheim am Glan: Hain. Emmer, M. (2005). Politische Mobilisierung durch das Internet? Eine kommunikationswissenschaftliche Untersuchung zur Wirkung eines neuen Mediums. München: Verlag Reinhard Fischer. Emmer, M., Füting, A. & Vowe, G. (2006). Wer kommuniziert wie über politische Themen? Eine empirisch basierte Typologie individueller politischer Kommunikation. Medien & Kommunikationswissenschaft, 54 (2), Zugriff am Verfügbar unter Emmer, M., Seifert, M. & Vowe, G. (2006). Internet und politische Kommunikation: die Mobilisierungsthese auf dem Prüfstand. Ergebnisse einer repräsentativen Panelstudie in Deutschland. In P. Filzmaier, M. Karmasin & C. Klepp (Hrsg.), Politik und Medien, Medien und Politik (S ). Wien: WUV. Emmer, M. & Vowe, G. (2004). Mobilisierung durch das Internet? Ergebnisse einer empirischen Längsschnittuntersuchung zum Einfluss des Internets auf die politische Kommunikation der Bürger. Politische Vierteljahresschrift, 45 (2), Emmer, M., Vowe, G., Wolling, J. & Seifert, M. (Hrsg.). (2011). Bürger online. Die Entwicklung der politischen Online-Kommunikation in Deutschland. Konstanz: UVK Verlagsgesellschaft mbh. Emmer, M. & Wolling, J. (2007). Leben in verschiedenen Welten? Themenagenden von Offlinern und Onlinern im Vergleich. In S. Kimpeler, M. Mangold & W. Schweiger (Hrsg.), Die digitale Herausforderung. Zehn Jahre Forschung zur computervermittelten Kommunikation (1. Aufl., S ). Wiesbaden: VS Verlag für Sozialwissenschaften. Emmer, M. & Wolling, J. (2009). 'Online Citizenship'? Die Entwicklung der individuellen politischen Beteiligung im Internet. In L. Novy (Hrsg.), Lernen von Obama? Das Internet als Ressource und Risiko für die Politik (S ). Gütersloh: Verl. Bertelsmann-Stiftung. Emmer, M. & Wolling, J. (2010). Online-Kommunikation und politische Öffentlichkeit. In K. Beck & W. Schweiger (Hrsg.), Handbuch Online-Kommunikation (1. Aufl., S ). Wiesbaden: VS Verlag für Sozialwissenschaften. Emmer, M., Wolling, J. & Vowe, G. (2010). Ein Medium wird erwachsen. Die Entwicklung der politischen Internetnutzung der Deutschen von In J. Wolling, M. Seifert & M. Emmer (Hrsg.), Politik 2.0? Die Wirkung computervermittelter Kommunikation auf den politischen Prozess (1. Aufl., S ). Baden-Baden: Nomos, Ed. Fischer. Füting, A. (2011). Wie kommunizieren die Deutschen über Politik? Eine typologische Längsschnittanalyse. In M. Emmer, G. Vowe, J. Wolling & M. Seifert (Hrsg.), Bürger online. Die Entwicklung der politischen Online-Kommunikation in Deutschland (S ). Konstanz: UVK Verlagsgesellschaft mbh. 22

23 Gabriel, O. W. & Brettschneider, F. (2002). Politische Partizipation. In O. Jarren (Hrsg.), Politische Kommunikation in der demokratischen Gesellschaft. Ein Handbuch mit Lexikonteil (1. Aufl.). Opladen: Westdt. Verl. Gorgs, C. & Meyer, P. (1999). Fernsehen und Lebensstile. Eine empirische Studie über den Medienkonsum am Beispiel des Grossraums Augsburgs. München: R. Hampp. Hasebrink, U. & Schmidt, J.-H. (2012). Informationsrepertoires der deutschen Bevölkerung. Konzept für eine regelmäßig durchzuführende bevölkerungsrepräsentative Befragung im Rahmen des Vorhabens "Erfassung und Darstellung der Medien- und Meinungsvielfalt in Deutschland" (Arbeitspapiere des Hans-Bredow-Instituts, Bd. 24, Aufl.). Hamburg: Hans-Bredow-Inst. für Medienforschung an der Univ. Hamburg Verl. Rössler, P. (2011). Skalenhandbuch Kommunikationswissenschaft. Wiesbaden: VS Verlag für Sozialwissenschaften. Seifert, M. (2012). Mobilisierung für alle? Sozial selektive Wirkungen des Internets auf die politische Kommunikation. Wiesbaden: VS Verlag für Sozialwissenschaften / Springer Fachmedien Wiesbaden GmbH, Wiesbaden. van Eimeren, B. & Frees, B. (2012). Ergebnisse der ARD/ZDF-Onlinestudie Prozent der Deutschen online neue Nutzungssituationen durch mobile Endgeräte. Media Perspektiven (7-8), Zugriff am Verfügbar unter Verba, S., Schlozman, K. L. & Brady, H. E. (1995). Voice and equality. Civic voluntarism in American politics. Cambridge, Mass: Harvard University Press. Verba, S., Nie, N. H. & Kim, J.-O. (1978). Participation and political equality. A seven-nation comparison. Cambridge: Cambridge Univ. Pr. Vowe, G., Emmer, M. & Seifert, M. (2007). Abkehr oder Mobilisierung? Zum Einfluss des Internets auf die individuelle politische Kommunikation. Empirische Befunde zu alten Fragen im Kontext neuer Medien. In B. Krause, B. Fretwurst & J. Vogelgesang (Hrsg.), Fortschritte der politischen Kommunikationsforschung. Festschrift für Lutz Erbring (S ). Wiesbaden: VS, Verlag für Sozialwiss. Wolling, J., Seifert, M. & Emmer, M. (Hrsg.). (2010). Politik 2.0? Die Wirkung computervermittelter Kommunikation auf den politischen Prozess (1. Aufl.). Baden-Baden: Nomos, Ed. Fischer. 23