Winning over Young People to Engage in Politics
Assessing how Political Communication affects Young People’s
Participation Preferences
Johanna M. Mertens11181362 Master’s Thesis
Graduate School of Communication Master’s programme Communication Science
Supervisor: Dr. Franziska Marquart 30th of June 2017
Abstract
Despite the vast research on changing patterns of young people’s engagement in traditional and new forms of participation, little is known about the impact of political communication on the choice of the participation form. This study uses an experiment with young people aged 18 to 24 (N = 209), testing how the type of political content (concrete vs. abstract), a distinguishing characteristic of traditional and new forms of participation, influences
participants’ intention to participate in a variety of political activities. The findings reveal that young people are more likely to take part in new compared to traditional forms of
participation, and that engaging in low-cost activities is more common compared to high-cost activities. However, young people who are exposed to either concrete or abstract political content do not differ with regard to their intention to participate. This relationship is neither mediated through perceived personal relevance of the topic nor personal affectedness.
However, the effects of perceived personal relevance and personal affectedness depend on the parents’ socioeconomic status. In conclusion, these findings suggest that young people are not per se disengaged from politics, but search for new means of participation. To encourage youth participation, the author recommends continuing to promote new forms of involvement and emphasizing how easy it is to engage in low-cost activities.
Keywords: political participation, young people, traditional and new forms, low- and high-cost forms, SES, perceived personal relevance, personal affectedness, media effects, experiment, moderated mediation, (dis)engaged youth
According to Bennett (2008, p. 1), “younger generations have disconnected from conventional politics and government in alarming numbers”: Voting turnout among young people is declining in Western democracies (Bastedo, 2015), and young people’s
engagement in traditional forms of participation is decreasing (Eckstein, Noack, &
Gniewosz, 2013). Moreover, many young people feel that politicans marginalize them, and they perceive formal politics as strange and inaccessible (Cammaerts, Bruter, Banaji, Harrison, & Anstead, 2014). Not only is there a participation gap between young and old people, but also between individuals with different socioeconomic statuses (SES, Brady, Verba, & Lehman Schlozman, 1995; Henn & Foard, 2013). As one example, research has found that the low voter turnout among young people can mainly be explained by young people with a lower SES who do not vote (Bastedo, 2015). This is particularly worrying since participation in politics is not only a fundamental right (Cammaerts et al., 2014), but also crucial for the functioning of democracies. When only people with shared characteristics (e.g., of similar age or higher education) participate, the electorate is not adequately represented anymore. This, in the long-term, may undermine democracy’s legitimacy (Ødegård & Berglund, 2008). The political socialization of young people is crucial, as they will be the political decision-makers of the future. Therefore, the European Union explicitly stresses the goal to encourage young people to take part in the democratic life (Treaty of Lisbon of 2007).
However, there is a glimmer of hope: Young people seem to be more likely to participate in so-called new forms of participation, such as taking part in flash mobs or signing petitions (Bennett, 2008; Norris, 2003; Ødegård & Berglund, 2008). Moreover, young people want to express their opinions and are willing to participate, but the right means of participation are still missing (Gaiser, De Rijke, & Spannring, 2010). As new forms of political participation are on the rise (Bennett, 2008; Dalton, 2008), it is necessary to ask whether the associated changes in the nature and level of participation have an impact on the social inequality within the public sphere, and if modernized participation forms require
different civic skills and effort (Norris, 2007). However, there is no academic consensus on whether these forms contribute to closing or widening existing participation gaps. According to the mobilization hypothesis, new forms of participation in general and online participation, in particular, are likely to attract new groups to participate (Oser, Hooghe, & Marien, 2013). This expectation is based on the assumption that barriers to participate are reduced as
information is more easily accessible and different civic skills are required to get involved. In contrast, the reinforcement hypothesis stresses that new forms of participation and the internet might even enlarge existing participation gaps as new skills are required which may also vary by SES (Oser et al., 2013).
The purpose of this study is to answer the research question how the type of political content promoting participation affects young people’s intention to engage in different forms of participation. As such, it aims to enhance the understanding how youth can be motivated to participate politically (Bennett, 2008). The contribution of this study is twofold: The concepts of new and traditional forms of participation are narrowed down, and key distinguishing characteristics are established. Building on this, one decisive criterion, namely whether either one concrete or multiple abstract issues are addressed (Norris, 2003), is identified. An
experiment is conducted to assess how political content in an online environment, addressing either a single concrete or multiple abstract issues, affects young people’s intention to engage in different forms of participation.
Theoretical Background Engaged or Disengaged Youth
According to Henn and Weinstein (2006), numerous authors are concerned about young people’s disengagement with politics. At the same time, other authors argue that there is a shift towards new forms rather than a decline in participation in general (Gil de Zúñiga, Veenstra, Vraga, & Shah, 2010; Norris, 2007). This implies that the same empirical evidence can be evaluated differently depending on the underlying normative assumptions regarding
what good citizens ought to do. According to the disengaged youth paradigm, the decrease in general civic involvement is threatening democracy. Since the government is seen as the center of democracy, the often associated personalization is perceived as harmful for
democracy (Bennett, 2008). In contrast, the engaged youth paradigm acknowledges that there are considerable changes in the social identity of young people that lead to a focus on individualized, expressive political behavior (Bennett, 2008). Accordingly, young people are not disengaged per se, but take part in different, new forms of participation to influence political decision-making (Bastedo, 2015; Cammaerts et al., 2014). This results in an increase of protests and a diversification of new forms of participation (Dalton, 2008; Norris, 2003). But how do these new forms of engagement differ from the traditional forms? Despite the fact that differences in the normative assumptions lead to different interpretations of the state of young people´s civic engagement, there are also methodological issues. Latent forms of political participation such as expressing the opinion online are often neglected, and the term “civic engagement” is used in multiple different ways (Ekman & Amnå, 2012). Hence, there is a need for a new and definite concept of political participation.
Traditional and New Forms of Political Participation
Research on political engagement consistently distinguishes between two types of participation, which can broadly be categorized as traditional and new forms. They are, for instance, labeled as conventional versus unconventional participation (Eckstein, Noack, & Gniewosz, 2013), or as formal versus cause-related political activities (Ødegård &
Berglund, 2008). Some scholars even distinguish between two related normative concepts of citizenship, namely the dutiful citizen, who perceives voting as the most important act of participation, and the engaged citizen, who has a broader understanding of political participation and values active, self-expressive participation (Bennett, 2008; Dalton, 2008).
Traditional forms of political participation focus on the formal representation of the citizen’s voice and do not go beyond electoral politics (Cammaerts et al., 2014; Dalton, 2008; Soler-i-martí, 2017). They are facilitated through highly institutionalized organizations (Norris, 2003), which usually focus on a political agenda, namely a set of issues or policies they want to address (Ødegård & Berglund, 2008). Examples for traditional forms of political participation are voting, party membership, volunteering for or donating money to political parties.
The definition of new forms of participation is less straightforward, as they are based on a much broader definition of participation comprising “all activity that aims to influence politics” (Ødegård & Berglund, 2008, p. 594). Hence, extra-representational activism such as political consumption is typical for new forms of participation (Soler-i-martí, 2017).
Activities are less government-centered, and non-governmental organizations or companies may also be the target of a campaign (Bennett, 2008; Norris, 2003). Overall, new forms of participation can be described as individualized (Bennett, 2008; Bolzendahl, 2013; Ødegård & Berglund, 2008), focused on single issues (Norris, 2003; Ødegård & Berglund, 2008; Soler-i-martí, 2017), related to people’s everyday life (Norris, 2003; Ødegård & Berglund, 2008), and are associated with a lack of trust in traditional mass media (Bennett, 2008; Gil de Zúñiga et al., 2010). Such activities are mostly organized through informal and loose networks (Bennett, 2008; Ødegård & Berglund, 2008). Examples for new forms of political
participation are boycotting, donating money for a cause or joining online political forums. However, it is not always possible to categorize participation forms based on such static definitions. Thus, van Deth (2014) argues that an operational definition including a set of decision rules based on prior research can be applied to any form of political activity. Along these lines, an operational definition is more helpful to recognize distinct modes of participation. He identifies three forms of political participation: the minimalist or
motivational political participation (here: new forms). Criteria for the minimalist definition are a) that behavior is concerned, b) which is voluntary, c) by citizens and not professionals or politicians and d) most importantly that the activity is located in the sphere of politics (van Deth, 2014). These definitions were used to categorize the participation forms that are subject of this study (Table 1).
Political Agenda versus Single Issues
The most striking difference between traditional and new forms of participation is that the latter have a focus on one specific issue rather than a political agenda including multiple issues (Norris, 2003). Particularly young people have diverse sets of highly individualized political preferences and values (Albert, Hurrelmann, & Quenzel, 2015). Thus, new forms of participation which “leave people free to support those issues they feel strongly about, without explicitly supporting a group of problems that include things with which they
disagree or find uninteresting” (Ødegård & Berglund, 2008, p. 597) are particularly attractive to youths.
Moreover, a focus on single issues is also likely to attract young people, who often have a cause-oriented political interest (Soler-i-martí, 2017). Accordingly, there are two distinct dimensions of political interest: Cause-oriented interest is concerned with concrete political causes, while institutional interest resembles the classical political interest as reflected in traditional forms of participation.
Furthermore, single issues addressed by new forms of activism are typically very specific problems (Teney & Hanquinet, 2012). In contrast, the multiple issues or political agenda dealt with by traditional forms of participation are usually more abstract (Ødegård & Berglund, 2008). According to Bastedo, transforming even complex political issues of national importance “into something personal and concrete” (2015, p. 662) is crucial to reach out to young people. Examples for concrete issues that are mentioned by young are the response of a mayor to homelessness, prices for housing, as well as police harassment. In
comparison, examples for more abstract issues are international trade agreements, democratic reforms or online piracy legislation (Bastedo, 2015). However, to my knowledge, there are no examples of political content, more precisely political ads, portraying abstract or concrete issues that allow for the testing of differences in the effects of concrete and abstract political content on the intention to participate politically. Overall, I expect:
H1: Young people who are exposed to concrete political content addressing a single issue have a higher intention to participate compared to young people who are exposed to abstract political content addressing multiple issues.
H2: Young people have a higher intention to take part in new compared to traditional forms of political participation.
High- and Low-cost Forms of Political Participation
Forms of political participation can also be distinguished based on the cost of involvement (Gil de Zúñiga et al., 2010). Low-cost activities such as liking a post on social media require little commitment and are easy-to-implement, whereas high-cost forms of participation such as joining a political organization require more commitment, time, and sometimes even financial resources (Michelsen, Grunenberg, Mader, & Barth, 2015). Hence, high-cost forms of participation are not feasible for some people (Gil de Zúñiga et al., 2010), and depend on high levels of problem awareness (Michelsen et al., 2015). Thus, as reflected in the engagement pyramid, a framework depicting levels of citizens’ voluntary involvement, many people are engaged in low-cost activities: They observe, follow and endorse an
organisation or cause (Rosenblatt, 2010). On the contrary, fewer people are involved in high-cost engagement, such as ongoing contribution, owning or leading an organisation.
It should be noted that the requirements for time and moral commitment do not vary between traditional and new forms of participation per se but between different types of organizations (Gaiser et al., 2010). Accordingly, being part of traditional organizations or associations is more costly than being part of an informal group or taking part in temporary
activities. Moreover, some forms of participation are perceived to require higher levels of civic skills than others. Hence people’s likelihood to take part in easy, little challenging forms of participation are less related to their confidence in their political competencies (Eckstein et al., 2013). Overall, I expect that:
H3: Young people have a higher intention to take part in low-cost compared to high-cost forms of political participation.
Perceived Personal Relevance
Political communication influences political behavior; however, the proposed media effects are mostly indirect effects (Holbert & Stephenson, 2003). Hence, I argue that the influence of political content type on participatory intention is mediated through the perceived personal relevance of the respective issue (Van Reijmersdal, Rozendaal, Smink, & Van Noort, 2016): The perceived personal relevance of a message is high if the object or behaviour speaks to the receiver's interests and acknowledges a person’s individual living conditions (Van Reijmersdal et al., 2016) or is associated with personal life-style, value, and self-image (Celsi, Chow, Olson, & Walker, 1992). Thus, the concept of personal relevance reflects the core idea of the actualizing citizen model and new forms of participation (Bennett, 2008). Similar concepts are often applied with regard to behavioral intentions and pro-environmental actions in particular, and there is broad consensus that personal relevance positively affects cognitive processing, attitudes, and behavior (Kang, Liu, & Kim, 2013; Van Reijmersdal et al., 2016).
Personal relevance is not an objective concept, and the perception may differ per individual.1 According to Frewer and colleagues (1999), it is therefore important to assess how people perceive personal relevance, rather than only manipulating it. Moreover, the way in which messages are structured also matters for perceived personal relevance (Frewer, Howard, Hedderley, & Shepherd, 1999). Thus, it is crucial to explain to people that their
actions can have an impact by emphasizing the personal relevance through alluding to the individual’s lifestyle, values, and self-image (Kang et al., 2013).
Thus, I expect:
H4a: The perceived personal relevance of the problem positively mediates the relationship between the type of political content and the intention to participate.
Perceived Personal Affectedness
People who are personally affected by a problem are more like to participate
(Lingenberg, 2010), examples are dangers such as hazards or epidemics they may be exposed to (Frewer et al., 1999). Making concrete statements about the problem and its effects reduces uncertainty, which is often used as an excuse not to act on an issue (Spence, Poortinga, & Pidgeon, 2012). Thus, concrete political content is expected to increase the perceived personal affectedness and hence the likelihood to participate. This is particularly important with regard to issues with a temporal distance, namely problems whose effects are only noticeable in the future such as climate change, which are often perceived as very distant threats with little relevance to the everyday life (Scannell & Gifford, 2013).
Thus, I expect:
H4b: The perceived personal affectedness by a problem positively mediates the relationship between type of political content and intention to participate.
The Moderating Role of SES
The SES of the family, which is defined as the social standing as indicated by occupation, education, income, and wealth, influences the likelihood to engage in any form of political participation (Henn & Foard, 2013; Ødegård & Berglund, 2008; Oser et al., 2013; Rogers, Bultena, & Barb, 1975, Simonson, Vogel, & Tesch-Römer, 2014). However, the underlying mechanisms linking SES to participation are often not disentangled (Brady et al., 1995). Rogers and colleagues (1975) suggest that people with a higher SES are predisposed to participate because they are generally more interested in politics (selection model). However,
according to the civic voluntarism model by Brady and colleagues (1995), money and civic skills, resources which are crucial for political participation, differ between socioeconomic groups. These differences explain participation gaps (Norris, 2003), as higher educated parents are for instance more likely to encourage their children to engage in educational activities (Snellman, Silva, Frederick, & Putnam, 2014). Moreover, the family’s financial resources determine whether young people can afford to participate, and many young people with a lower SES have to make sure to meet their basic needs, and hence have no capacities to participate (Cammaerts et al., 2014).
Young people with lower SES often miss out on opportunities to establish contacts and friendships outside their communities, which is one possible explanation why they tend to act mostly on issues which directly affect themselves, such as topics regarding family, race, or religion (Sherrod, 2003). This finding is supported by Bastedo (2015), who argues that young people with a lower SES care about topics that are “concrete and [impact] them directly” (p. 660). On the contrary, young people with a higher SES are interested in a broader range of issues.
Thus, I expect:
H5a: Young people’s SES moderates the relationship between perceived personal relevance and the intention to participate.
H5b: Young people’s SES moderates the relationship between perceived personal affectedness and the intention to participate.
Additional Variables
I will control for the effects of prior participation and political self-efficacy (Eckstein et al., 2013; Lopes, Benton, & Cleaver, 2009).2 Moreover, I will control for the effects of standard socio-demographic factors such as age, gender, and political left-right self-placement. In addition, general interest in politics and topic-specific interest (Lopes et al.,
2009) will be considered, and it will be controlled for the effects of political activity (Bastedo, 2015).
The conceptual model (Figure 1) summarizes the outlined hypotheses.
Method Experimental Design
To test the hypotheses and to examine causal effects, a one-factorial between-subjects experimental design with three groups (two experimental groups and one control group) was conducted. The independent variable was the type of political content (concrete vs. abstract), and the dependent variable was the intention to participate in different forms of political participation. A moderator and several mediators were included as well (Figure 1).
Experimental Manipulation Stimuli.
The experimental stimuli consisted of three political ads which differed in their level of abstractness: A political ad informing about a specific environmental issues in a concrete manner (Bennett, 2008; Norris, 2003) with an everyday ethical component (Soler-i-martí, 2017), a political ad informing about a group of environmental issues in an abstract manner (Ødegård & Berglund, 2008; Soler-i-martí, 2017), and a control group with a political ad regarding medicine prices which was neither abstract not concrete (see Appendix B).
The concrete and the abstract ad dealt with environmental issues, because young people tend to be very interested in this topic, whereas the control stimulus dealt with healthcare policies, that is, an issue less relevant to young people (European Parliament, 2014). To make the stimuli as realistic and authentic as possible, they were designed to look like a Facebook ad. A social media example was chosen as the amount of time young people spend online is increasing (Albert et al., 2015) and German politicians increasingly use social media, particularly Facebook, to connect to citizens (Oelsner & Heimrich, 2015). Each stimulus consisted of a picture, a header (45–50 characters), and a brief description (33-37
words). Moreover, the two experimental groups included a call to action (“Get active!”). The stimuli were selected based on a pre-test.
Pre-test.
Four pairs of experimental stimuli consisting of a concrete and an abstract ad each were tested regarding the perceived level of abstractness. Topics of the ads were climate change or plastic waste in the oceans. Participants in each condition were exposed to four randomly selected concrete and abstract ads (Table 2).3 All ads that were discussing only one problem in a concrete manner were rated as significantly less abstract than the ads discussing several environmental issues in a more abstract manner (all p < .05). Hence, the manipulation was successful for all pairs. The mean differences were the biggest for pair 1 (MD = 2.17) and pair 4 (MD = 2.10, Table 3). However, pair 4 was judged as more credible (M = 5.65, SD = .021) than pair 1 (M = 5.28, SD = .24), which is why pair 4 was selected as the most suitable pair for the main study.
Sampling and Participant Characteristics
The target population of this study were German-speaking youths aged 18 to 24 years.4The goal of this study was to include participants with different characteristics regarding their level of education and the family’s SES. Participants were recruited through various organizations (e.g., unions’ youth departments, associations for voluntary work, citizens’ initiatives), which distributed the invitation to participate in the online survey through their networks (newsletter, online forums).5 In addition, participants were recruited through a school for vocational education; in that case, participants filled out the
questionnaire in class.
Overall, 302 people accessed the online survey, but only 209 participants completed it (43% female).6 7% of participants attended “Hauptschule” (ISCED-level 2), 21% attended high school, 24% attended a school for a vocational training, and 48% of people attended university. A majority of people indicated to be politically interested (M = 6.94, SD = 2.61,
10-point scale from 1 to 10, range = 1 - 10). Many participants described themselves as leaning to the political left (48.31 %) or centre-left (33.50 %), and only few people described themselves as centre-right or right (18.19 %). The randomization was successful; there were no significant differences between the three conditions with respect to gender, η2 = 0.94, p = .642; age, F (2,206) = 1.39, p = .251; education, F (2,206) = .36, p = .698; and political left-right self-placement, F (2,206) = .25, p = .782.
Procedure
The experiment was conducted over a three-week period between the 1st and the 21st of Mai 2017. Participants were randomly assigned to one of the three experimental conditions. At the beginning of the questionnaire, participants were informed about the purpose of the study, asked to provide informed consent and to give their age. All participants who were not aged 18-24 were excluded. Participants were then exposed to one of the three experimental political, social media ads. Subsequently, they were asked to answer the post-test questions. After debriefing, participants were thanked for taking part in this study.
Most participants completed an online questionnaire created with Qualtrics (05-2017) using either a desktop computer or laptop (43.5%) or a mobile device (33.5%). For practical reasons, participants who were recruited through a school for vocational education completed the questionnaire on paper in class (23.0%).7 As some of them had difficulties in reading and understanding some words, they were provided with an additional sheet with explanations (e.g., participation, petition).8
Measurement Mediators.
Perceived personal relevance was measured using five items on a 7-point Likert scale, with answer options ranging from strongly disagree to strongly agree (α = .91, see appendix A for the full questionnaire). This measurement was based on an existing perceived personal relevance scale (PPR; Kang et al. , 2013) and adjusted to the context of this study. On
average, participants rated the perceived personal relevance as neither high nor low (M = 4.07, SD = 1.54). Participants were also asked to indicate whether they expect to be personally affected by the problem described in the political ad in the future. This concept was measured using one item on a 7-point-Likert scale (M = 5.10, SD = 1.76).
Moderators.
The education of both parents according to the International Standard Classification of Education was used as a proxy for the SES of the family (ISCED; UNESCO Institute for Statistics, 2012). 9 People who have completed tertiary education, defined as academic education or advanced vocational training or professional education, were considered to be highly educated (UNESCO Institute for Statistics, 2012). A variable combining the mother’s and the father’s highest educational attainment was created (α = .72, M = 5.06, SD = 1.75, on a scale from 1 to 8) and recoded into a single variable with the two options lower and higher educated parents.
Dependent Variables.
The overall intention to participate was defined as participants’ self-reported
likelihood to engage in different forms of political participation in the future. Answer options ranged from extremely unlikely to extremely likely on a 7-point Likert scale. Intention to participate overall was measured based on 20 items suggesting different forms of political participation (e.g. signing a petition) which were described in prior research (e.g. Dalton, 2008; Ekman & Amnå, 2012; van Deth, 2014). The reliability of this scale was very high, but participatory intention was comparably low overall (α = .91, M = 3.66, SD = 1.18, range = 1-6.75; Figure 2).
The intention to engage in different forms of political participation was subdivided into traditional (7 items, α = .88) and new forms of participation (13 items, α = .82) based on theory (van Deth, 2014). A factor analysis using varimax rotation showed that all seven items that were expected to measure the intention to participate in traditional participation formed a
single uni-dimensional scale (eigenvalue = 3.38). A second factor analysis revealed that the 13 items that were expected to measure the intention to participate in new forms of political participation formed a single uni-dimensional scale (eigenvalue = 5.53). The results
suggested that both scales, the intention to participate in new (M = 3.78, SD = 1.24, range = 1-6.86) and in traditional forms (M = 3.43, SD = 1.28, range = 1 - 6.75), measured the concepts they intent to measure.
Moreover, participation was also subdivided into low- and high-cost forms based on theory (Gil de Zúñiga et al., 2010; Michelsen et al., 2015). Two factor analyses using varimax rotation showed that all items that were expected to measure the intention to participate in low-cost forms (eigenvalue = 4.94) and high-cost forms of participation (eigenvalue = 4.20) formed single uni-dimensional scales. The reliability for both the low-cost scale (11 items, α = .88, M = 4.15, SD = 1.31, range = 1- 6.89) and the high-cost scale was high (9 items, α = .85, M = 3.06, SD = 1.24, range = 1- 6.73).
For an overview, individuals’ willingness to participate in the different political activities per item was summarized (Figure 2). Participants were most likely to vote for a party caring about the cause and least likely to write to a newspaper or media outlet to support the cause.
Additional Variables.
Participants were asked to indicate as how abstract they perceived the political ad they saw on a 7-point scale (M = 3.26, SD = 1.45). Furthermore, the credibility of the political ad was measured (M = 4.86, SD = 1.49).
Political efficacy was measured using the Political Efficacy Short Scale (PEKS; Beierlein, Kemper, Kovaleva, & Rammstedt, 2014) on a 7-point Likert scale ranging from strongly disagree to strongly agree. The reliability for internal efficacy (2 items; α = .80, M = 5.14, SD = 1.59) and external political efficacy (2 items; α = .77, M = 3.57, SD = 1.34) was
high. Moreover, I controlled for gender, education, age, left-right self-placement, general political interest as well as interest in topic and general political activity (Table 4).
Results Analysis
The direct effects were tested using a series of one-way ANOVA’s and Bonferroni post-hoc tests. Paired-sample t-tests were conducted to test for differences in the intention to participate between traditional and new forms as well as low- and high-cost forms of political participation. Moreover, I tested for moderated mediation using The Process Macro for SPSS (Version 2.16), with 10,000 bootstrap samples and 95% bias-corrected bootstrap-confidence intervals. All models controlled for age, gender, left-right self-placement, political interest, issue-specific interest, political activity as well as internal and external political efficacy.
Manipulation Check
To test whether the manipulation was successful, a one-way ANOVA was conducted to compare the perceived level of abstractness between the experimental conditions. The analysis revealed that there were significant differences between the three groups, F (2, 206) = 39.99, p <. 001, η2 = .18. A Bonferroni post-hoc test indicated that participants in the concrete condition (M = 2.58, SD = 1.17) perceived the ad as significantly less abstract than participants in the abstract condition (M = 4.06, SD = 1.51, p < .001). Moreover, there were significant differences between the concrete and control group (M = 3.15, SD = 1.27, p = .033), and the abstract and control group (p < .001). Moreover, participants were asked to describe the topic of the political ad in one to three words. A summary (Figure 3) confirmed that there are undoubtedly differences in the interpretation of the two political ads.
Hypotheses testing
A one-way ANOVA was conducted to test whether the general intention to participate differed between the three experimental conditions (Table 5). There was a significant
effect size was medium (η2 = .07). A Bonferroni test revealed that the concrete and abstract experimental groups did not differ with regard to the intention to participate (M difference = -.02, p = 1.00), but there was a significant difference between the concrete experimental condition and the control group (M difference = .67, p = .002) and between the abstract experimental condition and the control group (M difference = .69, p = .002). Thus, the first hypothesis was not supported10.
A paired samples t-test was conducted to test whether young people had a higher intention to take part in new compared to traditional forms of political participation. The difference was significant, t (208) = -5.45, p < .001, 95% CI [-.47, -.22], indicating that young people were more likely to participate in new (M = 3.78, SD = 1.25) than in traditional forms of political participation (M = 3.43, SD = 1.28). Even though the effect size was small
(Cohens d = 0.38), this finding confirmed the second hypothesis.
To test whether young people were more likely to engage in low- compared to high-cost forms of political participation, a paired samples t-test was conducted. There was a significant difference, t (208) = 15.61, p <.001, 95% CI [.95, 1.22], indicating that participants were more likely to participate in low-cost activities (M = 4.15, SD = 1.31) than in high-cost forms of participation (M = 3.06, SD = 1.24), and the effect size was large (Cohen’s d = 1.00). Thus, the third hypothesis was supported.
A second stage moderated mediation analysis using PROCESS based on ordinary least squares path analysis was conducted. It was assumed that the relationship between the
proposed mediators, perceived personal relevance and personal affectedness, and the
dependent variable, intention to participate, depends on the moderator (b-path) (Hayes, 2013). There was no significant direct effect of the type of political content on the proposed
mediators personal relevance (b = .23, 95% CI [-.22, .65]) and personal affectedness (b = .29, 95% CI [-.22, .80]) or the dependent variable (b = -.20, 95% CI [-.55, .15], Table 6 and Figure 4 and Figure 5). Hence hypothesis 4a and 4b were not supported. The analysis further
revealed that the proposed moderator (parents’ education) positively affected the relationship between personal affectedness and the intention to participate (b = .38, 95% CI [.18, .59]). Thus, hypothesis 5b was supported. Moreover, parents’ education also moderated the relationship between perceived personal relevance and individuals’ participatory intention; however, this effect was only marginally significant and not in the expected direction (b = -.22, 95% CI [-.44, .00]. Thus, hypothesis 5a was not supported. It is noteworthy that the moderating role of education of the parents was different for the two proposed mediators. Overall, there was no significant moderated mediation for the proposed mediators perceived personal relevance (index = -.05, 95% CI [-.24, .02]) and personal affectedness (index = .11, 95% CI [-.05, .39]).11
Additional Analysis
To test whether the relationship between the type of political content and the proposed mediators was moderated by the education of the parents (a-path), and whether the direct relationship between type of political content and intention to participate was moderated by the education level of the parents, a second moderated mediation analysis was conducted (Hayes, 2013). The analysis revealed that the independent variable, type of political content, as well as the proposed moderator, the parents’ education level, did not significantlty predict the outcome of the proposed mediator or on the general intention to participate (Table 7 and Figure 6). However, the proposed mediators. personal affectedness (b = .13, 95% CI [.24, -.01]) and perceived personal relevance (b = .47, 95% CI [.34, .60]) did predict the
participatory intention. Overall, there were no significant moderated mediation effects for the two proposed mediators perceived personal relevance (index = .04, 95% CI [-.40, .46]) and personal affectedness (index = -.01, 95% CI [-.23, .10]), as visually displayed in Figure 7. The model with the best fit was model A (R2 = .62) compared to model B (R2 = .57), indicating that there was a second-stage rather than a first-stage moderated mediation.
Additionally, three independent samples t-test were conducted to test whether the intention to engage in different forms of political participation differed for participants whose parents were higher or lower educated. Young people with higher educated parents had a higher intention to engage in all forms of political activities (Figure 8), but the differences were not significant for participation overall, t (186) = -.85, p = .397, 95% CI [-.15, .17], intention to participate in traditional, t (186) = -.32, p = .750, 95% CI [-.44, 32], and new forms of participation, t (186) = -1.07, p = .287, 95% CI [-.55, .16].
Discussion
This study provides new insights into young people’s engagement in new and
traditional forms of political participation (Bennett, 2008; Dalton, 2008). Even though voting for a party remains the most common activity to exert political influence (see also Gaiser, et al., 2010; Ogris & Westphal, 2005), young people are overall more likely to participate in new political activities. This indicates that new participation forms supplement rather than replace traditional forms (Schäfer, 2010). This impression is reinforced by a relatively high
correlation between the intention to engage in both types of participation (Shehata, Ekström, & Olsson, 2015).
Furthermore, this study confirms that young people have a higher intention to
participate in low- compared to high-cost forms of political participation (Gil de Zúñiga et al., 2010). Since many new means of participation in general and political online activities, in particular, require less commitment, time and resources, the frequency with which youths are politically active may increase. Some critiques, however, argue that such forms have a significant disadvantage: As there is a lack of open-government structures and regulations specifying how and to what extent decision-makers have to take, for instance, the people’s will voiced through petitions into account, it is doubtful how effective these new forms of participation are. Hence it is necessary that citizens and governments “explore reforms to facilitate these new participation channels” (Dalton, 2008, p. 94).
Being open to the public can be seen as an essential characteristic of political
participation (Bennett, 2008). However, some activities such as political consumerism are not publicly recognized. As a consequence fragmentation of the public sphere increases (Hacker & Dijk, 2000). There is a trend that young people often share their views within groups of like-minded people only and do not discuss politics with other citizens. Thus, they miss the opportunity to understand the positions of others, which may impede collective decision- according to the ideal of the participatory democracy (Dalton, 2008).
I expected that young people who are exposed to concrete political content are more likely to participate in political activities compared to young people who are exposed to abstract political content (Bastedo, 2015). However, this study found no significant difference in the intention to participate between the two experimental groups. There are two alternative explanations for this non-finding: Firstly, the manipulation might not have been strong enough since participants who were exposed to abstract political content regarding several global environmental issues perceived it as neither concrete nor abstract. Thus, future work should focus on stimuli including multiple related issues from more than one policy field. Secondly, the level of the described problem, namely whether an issue is framed as local or global, is a better predictor than the degree of abstractness (Heyward & Roser, 2016).
According to previous research, many people are more likely to act on local and manageable environmental than on more distant issues (Hart & Nisbet, 2012). Hence, the non-findings may be attributable to the fact that both ads had a reference to environmental issues as a global problem and were thus too similar.
Despite the fact that there was no significant main effect of type of political content, the findings suggest that the proposed mediators, perceived personal relevance and perceived personal affectedness, indeed affect the intention to participate. Young people who view the problem as important for themselves are more likely to engage in any form of political participation (Frewer et al., 1999; Kang et al., 2013). Interestingly, young people who think
that they will be affected by the problem in the future are less likely to engage. A possible explanation is that some people who think they will be affected might believe that the current climate change is irreversible and that actions that are taken will not make any notable
difference anymore and hence give up and take no actions on the topic anymore. However, personal relevance has a positive and personal affectedness has a negative influence on the participatory intention. This supports the finding that those who are most affected by societal change are not the ones who are most active and voice their opinions (Dryzek, 2001). This means that the perceived personal relevance is crucial for participation and suggests that the two variables indeed measure two different concepts.
Furthermore, the results of the experiment show that the relationships between perceived personal relevance and the participatory intention as well as the relationship between personal affectedness and the participatory intention are dependent on the
parents’education level. Young people whose parents hold lower educational degrees are less likely to engage in politics, which supports findings from previous studies (e.g., Henn & Foard, 2013; Ødegård & Berglund, 2008; Oser et al., 2013; Rogers et al., 1975).
This study provides the first experimental test of the effect of a distinct content feature (i.e., the abstractness of political content) on traditional and new participation forms. One objective of the study was to overcome existing conceptual ambiguities between traditional and new, bottom-up forms of political participation (Dalton, 2008; Fox, 2014). In reliance on existing literature and van Deth's (2014) decision rules, it was possible to categorize different political activities as traditional or new forms of participation. Moreover, this research sheds new light on the explanatory power of perceived personal relevance as a mediator and compares its effect to the mediating effect of perceived personal affectedness. The results suggest that there are different effects for both mediators, which stress the need for future studies to incorporate both concepts in their designs. Overall, this study represents another
step in understanding the role political communication plays regarding the mobilization of young people and sheds light on participation patterns among this group.
Limitations and Suggestions for Future Research
Although this study provides valuable new insights into the political participation of young people, some limitations need to be considered. First, the effect of type of political content on intention to participate was only tested for environmental issues. The topic was chosen because many young people are interested in it (European Parliament, 2014).
However, for the very same reason, the issue itself might be perceived as being relevant to the personal life and more tangible than other problems (Heyward & Roser, 2016). The finding that participants who were exposed to political content regarding environmental issues reported higher levels of perceived personal relevance than participants in the control group supports this explanation. Hence, future studies should compare the effects of concrete versus abstract content across different policy fields.
Second, even though most of the scales used here were highly reliable, assessing a family’s SES by questioning young people is challenging. I aimed to use a composite
measure, but the newly created scale was not reliable. A possible explanation is that children (particularly in bigger cities) tend to live and learn together with people who share similar demographics (Schulte, Hartig, & Pietsch, 2014). Therefore, relative measures have little relevance compared to independent, more objective measures like parents’ highest
educational attainment. As education of adults (older than 25) is an “excellent proxy measure of SES” (Oakes, 2008, p. 50) and a better predictor compared to financial resources or employment status (Simonson et al., 2014) combined score of the educational attainment of both parents was used as the indicator for a family’s SES. Moreover, it should be noted that the measurement of perceived personal affectedness is potentially flawed because a single-item scale was used following previous studies (Frewer et al, 1999; Metag, 2016).
This study examined political content’s effects on young people’s intention to
participate. I acknowledge that there is an intention-behaviour gap (Gaiser et al., 2010), which means that people are more likely to indicate that they are willing to engage than to actually participate. Nevertheless, measuring the willingness to engage in different participation forms allows drawing conclusions regarding the relative attractiveness of various activities. Future research should aim at assessing actual behavioural outcomes, such as signing up for events or taking part in activities. Furthermore, this study measured the intention to participate
immediately after exposure to the stimuli; hence it is not possible to make statements about long-term effects. One option to do so would be to ask participants about their intention to participate again after a period of two weeks.
Moreover, self-reported measures or misreporting are the most important potential sources of bias: Since political participation is highly valued in society (Keeter, Zukin,
Andolina, & Jenkins, 2002), answers may have been biased by social desirability. When using self-reported measures, participants may also vary regarding their understanding of the
concepts under study. To avoid misinterpretations, I asked about specific activities and provided explanations if necessary. Overall, potential measurement problems are not likely to affect the results as the test-retest consistency is high for most political participation measures (Keeter et al., 2002).
Difficulties also aroused during the recruitment phase: Even though numerous
organizations were approached, only few agreed to support the research project. The resulting sample size of this study was sufficient to detect overall trends but small enough to make it difficult to find significant relationships. In addition, young people who are members of organizations may systematically differ from young people who are not part of any organizations. However, as there was great diversity regarding participants’ education, political attitudes, and family circumstances, the results of the study are sufficiently generalizable. It should be noted that findings are only generalizable to young people in
Germany, as differences in the national context regarding the social and political system and culture also lead to differences in patterns of participation (Gaiser et al., 2010). Hence, cross-country comparisons might provide valuable additional insights.
This study tested which characteristics make new forms of participation attractive for young people, and how young people react to political content designed to emphasize such features. It would be interesting for future studies to test for the effect of additional distinctive characeristics, such as representative versus extra-representative participation (Soler-i-martí, 2017), the level of individualization and relevance to people’s daily life (Bennett, 2012; Ødegård & Berglund, 2008), and whether activities are organized through informal and loose networks or institutionalized organizations (Bennett, 2008; Norris, 2003). Moreover, I
recommend including macro level variables in future research designs, as for instance social inequality is also known to be a strong predictor of participation patterns (Schäfer, 2010).
Conclusion
Young people are more likely to engage in low-cost than in high-cost forms of participation. Moreover, new forms are overall more attractive to them than traditional forms of political participation and, hence, are a promising way to (re)connect young people to politics. Young people are not per se disengaged from politics but look for new means of engagement and expression, which is in line with the engaged youth paradigm (Bennett, 2008; Norris, 2007) and may give cause to optimism. Moreover, the results support the mobilization hypothesis as the intention to participate in new forms of participation is higher for youth from all socio-economic backgrounds and the participation gap between those with lower and higher educated parents does not grow significantly bigger.
Communication professionals, politicians, and youth workers striving to motivate young people to take activley part in politics may equally learn from this study. As young people appear to be more attracted to new participation forms, I recommend to highlight such activities. Moreover, whenever there are low-cost activities with low entrance barriers such as
signing a petition or taking part in a boycott, this should be emphasized as well, for instance by mentioning that signing a petition requires less than a minute or that it is possible to take part in a flash mob without further commitment to an organization. These insights are especially useful to create strong calls for actions. However, changes in communication strategies are not enough to bridge the participation gap. I firmly believe that it is essential to increase the general social acceptance for new modes of participation among people of all ages and levels of education (Bennett, 2008), and to further establish pathways from new forms of participation to traditional politics.
References
Albert, M., Hurrelmann, K., & Quenzel, G. (2015). 17. Shell Jugendstudie. Jugend 2015 [17th Shell Youth Study.Youths 2015]. Hamburg, Germany: Fischer.
Bastedo, H. (2015). Not “one of us”: Understanding how non-engaged youth feel about Politics and Political Leadership. Journal of Youth Studies, 18(5), 649–665. https://doi.org/10.1080/13676261.2014.992309
Beierlein, C., Kemper, C. J., Kovaleva, A. & Rammstedt, B. (2014). Political Efficacy Kurzskala (PEKS) [Political Efficacy Short Scale]. In ZIS - GESIS Leibniz Institute for the Social Sciences (Eds.). Zusammenstellung sozialwissenschaftlicher Items und Skalen. http://dx.doi.org/10.6102/zis34
Bennett, W. L. (2008). Changing Citizenship in the Digital Age. Civic Life online: Learning how Digital Media can Engage Youth, 1, 1–24.
https://doi.org/10.1162/dmal.9780262524827.001
Bennett, W. L. (2012). The Personalization Of Politics: Political Identity, Social Media, and Changing Patterns of Participation. The Annals of the American Academy of Political and Social Science, 644(1), 20–39. https://doi.org/10.1177/0002716212451428
Bolzendahl, C. (2013). Are “Good” Citizens “Good” Participants ? Testing Citizenship Norms and Political Participation across 25 Nations. Political Studies, 61, 45-65.
https://doi.org/10.1111/1467-9248.12010
Bowling, A. (2005). Mode of Questionnaire Administration can have serious Effects on Data Quality. Journal of Public Health, 27(3), 281–291.
https://doi.org/10.1093/pubmed/fdi031
Brady, H. E., Verba, S., & Lehman Schlozman, K. L. (1995). Beyond Ses: A Resource Model of Political Participation. American Political Science Review, 89(2), 271–294.
https://doi.org/10.2307/2082425
Youth Apathy: Young Europeans’ Critical Attitudes Toward Democratic Life. American Behavioral Scientist, 58(5), 645–664.
https://doi.org/10.1177/0002764213515992
Celsi, R. L., Chow, S., Olson, J. C., & Walker, B. A. (1992). The Construct Validity of intrinsic Sources of personal Relevance: An intra-individual Source of felt Involvement. Journal of Business Research, 25(2), 165–185.
https://doi.org/10.1016/0148-2963(92)90015-4
Dalton, R. J. (2008). Citizenship Norms and the Expansion of political Participation. Political Studies, 56(1), 76–98. https://doi.org/10.1111/j.1467-9248.2007.00718.x
Deutsches Instituts für Vertrauen und Sicherheit im Internet (2014). DIVSI U25-Studie. Kinder, Jugendliche und junge Erwachsene in der digitalen Welt [Children,
Adolescents and young Adults in a Digital World] (Research Report). Retrieved from DIVSI website https://www.divsi.de/publikationen/studien/divsi-u25-studie-kinder-jugendliche-und-junge-erwachsene-in-der-digitalen-welt/
Dryzek, J. S. (2001). Legitimacy and economy in deliberative democracy.Political theory, 29(5), 651-669. https://doi.org/10.1177/0090591701029005003
Eckstein, K., Noack, P., & Gniewosz, B. (2013). Predictors of Intentions to participate in Politics and actual political Behaviors in young Adulthood. International Journal of Behavioral Development, 37(5), 428-435. https://doi.org/10.1177/0165025413486419 Ekman, J., & Amnå, E. (2012). Political Participation and Civic Engagement: Towards a new Typology. Human Affairs, 22(3), 283-300. https://doi.org/10.2478/s13374-012-0024-1 European Parliament, (2014). Flash Eurobarometer of the European Parliament. European
Youth in 2014. Analytical Synthesis (EP Publication No. EB 395) Retrieved from http://www.europarl.europa.eu/pdf/eurobarometre/2014/youth/eb_395_synthesis_yout h_en.pdf
participation in Britain: The decline and revival of civic culture. Parliamentary Affairs 67(2), 495–505. http://doi.org/10.1093/pa/gss094
Frewer, L. J., Howard, C., Hedderley, D., & Shepherd, R. (1999). Reactions to Information about Genetic Engineering: Impact of source Characteristics, perceived personal Relevance, and Persuasiveness. Public Understanding of Science, 8(1), 35–50. https://doi.org/10.1088/0963-6625/8/1/003
Gaiser, W., De Rijke, J., & Spannring, R. (2010). Youth and political participation. Empirical results for Germany within a European context. Young, 18(4), 427–450.
https://doi.org/10.1177/110330881001800404
Gil de Zúñiga, H., Veenstra, A., Vraga, E., & Shah, D. (2010). Digital Democracy:
Reimagining Pathways to political Participation. Journal of Information Technology & Politics, 7(1), 36–51. https://doi.org/10.1080/19331680903316742
Hacker, K.L., van Dijk, J. (2001). Digital Democracy. Issues of Theory and Practice. London, United Kingdom: SAGE Publications.
Hart, P. S., & Nisbet, E. C. (2012). Boomerang Effects in Science Communication. Communication Research, 39(6), 701–723.
https://doi.org/10.1177/0093650211416646
Hayes, A. F. (2013). Introduction to mediation, moderation, and conditional process analysis: A regression-based approach. New York, NY: Guilford Press.
Hayes, A. F. (2016). The Process Macro for SPSS (Version 2.16) [Computer software]. Retrieved from http://www.processmacro.org
Henn, M., & Foard, N. (2013). Social Differentiation in young people’s political Participation: the Impact of social and educational Factors on youth political Engagement in Britain. Journal of Youth Studies, 17(3), 1–21.
https://doi.org/10.1080/13676261.2013.830704
young people vote? Policy & Politics, 34(3), 517–534. http://dx.doi.org/10.1332/030557306777695316
Heyward, C. & Roser, D. (2016). Climate Justice in a Non-Ideal World. Oxford, United Kingdom: Oxford University Press.
Kang, J., Liu, C., & Kim, S. H. (2013). Environmentally sustainable Textile and Apparel Consumption: The role of Consumer Knowledge, perceived consumer Effectiveness and perceived personal Relevance. International Journal of Consumer Studies, 37(4), 442–452. https://doi.org/10.1111/ijcs.12013
Keeter, S., Zukin, C., Andolina, M., & Jenkins, K. (2002, April). Improving the Measurement of Political Participation. Paper presented at the Annual Meeting of the Midwest Political Science Association, Chicago, IL.
Lingenberg, S. (2010). Europäische Publikumsöffentlichkeiten. Ein pragmatischer Ansatz [European Public Spheres. A pragmatic Approach.]. Wiesbaden, Germany: Verlag für Sozialwissenschaften.
Lopes, J., Benton, T., & Cleaver. E. (2009). Young people's intended Civic and Political Participation: Does Education matter? Journal of Youth Studies, 12(1), 1-20. http://dx.doi.org/10.1080/13676260802191920
Metag, J. (2016). Political communication and Opinion formation in Germany: A comparative study of local and national issues. European Journal of Communication, 31(2), 185-203. http://dx.doi.org/10.1177/0267323116629876
Michelsen, G., Grunenberg, H., Mader, C., & Barth, M. (2015). Greenpeace
Nachhaltigkeitsbarometer 2015. Nachhaltigkeit bewegt die Jugend [Sustainability is Moving the Younger Generation: Greenpeace Sustainability Barometer 2015]. (Research Report) Bad Homburg, Germany: VAS.
Norris, P. (2003, November). Young People & Political Activism. From the Politics of Loyalties to the Politics of Choice. Report for Council of Europe Symposium,
Strasbourg, France.
Norris, P. (2007). Political activism: New challenges, new opportunities. In C. Boix & S.C. Stokes (Eds.), The Oxford Handbook of comparative Politics (628-652).
http://doi.org/10.1093/oxfordhb/9780199566020.001.0001 Oakes, J. (2008). Measuring Socioeconmic Status. Retrieved from
http://www.esourceresearch.org/Portals/0/Uploads/Documents/Public/Oakes_FullCha pter.pdf
Ødegård, G., & Berglund, F. (2008). Political Participation in late Modernity among Norwegian Youth: an individual Choice or a Statement of Social Class? Journal of Youth Studies, 11(6), 593–610. https://doi.org/10.1080/13676260802247573 Oelsner, K., & Heimrich, L. (2015). Social Media Use of German Politicians: Towards
Dialogic Voter Relations? German Politics, 24(4), 451-468. http://dx.doi.org/10.1080/09644008.2015.1021790
Ogris, G., Westphal, S. (2005). Political Participation of Young People in Europe – Development of Indicators for Comparative Research in the European Union (EUYOUPART). (Research Report WP 10). Retrieved from
http://www.sora.at/fileadmin/images/content/Pages/euyoupart_ergebnisse_finalcompa rativereport.pdf
Oser, J., Hooghe, M., & Marien, S. (2013). Is Online Participation Distinct from Offline Participation? A Latent Class Analysis of Participation Types and Their Stratification. Political Research Quarterly, 66(1), 91-101.
https://doi.org/10.1177/1065912912436695
Qualtrics (2017). Qualtrics (05-2017) [Computer software]. Available from http://www.qualtrics.com
Rogers, D. L., Bultena, G. L., & Barb, K. H. (1975). Voluntary Association Membership and Political Participation: an Exploration of the Mobilization Hypothesis. Sociological
Quarterly, 16(3), 305–318. https://doi.org/10.1111/j.1533-8525.1975.tb00950.x Rosenblatt, G. (2010, February 1). The Engagement Pyramid: Six Levels of Connecting
People and Social Change. Retrieved from http://groundwire.org
Scannell, L., & Gifford, R. (2013). Personally Relevant Climate Change: The Role of Place Attachment and Local Versus Global Message Framing in Engagement. Environment and Behavior, 45(1), 60–85. https://doi.org/10.1177/0013916511421196
Schäfer, A. (2010). Die Folgen sozialer Ungleicheit für die Demokratie in Westeuropa [Effects of Social Inequality on Democracies in Western Europe]. Zeitschrift für vergleichende Politikwissenschaft, 4(1), 131-156. http://dx.doi.org/10.1007/s12286-010-0075-9
Schulte, K., Hartig, J., & Pietsch, M. (2014). Der Sozialindex für Hamburger Schulen [The Social Index for Schools in Hamburg]. In D. Fickermann & N. Maritzen (Eds.), Grundlagen für eine daten- und theoriegestützte Schulentwicklung. Konzeption und Anspruch des Hamburger Instituts für Bildungsmonitoring und Qualitätsentwicklung (IfBQ) (67-80). Münster, Germany: Waxmann.
Shehata, A., Ekström, M., & Olsson, T. (2015). Developing Self-Actualizing and Dutiful Citizens : Testing the AC-DC Model Using Panel Data Among Adolescents. Communication Research, 43(8), 1141-1169.
https://doi.org/10.1177/0093650215619988
Sherrod, L. R. (2003). Promoting the Development of Citizenship in diverse Youth. Political Science and Politics, 36(2), 287–292. https://doi.org/10.1017/S1049096503002233 Simonson, J., Vogel, C., & Tesch-Römer, C.(Eds.) (2016). Freiwilliges Engagement in
Deutschland: der Deutsche Freiwilligensurvey 2014 [Voluntary Engagement in Germany: German Survey of Volunteers 2014]. Berlin, Germany: Springer. http://dx.doi.org/10.1007/978-3-658-12644-5
Social Mobility and Extracurricular Participation among American Youth. The ANNALS of the American Academy of Political and Social Science, 657(1), 194–207. https://doi.org/10.1177/0002716214548398
Soler-i-martí, R. (2017). Youth political Involvement Update: Measuring the Role of cause-oriented political Interest in young People’s activism. Journal of Youth Studies, 18(3), 396–416. https://doi.org/10.1080/13676261.2014.963538
Spence, A., Poortinga, W., & Pidgeon, N. (2012). The Psychological Distance of Climate Change. Risk Analysis, 32(6), 957–972.
https://doi.org/10.1111/j.1539-6924.2011.01695.x
Teney, C., & Hanquinet, L. (2012). High Political Participation, high Social Capital? A relational analysis of youth social capital and political participation. Social Science Research, 41(5), 1213–1226. https://doi.org/10.1016/j.ssresearch.2012.03.012 Treaty of Lisbon amending the Treaty on European Union and the Treaty establishing the
European Community, C 306/01.(2007).
UNESCO Institute for Statistics (2012). International Standard Classification of Education ISCED 2011 (Research Report). Retrieved from
http://www.uis.unesco.org/Education/DOcuments/isced-2011-en.pdf
Van Deth, J. W. (2014). A conceptual Map of political Participation. Acta Politica, 49(3), 349–367. https://doi.org/10.1057/ap.2014.6
Van Reijmersdal, E. A., Rozendaal, E., Smink, N., & Van Noort, G. (2016). Processes and Effects of targeted Online Advertising among Children. International Journal of Advertising, 36(3), 396-414. https://doi.org/10.1080/02650487.2016.1196904
Table 1
Classification of Different Forms of Participation (Type and Cost of Participation) Form of participation
Type of participation
Cost of participation
Vote for a party Traditional Low
Convince someone to vote for a party Traditional Low Attend events/ discussions organized by
parties Traditional Low
Be active in (youth) party Traditional High
Donate to a (youth) party Traditional High
Become a member of a (youth) party Traditional High
Contact a politician Traditional High
Sign a petition New Low
Take part in flash mob/ creative protest New Low
Boycott New Low
Take part in demonstration New Low
Join a group in an online political forum New Low
Buycott New Low
Wear a political symbol New Low
Attend events/ discussions organized by
political organization (other than parties) New Low
Start your own project New High
Be active in political organization (other
than parties) New High
Write to newspaper/ media outlet New High
Work voluntary in organization New High
Donate money to organization (other than
Table 2
Overview: Stimuli per Experimental Group
Pair Exp. Group 1 Exp. Group 2
Pair 1 Concrete Abstract
Pair 2 Abstract Concrete
Pair 3 Concrete Abstract
Table 3
Results of t-tests and Descriptive Statistics on Dependent Variable Abstractness per Pair by Type of Political Ad (Pretest)
Group Concrete Ad Abstract Ad Pair M SD n M SD n t df p Pair 1 2.15 .99 20 4.32 1.29 19 -5.90 37 >.001 Pair 2 2.47 1.43 19 3.60 1.54 20 -2.37 37 .023 Pair 3 2.35 1.46 20 3.95 1.72 19 -3.14 37 .003 Pair 4 1.95 1.08 19 4.05 1.73 20 -4.52 32 >.001
Table 4
Descriptives of Control Variables
Variable M SD N Age 21 1.91 209 Left-right self-placement 3.87 1.82 209 Interest in politics 6.94 2.61 209 Interest in environmental policy 3.62 3.00 209
Interest in health policy 3.14 2.53 209
Political activity 5.54 2.73 209
Note. Age was measured in years. Political left-right self-placement was
measured on a 10-point scale from left to right. Interest in politics and political activity were also measured on 10-point-scales. Interest in
Table 5
Mean and Standard Deviation of Intention to Participate per Experimental Group
Exp. Group Mean SD
Group 1: Concrete political content 3.86 1.09 Group 2: Abstract political content 3.89 1.11 Group 3: Control 3.19 1.23
Table 6
Coefficients for Model A: Conditional indirect effect of Type of Content on General Intention to participate through personal Relevance and Personal Affectedness and the Conditional Effect of Type of Content on general Intention to Participate (second-stage moderation)
MA Personal Relevance MB Personal Affectedness Y Intention to participate
Predictor B SE p B SE p B SE p Antecedent X Type of content .23 .23 .313 .29 .26 .258 -.20 .18 .269 MA Personal Relevance .61 .09 < .001 MB Personal Affectedness -.31 .07 < .001 W Education of parents 1.31 .51 .011* int_1 MA x V -.22 .11 .052 int_2 MB x V .38 .11 < .001 *** int_3 X x V .30 .28 .280 Political activity .11 .06 .077 .29 .26 .258 .04 .04 .282 Political interest .01 .08 .860 -.02 .09 .851 -.01 .05 .845 Left-right self-placement -.05 .07 .527 -.04 .08 .661 -.08 .04 .083
Interest health policy -.05 .09 .597 .08 .10 .458 -.03 .05 .593
Interest environmental policy .39 .08 < .001 *** .39 .10 < .001 *** .09 .06 .121 Internal Efficacy -.10 .11 .327 .22 .12 .066 .06 .07 .329 External Efficacy .05 .09 .571 -.06 .10 .581 .02 .05 .769 Gender .75 .25 .003** .33 .28 .293 -.05 .15 .736 Age .06 .06 .302 .03 .07 .700 -.06 .03 .060 Constant .89 .97 .361 1.13 1.11 .311 2.77 .60 < .001 *** R² = .40 R² = .32 R² = .62 F(10, 117) = 7.63, p < .001 F(10, 117) = 5.53, p < .001 F(16, 111) = 11.06, p < .001 Note. Two mediators were included in the model: personal relevance and personal affectedness. Coefficients are unstandardized coefficients.
95% BCa CI = 95% bias-corrected accelerated confidence interval. Political content was coded as 0 = concrete, 1= abstract. *= p < .05, **= p < .01, ***= p < .001.
Table 7
Coefficients for Model B: Conditional Indirect Effect of Type of Content on General Intention to Participate through Personal Relevance and
. Personal Affectedness and the Conditional Effect of Type of Content on General Intention to Participate (first-stage moderation)
MA Personal Relevance MB Personal Affectedness Y Intention to participate
Predictor B SE p B SE p B SE p Antecedent X Type of content .19 .30 .520 .25 .34 .481 -.05 .14 .744 MA Personal Relevance .47 .05 <.001*** MB Personal Affectedness -.13 .06 .030* W Education of parents -.65 .30 .036* -.46 .35 .577 .21 .20 .281 X x W .09 .46 .852 .11 .53 .829 .38 .29 .118 Political activity .10 .06 .103 .05 .07 .535 .05 .04 .178 Political interest .03 .08 .738 -.01 .09 .937 .01 .05 .954 Left-right self-placement -.06 .07 .376 -.05 .08 .571 -.07 .05 .142
Interest health policy -.07 .09 .444 .06 .10 .540 -.05 .06 .404
Interest environmental policy .392 .08 <.001*** .41 .10 <.001*** .09 .06 .115 Internal Efficacy -.08 .10 .438 .24 .12 .053 .03 .07 .629 External Efficacy .06 .09 .482 -.05 .10 .631 .02 .05 .689 Gender .76 .24 .002* .34 .28 .228 -.03 .16 .839 Age .05 .06 .378 .02 .07 .783 -.07 .04 .045* Constant 1.04 .96 .280 1.23 1.11 .269 2.37 .60 < .001*** R² = .43, R² = .34 R² = .57 F(12, 115) = 7.25, p < .001 F(12, 115) = 4.83, p < .001 F(14,113) = 10.60
Note. Coefficients are unstandardized coefficients. Political content was coded as 0 = concrete, 1= abstract. 95% BCa CI = 95% bias-corrected accelerated
Figure 1. Conceptual Model of the Conditional Indirect Effect of Political Content on Intention to Participate through Perceived Personal Relevance and Perceived Personal Affectedness (second-stage moderation model).
