The Development of Political Interest Among Adolescents: A Communication Mediation Approach Using Five Waves of Panel Data

Abstract

Political interest is one of the most important individual-level predictors of news media use, public opinion formation, and engagement. When, how, and why some citizens develop a strong interest in politics is, however, less clear. This study analyzes the development of political interest during the formative years of adolescence, using a five-wave panel study among Swedish adolescents, covering a period of 4 years. Based on the citizen communication mediation model, we analyze how interest in political and current affairs news among family and friends influence adolescents’ political interest. Taken together, while the findings lend support for several of the hypotheses, mechanisms, and processes derived from the communication mediation model, parents appear more important than peers when it comes to shaping adolescents’ political interest.

Keywords

political interest socialization communication mediation adolescents news media use

The development and origins of citizens’ interest in politics has received growing research attention lately (Neundorf, Smets, & García-Albacete, 2013; Prior, 2010)—for reasons not difficult to imagine. Decades of studies indicate that political interest is one of the most important individual-level predictors of public opinion formation and engagement. Citizens who are politically interested are more likely to follow current affairs news and engage in motivated information processing (Prior, 2007; Zaller, 1992), display higher levels of political awareness and knowledge (Delli Carpini & Keeter, 1996; Luskin, 1990), and more inclined to participate in various civic and political activities (van Deth, Montero, & Westholm, 2007; Verba, Schlozman, & Brady, 1995). Put simply, political interest matters.

When, how, and why some citizens develop a strong interest in politics, while others do not, is, however, less clear. In one of the most extensive longitudinal analyses, Markus Prior (2010) recently concluded that aggregate and individual-level political interest remains highly stable over time among the adult population:

In order to figure out why political interest is higher among some people than others, it is necessary to understand how it forms in the first place. The stability of interest even among people in their twenties indicates that this formation happens quickly. (Prior, 2010, p. 765)

Therefore, to understand the origins of political interest, we need to focus on the formative years of adolescence and early adulthood (see also Neundorf et al., 2013).

Against this background, the present study analyzes the development of political interest during these formative years. By following adolescents over a period of 4 years, we are able to provide a comprehensive analysis of how family and friends influence the development of political interest. Theoretically, our approach builds on the communication mediation model (CMM) developed by Lee, Shah, and McLeod (2013, see also McLeod & Shah, 2009), highlighting “the centrality of communicative phenomena in the home, at school, among peers, and through media in the development of young citizens’ democratic competencies and motivations for engagement” (Lee et al., 2013, p. 670). More specifically, we focus on how interest in political and current affairs news among family and friends influence the development of political interest: Do adolescents with parents and peers who value following current affairs news and staying informed about political matters develop a stronger interest in politics? Thus, our analysis does not only consider the effects of classic socializing agents in the development of political interest, but focuses particularly on the role of communication and media practices.

The article is organized in the following way. In the next section, we elaborate more thoroughly on the concept of political interest and discuss research on its origins and development. This section is followed by a review of our main argument related to the role of political and current affairs news interest among family and friends, which forms the basis of our hypotheses. Our multiwave panel data are then presented along with measurements of key concepts. After presenting findings in the “Results” section, we close the article by discussing the main theoretical and societal implications of the study.

Political Interest: Definitions and Origins

At some point, all studies on democracy and political behavior touch upon issues related to the quantity, quality, or equality of political participation (Macedo et al., 2005). For several decades, political interest has been one of the most frequently used concepts in studies regarding various aspects of citizen behavior. It has not only been applied to analyses at both the individual (Sears, 1981, 1983) and country (Aalberg & Curran, 2012; Teorell, Torcal, & Montero, 2007) level. It has been at the center in studies on adults (Dalton & Shin, 2014), adolescents (Flanagan, 2013; Garcia-Albacete, 2013), and children alike (van Deth, Abendschön, & Vollmar, 2011).

Notwithstanding its unparalleled extensive application as well as its overwhelmingly manifested explanatory power at various fields of political and communication research, it may seem surprising that the concept of political interest has been lacking a more precise definition. The main idea behind the concept still focuses on individuals’ general self-rating (Lazarsfeld et al., 1948) of a specific content, “the degree to which politics arouses a citizen’s curiosity” (van Deth, 1990, p. 278). In that sense, political interest can be considered as an intrinsic motivation to pay attention to and engage in politics—not because of external pressures or as a means to achieve distinct outcomes, but for personal pleasure and the inherent satisfaction of doing so in itself (Ryan & Deci, 2000; Shani, 2009). As defined by Hidi and Renninger (2006), “[i]nterest as a motivational variable refers to the psychological state of engaging or the predisposition to reengage with particular classes of objects, events, or ideas over time” (p. 112). This definition also captures the common distinction among psychologists between situational interest and individual (or dispositional) interest. While situational interest denotes a personal-level state, characterized by momentarily increased attention as a reaction to specific environmental stimuli, individual interest “refers to a person’s relatively enduring predisposition to reengage particular content over time” (Hidi & Renninger, 2006, p. 113). Research on political interest has primarily focused on the latter dimension of the concept (Prior, 2010; Robison, 2017; Shani, 2009), which is also the case in the present study.

While research into political communication has produced numerous studies on the explanatory power of political interest, fewer studies have aimed at explaining political interest (Neundorf et al., 2013; Prior, 2010; Shani, 2009). Within political socialization research, this surely is a contested issue, especially when it comes to the when-question. Some have argued that political attitudes are open for change reflecting various general, personal, or generational experiences throughout lifetime (Jennings & Markus, 1984). Others claim that they form during a few impressionable years, although there is disagreement on when this occurs: during adolescence (Krosnick & Alwin, 1989) or later in early adulthood (Neundorf et al., 2013; Niemi & Hepburn, 1995). However, in his seminal study of more than 10 panel surveys, Prior (2010) found a high stability of political interest already laid down in the 20s. It led him to recommend greater attention to developmental processes during childhood and adolescence.

Adolescence is typically considered a distinct and unique life cycle phase in the socialization literature. During these years, important changes occur that may influence citizens’ political interest both directly and indirectly. With developing cognitive skills and expanding social experiences that comes with age, young citizens’ perceptions of politics as something relevant to their lives are assumed to crystalize—gradually strengthening their interest in politics as they grow older (Neundorf et al., 2013; Russo & Stattin, 2017; van Deth, 1990). Furthermore, adolescence is a phase characterized by significant changes in the relative importance of parents and peers as socialization agents. While the time adolescents spend with family members decrease substantially during these years, this is “paralleled by changes in adolescents’ relationship with peers, signaling the increasing importance of peers as a context for adolescents socialization” (Smetana, Robinson, & Rote, 2016, p. 64; see also Eder & Nenga, 2006). Based on these well-documented general observations, Dostie-Goulet (2009) argued that peers’ relative influence on adolescents’ political interest should increase as “children progressively disengage from family” (p. 407). Following the developmental processes and changing peer relations characterizing adolescence, we expect a general growth in political interest over time during adolescence to be observable in two ways:

Hypothesis 1 (H1): Adolescents’ interest in politics increases over time as they grow older.

Hypothesis 2 (H2): Adolescents’ frequency of political discussions increases over time (H2a), but peers become relatively more frequent discussion partners compared with parents (H2b).

Apart from these developmental processes taking place during adolescence, however, it remains unclear to what extent social influences can leave a lasting imprint on adolescents’ general interest in politics. If individual political interest resembles a latent trait—partly influenced by a genetic component (Bouchard, 2004; Funk, 2013)—waiting to bloom, interest in politics may be rather insensitive to characteristics of the social environment. On the contrary, psychologists interested in the formation of individual interest in general have argued that interest gradually develops into a predisposition only in the presence of conducive environmental support, including social network characteristics and values held by significant others among family and peers (Hidi & Renninger, 2006; Ryan & Deci, 2000).

Social Contexts, News Norms, and the Development of Political Interest

Research on political socialization typically relies on social learning theory to explain how attitudes, values, habits, and behavior transmit from family and peer groups to citizens (De Vreese & Möller, 2014; Jennings, Stoker, & Bowers, 2009). By observing the behavior of significant others, children gradually learn what is considered appropriate and socially rewarded, and what is not in different contexts (Bandura, 2009; Bukowski, Brendgen, & Vitaro, 2007). Similarly, cultural reproduction (cultural capital) theory of socialization has been used to explain how certain values and habits are reproduced following differential access to cultural resources, communication styles, and in-group norms that guide appropriate behavior (Eder & Nenga, 2006; Shani, 2009).

The idea that norms and communicative practices may influence adolescents’ development of political interest is based on the CMM of socialization (Lee et al., 2013; McLeod & Shah, 2009), which puts particular emphasis on how media and communicative practices at home and within social networks influence political socialization (see also the Deliberative Learning Model; McDewitt & Kiousis, 2006). Thus, apart from emphasizing the importance of norms, values, and habits among family and peers—following the more traditional accounts of socialization—the CMM explicitly specifies how agents influence political engagement through certain communicative practices (De Vreese & Möller, 2014; McLeod, 2000). When it comes to political engagement more generally, the CMM postulates that “parental, peer, and schooling factors influence civic engagement both directly and indirectly by spurring other communication behaviors” (Lee et al., 2013, p. 675). Both political discussions and informational media use are considered particularly important for understanding adolescents’ political engagement and communication competencies more broadly. So far, however, the CMM has only been used to explain behavioral dimensions of political engagement, such as participation. Whether norms and cultural practices among family and peers can leave a lasting imprint on adolescents’ individual political interest, that is, their intrinsic rather than extrinsic motivation to engage in politics, is far from clear.

We argue, however, that political interest develops in relation to communicative practices formed by social norms and behaviors within adolescents’ family and peer social networks. In particular, Lee et al. (2013) emphasized “surveillance norms” among peers as a critical factor behind adolescents own informational media use—which in turn shapes further political discussion and civic engagement (see also Gotlieb, Kyoung, Gabay, Riddle, & Shah, 2015). Surveillance norms refer to the extent to which individuals’ value staying informed about current affairs and following the news. However, in order to avoid confusion related to connotations instilled by the term surveillance, we prefer to use “political and current affairs news interest” as the key concept. In particular, they argue that “[y]oung people who interact with peers who value knowledge and discussion of public affairs content are likely to be encouraged to consume and reflect on news content” (Lee et al., 2013, p. 673) themselves. Thus, based on social learning theory in general, and the CMM in particular, we expect that adolescents with parents and peers who value political and current affairs knowledge will develop a stronger interest in politics over time, by increasing (1) adolescents’ political discussions with parents and peers and by (2) increasing adolescents’ own news media use.

There are basically two distinct arguments linking parents’ and peers’ current affairs news interest to adolescents’ political interest. First, the home and peer environment can be seen as providing learning opportunities for adolescents. By being surrounded by people who value staying informed about political and current affairs on a daily basis, young citizens are more frequently exposed to these topics and, thereby, have greater opportunities to encounter, experience, and explore public issues, political arguments, and differences. By learning more and becoming more politically aware, they become better able to understand the relevance of politics to their own life—a factor related to the development of interest (Jennings et al., 2009; Silvia, 2008; van Deth, 1990). Second, an element of social pressure may also account for the relationship. Adolescents may internalize values, norms, and interests from significant others for social identity reasons and feelings of belongingness (Hidi & Renninger, 2006; Ryan & Deci, 2000). By learning, through experience, what is socially rewarded and punished by people in their immediate environment, adolescents will gradually associate various activities with positive or negative feelings. Robison (2017) explicitly discusses how social rewards influence intrinsic political interest, arguing that “[w]hen an individual comes to associate engagement with positively evaluated goals, then they will likely form a positive attitude towards, or interest in, performing the activity in the future” (p. 4). Taken together, we expect political and current affairs news interest among parents and peers to influence the development of political interest among adolescents:

Hypothesis 3 (H3): Parents’ interest in political and current affairs news has a positive effect on the development of political interest among adolescents.

Hypothesis 4 (H4): Peers’ interest in political and current affairs news has a positive effect on the development of political interest among adolescents.

Irrespective of the precise mechanisms underlying the influence of parents and peers on political interest (learning opportunities or social pressure), the CMM suggests that certain processes generate these effects. In particular, we first expect a relationship between parents’ and peers’ news interest and the frequency of interpersonal communication concerning politics and current affairs issues:

Hypothesis 5 (H5): Strong interest in political and current affairs news among parents and peers will increase adolescents’ frequency of political discussions with parents (H5a) and peers (H5b).

In a second step, we expect these interpersonal encounters with politics and current affairs through parents and peers to encourage adolescents’ own news media use and promote further interest in politics. This expectation follows social learning theory in general and the CMM in particular. By regularly discussing politics and current affairs, parents and peers increase the salience of this topic, signaling its importance within the social networks to which adolescents belong—creating both learning opportunities and conditions for social pressure (Bukowski et al., 2007; Harris, 1995; Jennings et al., 2009; Lee et al., 2013). On a related third point, research also suggests that one motivation to monitor and follow the news media is anticipation of future political discussions (Boomgaarden, 2014; Eveland, Hayes, Shah, & Kwak, 2005). In that sense, adolescents who are part of social networks that value knowledge about politics and current affairs, and who interact with parents and peers on such matters, may prepare for future discussions by increasing their own news media use. Therefore, we pose the following two hypotheses:

Hypothesis 6 (H6): Adolescents who frequently discuss political and current affairs issues with parents (H6a) and peers (H6b) will increase their own news media use.

Hypothesis 7 (H7): Adolescents who frequently discuss political and current affairs issues with parents (H7a) and peers (H7b) will develop a stronger interest in politics.

In a final step, we also believe that part of the effect on political interest is mediated by increases in adolescents’ own news media use. Political interest and news media use are closely intertwined, most likely resembling a reciprocal causal relationship (Boulianne, 2011; Kruikemeier & Shehata, 2017; Möller & De Vreese, 2015). Therefore, growing interest should increase adolescents’ own news media use. By exposing adolescents to societal information, perspectives, and arguments, however, we expect news media use to increase interest further (Strömbäck & Shehata, 2010). Our goal here is not to disentangle these reciprocal effects but rather to test whether increases in adolescents’ news media use are related to increases in political interest.

Hypothesis 8 (H8): Adolescents who increase their own news media use will develop a stronger interest in politics.

Finally, given the particular circumstances that characterize adolescence, we hypothesize that peer norms will have a stronger effect on the development of political interest than parents’ norms. As noted above, the relative significance of family as a socialization force diminishes during adolescence while peer relations gain importance (Eder & Nenga, 2006; Smetana et al., 2016). This general trend should matter for influences on adolescents’ political interest as well (Dostie-Goulet, 2009):

Hypothesis 9 (H9): Among adolescents, peer news norms have a stronger effect on the development of political interest than parents’ news norms.

In sum, Figure 1 displays our set of hypothesized relationships on how parents’ and peers’ interest in political and current affairs news influences adolescents’ development of political interest, as outlined above. Most importantly, we expect that norms and practices within family and peer networks promote adolescents’ interest through a multistep process involving both interpersonal communication and adoption of news media habits. In the present study, we will test this model using five waves of panel data collected among adolescents. These data are described more extensively in the next section.

Figure 1.

A communication mediation model of political interest (hypothesized effects).

Research Design, Data, and Measures

The present study uses panel data from a large-scale research project on political socialization conducted in Sweden. With a school-based sampling strategy and a controlled questionnaire administration, the study has been able to maintain a very high response rate across panel waves.

For the purpose of this study, we rely on data from five panel waves, conducted annually between 2010 and 2014. The study is based on a sample of adolescents residing in Örebro, a middle-sized city of approximately 130,000 inhabitants. The city represents the national population well in terms of unemployment rate, income level, population density, and political leanings. The sample used for the present study consists of adolescents aged 13 to 14 years at the time of the first panel wave. The schools included were strategically selected in order to represent both theoretical and vocational programs; schools in the city center, in suburbs, and in rural areas; and schools in neighborhoods with differences in terms of social, economic, and ethnic background characteristics. In total, 10 lower secondary schools (högstadium) were selected at the outset of the fieldwork. All students within the specific age group at these schools were invited to participate in the study. The questionnaires were distributed and administered by trained research assistants, and filled out by students during scheduled school hours. By using such an approach, the study managed to secure exceptionally high response rates across panel waves. An overview of sample sizes and response rates throughout the fieldwork is provided in Table 1 (45% of all respondents participated in all five waves of the panel).

Table 1.

Sample and Participation Rate for Each Panel Wave.

Year	Cohort 1 (13-14 years old)
Year	Sample	Participation rate
2010	960	94% (N = 904)
2011	987	89% (N = 883)
2012	954	88% (N = 843)
2013	934	77% (N = 723)
2014	935	74% (N = 695)

Measures

Our main dependent variable is political interest, which was measured identically at each wave of the panel based on the following two survey items: (1) “How interested are you in politics?” and (2) “How interested are you in what is happening in society?” with response categories ranging from 1 (not at all interested) to 5 (very interested). These items were recoded to range between 0 and 1 before averaged into a single political interest scale (Pearson’s r = .46, at Wave 1).

The independent variables of primary interest here are parents’ and peers’ interest in political and current affairs news. This concept was measured using a battery of both positively and negatively framed agree–disagree statements: (1) My parents [friends] are interested in what happens around the world, (2) My parents [friends] follow the news, (3) My parents [friends] do not care much about what happens around the world, and (4) My parents [friends] are not particularly interested in politics or societal issues. The parents’ interest in political and current affairs news measure is based on all four items, with Statements 3 and 4 reversed before averaged into an additive index ranging from 0 to 1 (Cronbach’s α = .64, Wave 1). The peers’ interest in political and current affairs news was created similarly based on the first two items only, due to significantly higher reliability of the two-item (Cronbach’s α = .71/Pearson’s r = .56, Wave 1) compared the four-item scale (Cronbach’s α = .48, Wave 1). To assess the robustness of our findings, all analyses were conducted using both the two-item and the four-item scale of peer news interest.¹

With respect to mediating variables, we measured frequency of political discussion with parents and friends separately using two items focusing on the frequency of discussions about (1) what you have heard on the news about what has happened in Sweden or around the world, and (2) politics and society, with response categories ranging from 1 (very often) to 4 (never). The scales were inverted and recoded to range between 0 and 1 before added into a parent (Wave 1 Pearson’s r = .54) and peer (Wave 1 Pearson’s r = .46) political discussion index. Adolescents’ own news media use was measured by asking respondents how often they follow the news: (1) by reading printed daily newspapers, (2) by listening to radio news, (3) by watching television news, and (4) by using the Internet to follow the news—with response categories ranging from 1 (at least 5 days a week) to 5 (never). The scales were inverted and recoded to range between 0 (minimum value) and 1 (maximum value) before added into a total news media use index (Cronbach’s α = .69, Wave 1). See Appendix for complete list of mean values and standard deviations for each variable.

Data Analysis

As we have data on all key independent and dependent variables from each wave of the panel study, two different panel analytic approaches are used to test our hypotheses.

The first test relies on two-way fixed effects panel models (Allison, 2009; Finkel, 2008). These models account for unobserved (unmeasured) heterogeneity between units by relying on within-person variation over time only. Thus, we are able to test whether within-person changes in parent and peer news interest are related to within-person changes in political interest (as well as the mediation variables)—thereby using “each individual as his or her own control” (Allison, 2009, p. 1). By also including time dummy variables representing panel waves, these models further account for the general growth trend as well as period-specific factors that influence adolescents’ political interest equally at any given time point. Although these models have been characterized as providing “the most stringent causal test possible outside of an experimental setting” (Dilliplane, Goldman, & Mutz, 2013, p. 241), it must be noted that they do not account for time-variant factors that may change over time—and therefore are no panacea to the problem of causal inference.

The second test is based on autoregressive panel models estimated using structural equation modeling (Acock, 2013; Finkel, 2008). In a first step, we test whether parents’ and peers’ interest in current affairs news measured in Wave 1 predict adolescents’ political interest in subsequent waves, controlling for their initial level of political interest through a lagged dependent variable. Second, a comprehensive test of the full mediation processes implied by the CMM is conducted by including political discussions and adolescents’ news media use as mediator variables. By doing so, this autoregressive model captures the extent to which parent and peer news interest influence changes in political interest through over time changes in the mediators. To better account for random measurement error in the dependent variable, all structural equation models specify political interest as a latent variable—using the two items presented above as indicators (Acock, 2013; Finkel, 2008). The details of these models are provided in the “Results” section.

Apart from these two main modeling strategies, the sensitivity of our findings was assessed using alternative operationalizations of the dependent variable (one-item indicator) as well as instrumental variable approaches (Arellano-Bond dynamic panel models). Results from these analyses are reported in the “Results” section.

Results

In this section, we present our findings related to the influence of parents’ and peers’ political and current affairs news interest on adolescents’ political interest in several steps. Before testing the hypotheses more thoroughly using both fixed effects and structural equation panel models, some basic descriptive data are provided to illustrate how adolescents’ political interest and interpersonal communication with parents and peers develop over time.

Looking first at the main dependent variable, Figure 2 displays how our two indicators of political interest develop during the 4-year period. The two lines represent the share of respondents who report they were either “very” or “somewhat” interested in “politics” (Black) as well as “what happens in society” (Gray), respectively. Overall the trends lend strong support to H1, indicating that political interest increases over time as adolescents grow older. Even though there are clear differences between the two items—significantly more respondents report an interest in society than in politics—the growth pattern is very similar in both cases. It is also worth noting that the correlation between the two items gradually increases over time, from r = .46 at Wave 1 to r = .69 in Wave 5, suggesting a growing perceived connection between “politics” and “society” over time.

Figure 2.

Levels and developments of political and societal interest (%).

Given the changes taking place during adolescence, we also expected a growing political interest to be manifested in more frequent political discussions with family and friends (H2a), but that peers become relatively more frequent discussion partners compared with parents over time (H2b). Again, this is precisely what we see in Figure 3, which shows how interpersonal communication develops over time. When it comes to the frequency of political conversations, parents appear more important than friends, and this is particularly the case when adolescents are young (parents = 0.45 and peers = 0.32 at Wave 1). As time goes by, however, friends become increasingly important and the gap between parents and peers as conversation partners narrows significantly (parents = 0.50 and peers = 0.45 at Wave 5).²

Figure 3.

Levels and developments of political discussion (mean values).

Against this background, we return to the main research question of this study: Do adolescents with parents and peers who value following current affairs news and staying informed about political matters develop a stronger interest in politics?

The Influence of Parents and Peers: A First Test

Our first test of the influence of parents’ and peers’ on adolescents’ political interest relies on a two-way fixed effects panel model, focusing on within-person variation over time. That is, are increases in parents’ and peers’ news norms related to increases in adolescents’ own political interest—above and beyond the general growth trend? Table 2 presents the findings, and Model 1 lends support to H3 and H4 regarding the positive effect of both parents’ (b = 0.20, p < .001) and peers’ (b = 0.21, p < .001) news norms on adolescents political interest. Thus, when focusing solely on within-person changes over time—controlling for all stable adolescent-level characteristics—the findings indicate that when news interest increases among family and friends, so does political interest among adolescents.

Table 2.

Two-Way Fixed Effects Panel Models Predicting Change in Political Interest.

	Model 1	Model 2	Model 3
Current affairs news interest
Parents’ news interest	.20*** (.03)	.14*** (.03)	.12*** (.03)
Peers’ news interest	.21*** (.02)	.10*** (.02)	.08*** (.02)
Mediation variables
Political talk with parents	—	.17*** (.02)	.14*** (.02)
Political talk with peers	—	.21*** (.02)	.19*** (.02)
News media use	—	—	.19*** (.03)
Time
Wave 2	.01 (.01)	.00 (.01)	.00 (.01)
Wave 3	00 (.01)	–.01 (.01)	–.01 (.01)
Wave 4	.05*** (.01)	.03** (.01)	.03*** (.01)
Wave 5	0.08*** (.01)	.05*** (.01)	.06*** (.01)
R² (within)	.15	.24	.26
N (observations/individuals)	3,877/1,015	3,856/1,014	3,845/1,013

Note. Unstandardized b values with unit-level clustered standard errors in parentheses.

p < .05. **p < .01. ***p < .001.

Models 2 and 3 introduces our two main mediating variables: interpersonal political discussions and news media use. A few things are worth highlighting from these models. First, the coefficients representing interpersonal communication from Model 2 clearly show that within-person changes in discussion with both parents (b = 0.17, p < .001) and peers (b = 0.21, p < .001) are related to within-person changes in political interest over time in the directions predicted by H7a and H7b. Second, this is also the case with respect to adolescents’ own news media use, as shown in Model 3 (b = 0.19, p < .001), lending support to H8. Third, when these mediating variables are introduced, the overall effects of both parents’ (b = 0.12, p < .001) and peers’ (b = 0.08, p < .001) news norms are reduced substantially. This suggests that the impact of these variables may be channeled (mediated) through both interpersonal discussions and adolescents’ news media use. A more explicit test of this assumption is provided in the next section.³

The Influence of Parents and Peers: A Second Test

In a final test of our hypotheses, we turn to the results from a series of autoregressive structural equation models predicting how parents’ and peers’ interest in news measured in Wave 1 influence changes in adolescents’ political interest over several years. Before analyzing the hypothesized mediation processes, Table 3 provides an initial test of the basic argument by showing how Wave 1 parental and peer news interest influence adolescents’ political interest 1, 2, 3, and 4 years later—controlling for their initial political interest. As can be seen, while parents’ news interest is positively related to adolescents’ political interest 1 year (b = 0.12, p < .001), 3 years (b = 0.18, p < .001) and 4 years (b = 0.11, p < .05) later, there are no statistically significant effects of peers. These findings clearly indicate that parents are more important when it comes to leaving a more long-term imprint on adolescents’ interest in politics.⁴

Table 3.

Autoregressive Panel Models Predicting Political Interest at Differing Time-Lags (Standardized Coefficients).

	1-year lag (W2)	2-year lag (W3)	3-year lag (W4)	4-year lag (W5)
Parents’ news interest (W1)	.12*** (.03)	.07 (.05)	.18*** (.05)	.11* (.05)
Peers’ news interest (W1)	.03 (.05)	.04 (.06)	.03 (.06)	–.10 (.06)
Political interest (W1)	.66*** (.06)	.47*** (.04)	.37*** (.04)	.56*** (.04)
n	675	652	573	560
Degrees of freedom	3	3	3	3
χ² (p value)	4.73 (p = .19)	1.88 (p = .60)	3.68 (p = .30)	1.03 (p = .79)
RMSEA	.03	.00	.02	.00
CFI	.998	1.00	.99	1.00

Note. Standardized b values with standard errors in parentheses. All models estimated using structural equation modeling, with political interest as a latent variable. Correlations between Wave 1 measures allowed, as well as between each measurement indicator over time (political interest). RMSEA = root mean square error of approximation; CFI = comparative fit index.

p < .05. **p < .01. ***p < .001.

In a final test of the mediation processes suggested by the CMM, we estimate an autoregressive structural equation model based on four waves of panel data. The main endogenous variable here is political interest measured in Wave 4, which is regressed on its lagged value from Wave 1 as well as our main independent variables from Wave 1 (parents’ and peers’ news interest) and the mediators from Wave 2 (interpersonal political discussion with parents and peers) and Wave 3 (adolescents news media use). Each equation in the model is autoregressive—with lagged values of the endogenous variables included as controls—yielding a model focusing on changes in all endogenous variables over time. In that sense, we are able to test whether news interest among parents and peers are related to changes in political interest, mediated by changes in interpersonal communication and news media use. The details of the path model are provided in the figure note.

To begin with, it is worth noting that the estimated “total” effect of parents’ (b = 0.20, p < .001) and peers’ (b = 0.03, p > .05) political and current affairs news interest on Wave 4 political interest corresponds to the results presented in Table 3.⁵ Furthermore, the findings presented in Figure 4 also reveal that there are distinguishable direct effects of parent (b = 0.11, p < .05) but not peer (b = −0.02, p > .05) news interest on changes in political interest. Regarding H5, we expected that strong news interest among parents and peers would increase the frequency of political discussions with parents (H5a) and peers (H5b). This expectation is supported only with respect to parents (b = 0.27, p < .001) but not in relation to peers (b = 0.07, p > .05). Furthermore, H6 predicted that adolescents who frequently discuss political and current affairs issues with parents (H6a) and peers (H6b) would increase their own news media use. This hypothesis is, however, fully supported. Talking about politics with parents has positive effects on changes in news consumption (b = 0.20, p < .001), and a similar effect is found for interpersonal communication with peers (b = 0.10, p < .05). In addition, H7—predicting that adolescents who frequently discuss political and current affairs issues with parents (H7a) and peers (H7b) would develop a stronger interest in politics—is also supported with respect to parents (b = 0.24, p < .001), but not peers (b = 0.08, p > .05). Finally, adolescents’ own news consumption is a positive predictor of political interest (b = 0.27, p < .001), lending support to H8. In terms of differential effects of parents and peers, not only do the total effects point in favor of parents but also the estimated indirect effects—thereby giving no support to H9 regarding the expected stronger influence of peers.⁶

Figure 4.

Path models predicting changes in political interest (standardized path coefficients).

Conclusion and Discussion

This study has focused on the development of political interest among adolescents. A long history of research into political behavior (Campbell, 2013; Delli Carpini & Keeter, 1996), media use (Boulianne, 2011; Strömbäck, Djerf-Pierre, & Shehata, 2013), media effects and public opinion formation (Zaller, 1992) has demonstrated the importance of this personal-level motivation in relation to a range of outcomes. As noted by Prior (2010), “political interest is typically the most powerful predictor of political behaviors that make democracy work. Politically interested people are more knowledgeable about politics, more likely to vote, and more likely to participate in politics in other ways” (p. 747). In addition, some argue that political interest may become increasingly important as a predictor of current affairs news consumption and awareness following the gradual transformation from a low-choice to a high-choice media environment (Aalberg & Curran, 2012; Prior, 2007; Strömbäck et al., 2013).

Therefore, understanding the origins and development of political interest is important (Neundorf et al., 2013; van Deth, 1990). The present study has aimed at making a contribution in this regard. More specifically, we have relied on five waves of annually gathered panel data among Swedish adolescents to study how interest in political and current affairs news within family and among peers influences young people’s political interest. Theoretically, the study builds on insights from political socialization research in general, and the CMM in particular, in order to recognize the crucial interdependencies between human agency and social context (Emler, 2015; Huckfeldt, 2014; Lee et al., 2013)—as well as by focusing particularly on the role of media and communication practices. More specifically, we asked whether political and current affairs news interest among parents and peers—putting value in following current affairs and staying informed about societal matters—was related to adolescents’ political interest.

In sum, the findings from our five-wave panel survey lent support to several of the hypotheses. To begin with, general levels of political interest grew significantly over time as adolescents got older. This was reflected not only in higher reported levels of political interest across panel waves (H1) but also in more frequent political discussions over time (H2a). Also in line with the socialization literature, the frequency of political discussions with peers increased significantly relative to parent discussions during adolescence (H2b)—a time when young people strive for autonomy and gradually spend more time with friends. With respect to the impact of parents and peers on adolescents’ political interest, however, the findings clearly indicate that parents are more influential, thereby lending no support to H9. Although the fixed effects models showed clear effects of both parents’ and peers’ interest in current affairs news on changes in adolescents’ political interest over time, the autoregressive panel models supported more lasting (long-term) effects from parental influences. While effects of both parents’ and peers’ news interest were distinguishable in the short run, only the parent effects lasted over several years. These differences were also evident in the final model, explicitly testing the mediation processes suggested by the CMM. Not only did parents’ news interest have a direct effect on changes in adolescents’ political interest over several years, but part of the total effect was also mediated by increases in political discussions with parents (H5a) as well as increases in adolescents’ own news media use (H6a). There was, however, no significant total effect of peers’ news interest, and support for the overall mediation process was significantly weaker with respect to peers—although support for single paths was found. Taken together, these findings have several implications and contribute to the literature in a number of ways.

To begin with, compared with most other socialization studies analyzing various aspects of political participation, focusing on political interest as an intrinsic motivation is important in itself. While participatory behaviors may be subject to clear social influences and pressures through mechanisms of external motivation, it is much less obvious whether such factors can also leave a lasting imprint on a deeply held intrinsic motivation such as political interest. If individual interest resembles a personality trait, as suggested by developmental psychologists (Hidi & Renninger, 2006; Ryan & Deci, 2000), and has a significant genetic component (Bouchard, 2004; Funk, 2013), social norms among significant others may have limited influence on adolescents’ development of political interest. The findings presented here thus extend research on the CMM by focusing on nonbehavioral dimensions political engagement, revealing its value not only when it comes to understanding the social origins of political participation but also with respect to intrinsic motivations to engage (and reengage) in politics—which may be even more significant in a long-term perspective.

Furthermore, the findings also suggest that parents’ political and current affairs news interest is more important than peers’ interest when it comes to the development of political interest during adolescence. This was a somewhat unexpected finding, running counter to our hypothesis derived from previous research. The socialization literature has repeatedly noted how peers gradually gain importance these years, while less time is spent with family (Eder & Nenga, 2006; Smetana et al., 2016). While the results support this general development, showing that friends become relatively more important as political conversations partners over time, most discussions about political and societal matters still take place with parents. This may be due to the fact that the family “represents a relatively safe domain for teenagers to try out opinions” (McDewitt & Kiousis, 2006, p. 252). More importantly, the effects of parents on adolescents’ political interest also appear stronger and more consistent in the analyses presented here. There are at least three potential explanations to these findings. First, although the socialization literature emphasizes that peer groups are more influential at this age as adolescents struggle with identity issues and feelings of belongingness while simultaneously seeking greater autonomy from parents (Bukowski et al., 2007; Harris, 1995), it may well be the case that politics plays a minor and marginal role in such identity-seeking. What really matter to adolescents in relation to peers are other things in life and other domains of interest. Thus, friends are important, although not when it comes to political matters. Second, the seemingly weaker influence of peers may be related to the distinction between political interest as an intrinsic motivation, on one hand, and actual political behavior, on the other hand. If group pressures are important during adolescence, such influences may have greater impact on political activities through external motivations, than on intrinsic motivations to engage in politics. Finally, it may also be the case that the impact of peers on adolescents’ interest in politics is more immediate and transient than parental influences. While parents remain, best friends and peer groups can change over time during adolescence. As a consequence, it may be unlikely to find lasting effects of peer networks on young citizens’ political interest based on time-lags of several years. This interpretation is actually in line with the findings presented here, where peer effects appeared only in the short run. From this perspective, rather than dismissing peer influence as irrelevant in comparison with parents, the effects of family and friends may be qualitatively different, operating through different time frames.

Finally, while this study has enabled a comprehensive longitudinal analysis of adolescents’ political interest over several years, the study is not without limitations. Most importantly, we have relied on respondents’ self-reports of their parents’ and peers’ current affairs news interest—rather than independent measures from these people. As such, the measures used here focus on adolescents’ perceptions of their parents’ and peers’ interest in political and current affairs news, how changes in such perceptions are related to changes in their own political interest, and how these perceptions predict interest several years later. From a methodological point of view, it could be argued that these measures are not strictly exogenous, and future research should therefore try to incorporate independent measures from parents and friends. On the contrary, it could also be argued that it is precisely adolescents’ perceptions of their parents and peers that should matter. Only when citizens are aware of—and experience—the interests and habits of people in their immediate surroundings, can they exert a real influence. Furthermore, while this study utilized longitudinal panel data, suggesting that norms and interests among family and friends are related to the development of political interest among adolescents, these analyses do not solve the question of causality. Endogeneity cannot be fully accounted for in research designs like these. The fact that the adolescents in our study are nested in a variety of contexts such as schools, classes, and social networks—that are also changing over time in complex ways—raises statistical challenges difficult to completely account for with the methods applied here. Therefore, research on the everyday interplay between adolescents’, parents’, and friends’ interests, norms, and habits would gain from other research designs involving both in-depth qualitative approaches, in-depth network analyses and experiments with strictly exogenous treatments.

Footnotes

Appendix

Mean Values and Standard Deviations of Each Measure.

	Wave 1	Wave 2	Wave 3	Wave 4	Wave 5
Political interest	.45 (.23)	.47 (.25)	.46 (.25)	.53 (.25)	.56 (.24)
Parents’ news interest	.67 (.18)	.69 (.17)	.70 (17)	.72 (.17)	.75 (.17)
Peers’ news interest	.45 (23)	.48 (.23)	.49 (.21)	.53 (.21)	.56 (.21)
Political talk with parents	.45 (.25)	.48 (.25)	.48 (.24)	.52 (.24)	.50 (.24)
Political talk with peers	.32 (.22)	.35 (.23)	.37 (.22)	.42 (.23)	.45 (.23)
News media use	.39 (.21)	.41 (.21)	.39 (.22)	.39 (.21)	.40 (.20)
n (minimum)	882	847	803	706	677

Note. Mean values with standard deviations in parentheses.

Authors’ Note

This study was made possible by access to data from the Political Socialization Program, a longitudinal research program at YeS (Youth & Society) at Örebro University, Sweden. Professors Erik Amnå, Mats Ekström, Margaret Kerr, and Håkan Stattin were responsible for the planning, implementation, and financing of the collection of data. The authors also want to thank the three anonymous reviewers as well as Per Oleskog Tryggvason, Silvia Russo and Maria Solevid for helpful comments on earlier versions of this manuscript.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: The data collection and the study were supported by grants from Riksbankens Jubileumsfond.

Notes

Author Biographies

Adam Shehata is Associate Professor, Department of Journalism, Media and Communication, University of Gothenburg. His research focuses on media use and effects.

Erik Amnå, PhD, is Professor in Political Science at Örebro University, Sweden. His research deals mainly with democratic impact of civil society and civic education as well as with young people’s development of civic and political engagement over time.

References

Aalberg

Curran

(2012). How media inform democracy: A comparative approach. London, England: Routledge.

Acock

(2013). Discovering structural equation modeling using tata. College Station, TX: Stata Press.

Allison

P. D.

(2009). Fixed effects regression models. Los Angeles, CA: SAGE.

Bandura

(2009). Social cognitive theory of mass communication. In Bryand

Oliver

M.-B.

(Eds.), Media effects: Advances in theory and research (pp. 94-124). New York, NY: Routledge.

Baum

(2006). An introduction to modern econometrics using stata. College Station, TX: Stata Press.

Boomgaarden

H. G.

(2014). Interpersonal and mass mediated political communication. In Reinemann

(Ed.), Political communication. Handbook of communication sciences (pp. 469-488). Berlin, Germany: De Gruyter Mouton.

Bouchard

T. J.

(2004). Genetic influence on human psychological traits: A survey. Current Directions in Psychological Science, 13, 148-151. doi:10.1111/j.0963-7214.2004.00295.x

Boulianne

(2011). Stimulating or reinforcing political interest: Using panel data to examine reciprocal effects between news media and political interest. Political Communication, 28, 147-162. doi:10.1080/10584609.2010.540305

Bukowski

W. M.

Brendgen

Vitaro

(2007). Peers and socialization: Effects on exernalizing and internalizing problems. In Grusec

Hastings

(Eds.), Handbook of socalization (pp. 355-381). New York, NY: Guilford Press.

10.

Campbell

D. E.

(2013). Social networks and political participation. Annual Review of Political Science, 16, 33-48. doi:10.1146/annurev-polisci-033011-201728

11.

Dalton

R. J.

Shin

D. C.

(2014). Reassessing the civic culture model. In Dalton

Welzel

(Eds.), The civic culture transformed (pp. 91-115). New York, NY: Cambridge University Press.

12.

De Vreese

Möller

(2014). Communication and political socialization. In Reinemann

(Ed.), Political communication (pp. 529-546). Berlin: De Gruyter.

13.

Delli Carpini

M. X.

Keeter

. (1996). What Americans know about politics and why it matters. New Haven, CT: Yale University Press.

14.

Dilliplane

Goldman

S. K.

Mutz

D. C.

(2013). Televised exposure to politics: New measures for a fragmented media environment. American Journal of Political Science, 57, 236–248. doi:10.1111/j.1540-5907.2012.00600.x

15.

Dostie-Goulet

(2009). Social networks and the development of political interest. Journal of Youth Studies, 12, 405-421. doi:10.1080/13676260902866512

16.

Eder

Nenga

S. K.

(2006). Socialization in adolescence. In Delamater

(Ed.), Handbook of social psychology (pp. 157-182). New York, NY: Kluwer Academic.

17.

Emler

(2015). Explaining political participation: Integrating levels of analysis. In Barrett

Zani

(Eds.), Political and civic engagement: Multidisciplinary perspectives (pp. 146-161). Hove, UK: Routledge.

18.

Eveland

Hayes

Shah

Kwak

(2005). Understanding the relationship between communication and political knowledge: A model-comparison approach using panel data. Political Communication, 22, 423-446. doi:10.1080/10584600500311345

19.

Finkel

(2008). Linear panel analysis. In Menard

(Ed.), Handbook of longitudinal research: Design measurement and analysis (pp. 475-504). Burlington, MA: Academic Press.

20.

Flanagan

(2013). Teenage citizens: The political theories of the young. Cambridge, MA: Harvard University Press.

21.

Funk

(2013). Genetic foundations of political behavior. In Huddy

Sears

Levy

(Eds.), The Oxford handbook of political psychology (pp. 237-261). New York, NY: Oxford University Press.

22.

Garcia-Albacete

(2013). Promoting political interest in school: The role of civic education. In Abenschön

(Ed.), Growing into politics: Contexts and timing of political socialisation (pp. 91-113). Oxon, UK: European Consortium for Political Research Press.

23.

Gotlieb

Kyoung

Gabay

Riddle

Shah

(2015). Socialization of lifestyle and conventional politics among early and late adolescents. Journal of Applied Developmental Psychology, 60-70. doi:10.1016/j.appdev.2015.06.004

24.

Harris

J. R.

(1995). Where is the child’s environment? A group socialization theory of development. Psychological Review, 102, 458-489. doi:10.1037/0033-295X.102.3.458

25.

Hidi

Renninger

K. A.

(2006). The four-phase model of interest development. Educational Psychologist, 41, 111-127. doi:10.1207/s15326985ep4102_4

26.

Huckfeldt

(2014). Networks, contexts, and the combinatorial dynamics of democratic politics. Political Psychology, 35, 43-68. doi:10.1111/pops.12161

27.

Jennings

M. K.

Markus

G. B.

(1984). Partisan orientations over the long haul: Results from the three-wave political socialization panel study. American Political Science Review, 78, 1000-1018. doi:10.2307/1955804

28.

Jennings

M. K.

Stoker

Bowers

(2009). Politics across generations: Family transmission reexamined. The Journal of Politics, 71, 782-799. doi:10.1017/S0022381609090719

29.

Krosnick

J. A.

Alwin

D. F.

(1989). Aging and susceptibility to attitude change. Journal of Personality and Social Psychology, 57, 416-425.

30.

Kruikemeier

Shehata

(2017). News media use and political engagement among adolescents: An analysis of virtuous circles using panel data. Political Communication, 34, 221-242. doi:10.1080/10584609.2016.1174760

31.

Lazarsfeld

Berelson

Gaudet

(1948). The people’s choice: how the voter makes up his mind in a presidential campaign (2nd ed.) New York, NY: Columbia University Press.

32.

Lee

N.-J.

Shah

McLeod

J. M.

(2013). Processes of political socialization: A communication mediation approach to youth civic engagement. Communication Research, 40, 669-697. doi:10.1177/0093650212436712

33.

Luskin

R. C.

(1990). Explaining political sophistication. Political Behavior, 12, 331-361. doi:10.1007/BF00992793

34.

Macedo

Alex-Assensoh

Berry

Brintnall

Campbell

D.E.

Fraga

L.R.

. . . Cramer Walsh

(2005). Democracy at risk: How political choices undermine citizen participation and what we can do about it. Washington, DC: Brookings Institution.

35.

McDewitt

Kiousis

(2006). Deliberative learning: An evaluative approach to interactive civic education. Communication Education, 55, 247-264. doi:10.1080/03634520600748557

36.

McLeod

Shah

(2009). Communication and political socialization: Challenges and opportunities for research. Political Communication, 26, 1-10. doi:10.1080/10584600802686105

37.

McLeod

(2000). Media and civic socialization of youth. Journal of Adolescent Health, 27, 45-51. doi:10.1016/S1054-139X(00)00131-2

38.

Möller

De Vreese

(2015). Spiral of political learning the reciprocal relationship of news media use and political knowledge among adolescents. Communication Research, doi:10.1177/0093650215605148

39.

Neundorf

Smets

García-Albacete

G. M.

(2013). Homemade citizens: The development of political interest during adolescence and young adulthood. Acta Politica, 48, 92-116. doi:10.1057/ap.2012.23

40.

Niemi

R. G.

Hepburn

M. A.

(1995). The rebirth of political socialization. Perspectives on Political Science, 24, 7-16. doi:10.1080/10457097.1995.9941860

41.

Prior

(2007). Post-broadcast democracy: How media choice increases inequality in political involvement and polarizes elections. New York, NY: Cambridge University Press.

42.

Prior

(2010). You’ve either got it or you don’t: The stability of political interest over the life cycle. The Journal of Politics, 72, 747-766. doi:10.1017/S0022381610000149

43.

Robison

(2017). The social rewards of engagement: Appealing to social motivations to stimulate political interest at high and low levels of external efficacy. Political Studies, 65, 24-41. doi:10.1177/0032321715619431

44.

Russo

Stattin

(2017). Stability and change in youths’ political interest. Social Indicators Research, 132, 643-658. doi:10.1007/s11205-016-1302-9

45.

Ryan

Deci

(2000). Intrinsic and extrinsic motivations: Classic definitions and new directions. Contemporary Educational Psychology, 25, 54-67. doi:10.1006/ceps.1999.1020

46.

Sears

D. O.

(1981). Life stage effects upon attitude change, especially among the elderly. In Kiesler

S. B.

Morganb

J. N.

Oppenheimer

V. K.

(Eds.), Aging social change (pp. 183-204). New York, NY: Academic Press.

47.

Sears

D. O.

(1983). The persistence of early political predispositions: The roles of attitude object and life stage. Personality and Social Psychology Review, 4, 79-116.

48.

Shani

(2009). On the origins of political interest (Doctoral dissertation). Princeton University, NJ.

49.

Silvia

P. J.

(2008). Interest—The curious emotion. Current Directions in Psychological Science, 17, 57-60. doi:10.1111/j.1467-8721.2008.00548.x

50.

Smetana

Robinson

Rote

W. M.

(2016). Socialization in adolescence. In Grusec

Hastings

(Eds.), Handbook of socialization research (pp. 60-84). New York, NY: Guilford Press.

51.

Strömbäck

Djerf-Pierre

Shehata

(2013). The dynamics of political interest and new media consumption: A longitudinal perspective. International Journal of Public Opinion Research, 25, 414-435. doi:10.1093/ijpor/eds018

52.

Strömbäck

Shehata

(2010). Media malaise or a virtuous circle? Exploring the causal relationships between news media exposure, political news attention and political interest. European Journal of Political Research, 49, 575-597. doi:10.1111/j.1475-6765.2009.01913.x

53.

Teorell

Torcal

Montero

J. R.

(2007). Political Participation: Mapping the Terrain. In van Deth

J. W.

Montero

J. R.

Westholm

(Eds.), Citizenship and Involvement in European Democracies: A Comparative Analysis, pp. 334-357. London & New York: Routledge.

54.

van Deth

J. W

. (1990). Interest in politics. In Kent Jennings

(Ed.), Continuities in political action: A longitudinal study of political orientations in three Western democracies (pp. 275-312). Berlin, Germany: Walter de Gruyter.

55.

van Deth

J. W.

Abendschön

Vollmar

. (2011). Children and politics: An empirical reassessment of early political socialization. Political Psychology, 32, 147-174. doi:10.1111/j.1467-9221.2010.00798.x

56.

van Deth

J. W.

Montero

J. R.

Westholm

. (Eds.). (2007). Citizenship and involvement in European democracies: A comparative analysis. New York, NY: Routledge.

57.

Verba

Schlozman

K. L.

Brady

H. E.

(1995). Voice and equality: Civic voluntarism in American politics. Cambridge, MA: Harvard University Press.

58.

Zaller

(1992). The nature and origins of mass opinion. New York, NY: Cambridge University Press.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.14 MB