From Clocks and Clouds VOL. 6 NO. 2
Revitalizing Young-Adult Citizenship: An Analysis of High-School Predictors on Civic Engagement
IN THIS ARTICLE
Civic engagement and political participation among the US population are waning, and this is particularly apparent in the young adult citizenry. This research paper seeks to assess which variables in civic education and the high school experience have the strongest impact on young adults' levels of civic engagement. By using a nationally representative sample of 4,483 young adults aged 18-24, this study employs multiple regression analyses to investigate the influence on civic education, classroom climate, extracurricular activities, and community service on respondents' civic engagement levels. The results indicate that high quality civic education, extracurricular activity involvement, and community service are significantly related to civic engagement levels. In particular, community service as a predictor, both when a volunteer activity and as a mandatory activity, were significant. The findings suggest that high quality civic education, rather than simply the quantity of classes taken, may be a more important factor in civic engagement. Additionally, the study indicates that out-of-classroom experiences can have a meaningful impact on young adults' levels of civic engagement. These findings can support educators and policymakers as they develop initiatives to improve political engagement among young adults.
In the United States, mandates for public schools to provide civic education to their students are included in several state constitutions. But despite this being a chief responsibility for public school systems, the present outcomes of such civic education programs are less than promising. Civic engagement and political participation among Americans has been waning, with young adults appearing to be even more disengaged than the rest of the adult population (Putnam 2000). In fact, since 1964, young-adult voting rates have been in an almost constant decline, except for slight increases in 1992, 2004, and 2008 (File 2013). After the 1972 presidential election, voter turnout for the voting-age population between 18 and 24 years never exceeded 50 percent participation, with only 38 percent of this age group voting in the most recent 2012 presidential election (File 2013).
In addition to the decline in voter turnout, Putnam (2000) has also documented a decline in other forms of civic engagement, including participation in political campaigns, rallies, and organizational membership within communities. This downward trend has been particularly notable among young adults. (Putnam 2000).
In response to these alarming figures of non-engagement, educators, policy makers, and political scientists have questioned how civic education can be modified and improved to foster greater levels of civic engagement among young adults. In January 2012, the U.S. Department of Education released a report recognizing that, "unfortunately, civic learning and democratic engagement are add-ons rather than essential parts of the core academic mission in too many schools and on too many college campuses today" (U.S. Department of Education 2012, 1). Similarly, at the university level, the U.S. Department of Education commissioned a 2011 report calling to enhance educational practices at the postsecondary level of education (U.S. Department of Education 2011).
These recent reports, coupled with Putnam's Bowling Alone, have revitalized the effort to explore the nuances of civic education. More specifically, recent research has attempted to evaluate the influence of several aspects of civic education on student's present and future political participation and involvement. This research paper seeks to assess how variables related to a high school experience, including civic education, extracurricular activities, community service, and classroom climate, impact young adult civic engagement.
Civic education is best understood by political scientist David Campbell's definition as "the knowledge, skills, attitudes, and experiences to prepare someone to be an active, informed, participant in democratic life" (Campbell, Levinson, & Hess 2012, 1). The research in civic education is largely divided into two trajectories of literature. The first body covers traditional classroom factors. In other words, this scholarship investigates the influence of factors inside the classroom, such as classroom climate, diversity, discussion of current events, and the number of civics classes taken. The second school of thought, however, argues that socialization experiences, which include community service and extracurricular activities related to civic education best cultivate future civic engagement.
Traditional Classroom Influences
A focus on the way classroom factors influence future rates of civic engagement received significant attention in the mid-twentieth century, and still persists today. Both quantitative and qualitative data have been evaluated in this process.
Early research in the mid 20th century sought to explore the connection between civic education and civic engagement. However, doubts were raised about the effectiveness of civic education when studies failed to find a significant association between taking civics courses in high school and the development of students' political knowledge and interest. (Langton & Jennings 1968). Furthermore, when Ehman, a professor of education at Indiana University considered the number of social studies courses a student has taken, the results were not conclusive in regards to the classes' influence on a student's political socialization (Ehman 1969).
More recently, Manning and Edwards (2014) conducted a metareview of the literature surrounding the relationship between the number of civics courses a student has taken and future political participation. Overall, they found that based on the present literature, "no clear pattern emerges from the findings" linking the amount of civic education received with future political participation (Manning and Edwards 2014, 37). From these ambiguous results, other factors may need to be considered when examining connections between civic education and future political engagement.
Ehman (1969) also investigated the impact of classroom climate on political socialization. Classroom climate can be described as the degree to which students feel comfortable discussing controversial issues and disagreeing with their peers and teachers (Campbell 2008). The study found that discussing controversial issues within an open-minded classroom climate was positively and significantly related to students' political socialization in a Detroit public high school (Ehman 1969).
More recent research has also studied factors such as classroom climate and levels of diversity rather than just the number of civic education classes a student has taken. Campbell (2007) found that higher rates of racial diversity are negatively correlated with the amount of political discussion that takes place within the classroom. As a result, this lack of open classroom discussion was correlated with high school students indicating that they would be less likely to become informed voters (Campbell 2007).
A year later, Campbell (2008) conducted another study, this time focusing on classroom climate. For students who perceived their classroom to have an open climate, they scored higher on civic knowledge questions and indicated that they would be more likely to vote in future elections (Campbell 2008). The impact of the open classroom climate was greater for students with lower socioeconomic statuses (Campbell 2008). Campbell suggested that open classroom climate in civic education courses can help to mitigate the typical gaps of civic engagement between students with high socioeconomic statuses and low socioeconomic statues (Campbell 2008). From this scholarship, the idea originates that the quality and environment of the classroom experience may be of greater importance than just the quantity of exposure to civic education.
A second school of thought reasons that extracurricular activities and community service best engage and equip students for future civic and political involvement. A wide array of research has reached a consensus that students who participate in extracurricular activities are more likely to be civically engaged in their communities as young adults (Hart et al. 2007, Kahne and Sporte 2008, Annette 2005, Glanville 1999). These studies also suggest that extracurricular activities equip students with the skills necessary for participation in civic life. From holding meetings, collaborating with others, and making decisions, these skills arguably can then be transferred to the civic and political arenas as adults (Hart et al. 2007, Kahne and Sporte 2008, Annette 2005, Glanville 1999).
Not all extracurricular activities, though, are seen to produce equal outcomes. According to sociologist Jennifer Glanville (1999), the type of extracurricular activity matters when related to future civic and political involvement. Glanville (1999) categorized extracurricular activities into two groups: instrumental organizations and expressive activities. Instrumental organizations, such as debate teams, school yearbooks, student government, and political club participation are associated with greater expected political participation in early adulthood, including voting (Glanville 1999). Participation in expressive activities, such as sports teams, academic teams, and the performing and fine arts, however, was not significantly correlated with expected political engagement (Glanville 1999).
These results may be in part due to different types of skills gained from the extracurricular activities, with instrumental organizations being more likely to nurture skills for civic life, such as negotiation, decision-making, and voting (Glanville 1999). It is also possible that students who self-select into activities such as student government rather than a performing art such as chorus, already have expressed political interests (Glanville 1999)
What is difficult to extrapolate from this body of literature is whether the students who self-select into extracurricular activities are predisposed to political and civic engagement, or whether these activities help students to develop these interests and behaviors. Furthermore, whether students were required to participate in such activities, or whether they were chosen voluntarily was not measured in these studies. This study will interact with this gap in the literature by seeking to understand whether there are differences, in particular, between students who opt-in to voluntary community service, in comparison to those who complete mandated service.
The literature regarding community service suggests a connection between service and civic engagement. A study conducted in 2000 found that participating in community service while in high school promoted political knowledge and participation skills (Niemi, Hepburn & Chapman 2000). Some civic education experts hypothesize that community service can impact students' future civic and political participation because it connects abstract concepts and problems addressed inside the classroom to real-world, tangible scenarios outside the classroom (Youniss & Yates 1997, Crystal & DeBell 2002). Youniss & Yates (1997) argue that this personal involvement connects students directly with the political and social issues within a community. Additionally, community service may facilitate the development of social networks that revolve around community concerns (Crystal & DeBell 2002).
However, when it comes to mandatory community civic engagement. For example, Stukas, Snyder & Clary (1999) suggest that mandatory service requirements for a class or as graduation requisite could foster attitudes of resentment which may counteract the spirit of civic and political involvement. Despite this, one study found that high school community service was significantly, positively correlated with higher rates of voting and volunteering eight years after the student graduated from high school, even when the service was required (Hart et al. 2007). Hart et al. suggested that these results may be explained by the idea that community service in adolescence fosters identity formation and the development of civic skills that can be utilized in adulthood (Hart et al. 2007).
Similar to there being different types of extracurricular activities, Youniss (2012) differentiates between one-time charitable acts and ongoing service that engages the student with social and political issues within the community. Youniss suggests that forms of service relating to social justice, environmental issues, or income inequality, in which students are educated about the topics and become aware of structural implications, act as "potential founts of policy and political education" (Youniss 2012, 129). While this concept has not been investigated at a national level relating to civic engagement, it is a compelling thought worthy of further research.
This paper hypothesizes that when performing separate multiple regression analyses for civic education classes, classroom climate, extracurricular activities, and community service, each model will see these factors to be significantly related to civic engagement levels in their respective models. This paper also hypothesizes that when performing a multiple regression containing all of the independent variables without one analysis, community service participation will serve as the strongest predictor among high school students' future civic engagement levels.
This analysis utilizes data from The Commission on Youth Voting and Civic Knowledge Youth Post-Election Survey (Levine 2012). The data for this survey is compiled from a national representative sample of 18-24 year-old United States citizens from all fifty states. Due to this study including specific information related to students' classroom, community service, and extracurricular experiences, this survey served as a strong data source to answer the research question presented. Included in the data is individual-level information regarding participants' voting behavior, civic education experiences, and political knowledge. In total, 4,483 young-adults participated in phone interviews within the six weeks following the 2012 presidential election. Four main school-related independent variables were evaluated for their influence on the dependent variable, which is respondents' civic engagement.
Dependent Variable: Civic Engagement
The dependent variable, civic engagement, can be defined for this study as the rate of respondents' involvement in community issues, service opportunities, and electoral activities. Civic engagement was compiled from five main factors that respondents may have participated in after graduating from high school. These five factors include: being registered to vote, voting in the 2012 presidential election, volunteering for a political candidate or campaign, attending a public meeting about community affairs, and working in the neighborhood as a community service opportunity. Each of these indicators was coded zero for a "no" response, and one for a "yes" response. Thus, the civic engagement scale ranges from 0-5, with "0" coded as "no engagement", 1 coded as "very low engagement" 2 coded as "low engagement", 3 coded as "moderate engagement," 4 coded as "high engagement," and "5" coded as" very high engagement."
The independent variables for this study include four indicators that may influence young-adult civic engagement. These variables include civic education exposure, classroom climate, extracurricular activities, and community service involvement.
As referenced in the literature review, previous research has suggested that an open classroom climate may play a role in fostering future civic engagement (Campbell 2008). Classroom climate is measured by two questions about respondents' classroom experience.
The two indicators listed above were chosen because they embody Campbell's concept of classroom climate, which includes components such students feeling comfortable to disagree with teachers about political issues, teachers encouraging students to express their opinions, and students' comfort in expressing opinions that differ from their peers (Campbell 2008, 443). For these two statements, the responses have been coded on a five point scale with "1" was coded as "strongly disagree," "2" was coded as "disagree," "3" was coded as "neutral," "4" was coded as "agree," and "5" was coded as "strongly agree."
The second measurement of this study includes the respondents' civic education experience. First, this study assesses whether the participants took at least one civic education course in high school. A "yes" response was coded as ‘1' and a "no" response was coded as ‘0'. Then, five qualifying follow-up questions were asked to assess the respondents' perception of the course's quality.
With each of these qualifying questions being re-coded for 0 meaning "no" and 1 meaning "yes," a Likert Scale of 0-5 was created to assess the perceived quality and usefulness of the course, with "0" being "very low quality" to "5" being "very high quality." Then, this information was visually binned so than an equal percentile was "low quality civic education" coded as 0 and "high quality civic education" coded as 1.
The third measurement as an indicator of civic engagement is participation in extracurricular activities during high school. Two measurements of extracurricular activities were utilized for this study. First, respondents were asked whether they were involved in any form of extracurricular activities during high school. Those involved in zero activities was coded as "0." Involvement in one activity was coded as "1," two activities was coded as "2," three activities was coded as "3," and four or more activities was coded as "4."
Second, the data also measured whether the extracurricular activities were related to "social or political issues." This relates to the divide between expressive and instrumental activities as described in the literature review, in which previous literature suggests students involved with instrumental activities are more likely to be politically and civically engaged as young adults (Glanville 1999). Respondents involved in at least one socially or politically related extracurricular activity were coded as "1". Those who were not involved in socially or politically related extracurricular activities were coded as "0".
The fourth and final indicator used for the independent variables relates to community service. To assess community service experiences within students' high school careers, respondents were asked whether they took part in community service while in high school, and whether such community service was voluntary or mandatory. Dummy variables were created for these community service indicators. For the voluntary community service dummy variable, voluntary service was coded 1 and all other responses (including no service and mandatory service) were coded 0. For the mandatory community service dummy variable, mandatory service was coded 1 with all other responses being 0. Finally, the third dummy variable of no service was coded 1 while all other responses were coded as 0.
Demographic Control Variables
Individual-level demographic variables that have previously been associated with civic engagement were also included within the analysis for the purpose of controls.
Race is included because racial minorities tend to have lower levels of political engagement when compared to non-minorities (Tourney-Purta et al. 2007).
Gender is also included as a control variable. In past research, females have tended to report lower levels of political engagement when compared to their male counterparts (Burns et al. 2001).
This study also utilizes a control variable for the type of school (traditionally public, private, religious, or charter) that the student attended. A meta-analysis conducted by Patrick Woolf in Education Next suggests that students attending private and charter schools tend to exhibit higher rates of community volunteerism and political participation (Woolf 2007).
An individual's expected level of education will also be utilized as a control variable. As a measurement, this paper uses the respondents' mothers' obtained level of with "college or higher" coded 1 and "below college" as 0. Rimukete et al. (2012) has indicated that the relationship exists between adolescents' expected education levels and levels of education obtained by their mothers or close female relatives. Finally, political engagement has been cited as being significantly higher among those with higher socioeconomic statuses (Verba et al. 1995). A common indicator for SES status among adolescents and young-adults in the perceived number of books in the home as a child. Thus, books will serve as a control variable for socioeconomic status. "Books to fill several bookshelves" were coded as "1", and any responses indicating less than several bookshelves worth of books were coded as "0".
Method of Analysis
This study includes multiple independent variables, and seeks to measure the effect of each variable on the dependent variable of civic participation. To do so, the study will use a compilation of bivariate comparisons and multiple regressions to estimate the effects of the independent variables. The bivariate correlation will ensure that each independent variable can be distinguished from the others. Thereafter, the multiple regression analyses will be utilized to measure the relative effects of the independent variables on rates of civic engagement. Four multiple regression analyses will separately measure the impact of the four specified categories of independent variables. Then, a final multiple regression analysis will include all of the variables together in one model. This final analysis allows for each predictor variable to serve as a control against the others.
In the first analysis, as shown in Table 1, a multiple regression analysis was performed with demographic controls and civic education predictors. Of the respondents of the survey, 81.5 percent of the respondents (n=3654) reported having taken a civics course in high school, while 16.7 percent (n=748) indicated they had not taken a civics class while in high school. Two dummy variables were utilized for the civic education predictors. The first dummy variable was coded with high quality civic education responses as ‘1,' and all other responses as zero. The second dummy variable was coded with low quality civic education as ‘1,' and all other responses as zero. Thus, those students not taking part in a civic education course in high school were coded as ‘0' in both cases, and serve as the reference group against which the dummies estimated.
Table 1: Level of Civic Engagement following the 2012 Presidential Election among 18-24 year-old US Citizens with Civic Education Predictors
In this first analysis, as shown in Table 1, receiving high quality civic education in high school yielded statistically significant results as a predictor of civic engagement. In contrast, low quality civic education was not a significant variable. Based on the data set, the mean civic engagement score for students not taking civic education courses was 1.7. For those that rated their civic education courses as low quality, these students' average civic engagement score was 1.9, and for those that indicated their civic education to be high quality, they had an average civic engagement score of 2.3. From this it appears that the quality of civic education is a more important factor than simple general participation rates.
However, some caution should be exercised when viewing these results related to civic education. The data does not account for which students took civic education out of prior interest, or how many civics classes a student took. With that, students who were already interested and engaged in politics or government may have self-reported higher scores on the civic education quality. Additionally, these students who were already engaged in civic education may also be more likely to exhibit high levels of other civic engagement variables reflected in the dependent variable.
In this analysis, certain demographic variables were also statistically significant predictors of a respondents' civic engagement. The three demographic variables that produced significant results included type of high school, socioeconomic status, and expected education. With socioeconomic status and expected education, a positive relationship resulted. In other words, high socioeconomic status and an expected college education were positive predictors of higher civic engagement. For type of high school, the significant relationship is negative. This suggests that students who attend private, religious, or alternative public high schools were more likely to be civically engaged than their peers in traditional public high schools.
The second multiple regression performed analyzed the influence of extracurricular activity involvement in high school on civic engagement for young adults. 63 percent (n=1,852) of the sampled participants indicated that they had participated in at least one extracurricular activity at school, while 37 percent (n=1092) reported that they were in no extracurricular activities at school.
For this analysis, dummy variables were utilized in the regression. For the variable "EC concerned with social or political issues," a "yes" response was coded as 1, and all else was coded as "0." I utilized a second dummy variable, "EC not concerned with social or political issues" with "true" coded as "1" and "all else" coded as "0," serving the reference category for those not involved in any school activities. The third variable, as mentioned in the study design, measures how many extracurricular activities a student was involved in. Table 2 presents the results of the multiple regression analysis.
Table 2: Level of Civic Engagement following the 2012 Presidential Election among 18-24 year-old US Citizens with Extracurricular Involvement Predictors
Based on the analysis, the number of extracurricular activities a student was involved with is a statistically significant predictor of civic engagement among US citizens ages 18-24. Students not involved in an extracurricular activity at school had a mean civic engagement score of 1.7. This average score increased to 2.1 for students in two activities, and students in four or more activities had the highest average civic engagement score, at 2.7.
Additionally, having at least one of the extracurricular activities relate to a social or political issue was significant. These results support the hypothesis that the type of extracurricular involvement is important and matters in relationship to the outcome of civic engagement. This parallels Glanville's findings (1999), where the study indicated that instrumental activities were more influential than expressive activities with political engagement.
Contrasting this is the negative, significant correlation between involvement in extracurricular activities that do not relate to political or social issues and civic engagement. It is not immediately apparent why there is a negative relationship. A review of the literature provides some insight. Those students who participate in extracurricular activities, but not activities that are political in nature, may not develop the skills that prepare these students for civic engagement, such as voting and negotiating (Glanville 1999). It would also be worth exploring whether students in this category were already uninterested or turned off by politics, and therefore self-selected into nonpolitical activities. It could be that those who are so focused on non-political activities become isolated from political activities. This is an area where further research could help better explain this relationship.
Several demographic variables were also significant in the multiple regression analysis. Socioeconomic status and expected education had both positive and significant relationships to the dependent variable, civic engagement. However, students who attended traditional public schools, when compared to all other types of secondary institutions, had negative, statistically significant relationships, suggesting that public school is a predictor of lower civic engagement scores in this model.Continued on Next Page »