Does the higher education experience affect political interest, efficacy, and participation? Comparing dropouts to graduates and ‘non-starters’

ABSTRACTIn this paper, we focus on non-monetary and potential societal benefits of higher education and ask whether the higher education experience fosters political interest, internal political efficacy, and participation irrespective of completing a degree. Increasing enrolment rates in higher education also increase the number of higher education dropouts. Previous literature has found that dropouts fare equally well on the labour market as young adults who never enter higher education (‘non-starters’), suggesting that higher education does not pay off without obtaining a degree. However, preparing students for the labour market is only one goal of higher education. Therefore, we focus on broader returns to the higher education experience and its potential contribution to society. We use a representative sample of adults in Germany and apply linear regression models to compare dropouts to both graduates and ‘non-starters’. Our results confirm that higher education experience is positively related to political outcomes. However, except for internal political efficacy, the differences between dropouts and ‘non-starters’ do not appear to change substantially with the time spent in higher education. These findings call into question the role of higher education in creating politically informed and responsible citizens.KEYWORDS: Consequences of dropping outhigher educationpolitical efficacypolitical interestpolitical outcomepolitical participation AcknowledgementThis paper uses data from the National Educational Panel Study (NEPS): Starting Cohort 6 – Adults, doi:10.5157/NEPS:SC6:12.0.1. From 2008 to 2013, NEPS data were collected as part of the Framework Programme for the Promotion of Empirical Educational Research funded by the German Federal Ministry of Education and Research (BMBF). As of 2014, the NEPS survey is carried out by the Leibniz Institute for Educational Trajectories (LIfBi) at the University of Bamberg in cooperation with a nationwide network.Disclosure statementNo potential conflict of interest was reported by the author(s).Notes1 OECD considers the period of three years after the end of the theoretical duration of the programme for calculating dropout rate.2 There are subtle differences between signalling approaches (Spence Citation1973) and credentialism (Collins Citation1979), in that the former assume diplomas to be generally valid signals of true productivity while the latter tend to view diplomas as symbolic capital, biased towards middle-class values. Further investigation of such differences is beyond the scope of our paper. Here, we simply note that both approaches are compatible with the reported findings that higher education only pays off with a degree.3 The restricted sample excludes 320 out of 5,390 individuals (about 6%) with a higher education entrance certificate. These 320 individuals appear to have graduated from higher education within two years or less, or to have spent more than 12 years in higher education. Although sensitivity analyses confirm that keeping these individuals in the sample does not substantially alter our results, we deem these cases implausible or at least atypical and exclude them from the main analyses. The full details of our sample restrictions are available in terms of Stata (StataCorp., Citation2019) syntax scripts from the authors upon request.4 Note that the respective item (Table 1) measures subjective believes about competence rather than competence itself. For a more in-depth discussion, see Saris and Torcal (Citation2009).5 Note the subtle difference between individuals and observations when the political outcome is measured repeatedly over time. From the repeated observations within individuals, we exclude only those observations with missing values. Sensitivity analyses confirm the robustness of our results, irrespective of whether we include observations with missing values on political outcomes or not. We prefer the restricted samples as they facilitate the standardization and, thus, interpretation of the results.6 Additional tables comparing covariate distributions between included and excluded observations for each of the six political outcomes are in the Appendix B.7 For binary outcomes, yt, the linear regression model in (1) is also referred to as the linear probability model (LPM). In the LPM, the coefficients β1 and β2 represent the differences to the predicted probabilities for dropouts. We ran sensitivity analyses, estimating average marginal effects from binary logistic regression models. The results closely resemble those from the LPM. We prefer the LPM as it facilitates dealing with missing values in the covariates by (full information) maximum likelihood estimation, which would otherwise require specialized software.8 Voter turnout suffers from over-reporting. While official statistics report voter turnouts of 71.5% in 2013 (Böth and Kobold Citation2013) and 76.2% in 2017 (Stemmer Citation2017), the (unweighted) estimates from our sample are above 90% for 2013 and near 100% for the 2017 elections (see Table 1).9 We exclude higher education graduates from the further analyses because graduating from higher education requires a minimum amount of time. Thus, there is a ‘mechanical’ correlation between the time spent in higher education and graduating, making it harder to interpret the meaning of time. Dropout, on the other hand, can occur at any time after enrolment.10 For a related discussion, see Milligan, Moretti, and Oreopoulos (Citation2004, 1668), who distinguish between increased cognitive skills and increased political interest as two channels through which education might influence political outcomes.Additional informationNotes on contributorsShweta MishraDr. Shweta Mishra is the head of the research area ‘Students and Graduates’ at INCHER-Kassel. Her research interests include social network analysis as well as multivariate statistics focusing on the higher education experiences and academic success of students, particularly students from low social/educational backgrounds, migrant students as well as students with disabilities.Daniel KleinDaniel Klein is a research assistant at the German Centre for Higher Education Research and Science Studies (DZHW). His research interests are in the fields of sociology of education and social stratification.Lars MüllerDr. Lars Müller is a sociologist and researcher at the Center for Teacher Training (ZfL) at the Justus-Liebig-University Giessen. His research interests include student engagement and student dropout with a special focus on underrepresented student groups.