![Sample our Politics & International Relations journals, sign in here to start your FREE access for 14 days]({"type":"image" , "src" : "/sda/5947/TOC-Subject-Sample-2021-Politics-International-Relations.jpg"})
Abstract
This article attempts to explain declining levels of voter turnout among young adults in terms of decreasing levels of political knowledge. Using data from a representative national survey of the Canadian electorate conducted in 2007, we find through descriptive and inferential statistics that younger individuals are more politically illiterate than older generations by a margin of 20–30 percentage points. We also detect that this generational political knowledge gap accounts for approximately half of the turnout gap that exists between voters in their early 20s and voters in their 50s. Our results further demonstrate that eliminating the knowledge gap would significantly increase turnout among young voters by 15 percentage points.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1. Putnam (Citation2001) and Hay (Citation2007) explore the erosion of social interest in political life, including voting.
2. More generally, the literature concurs that youth participation has declined between 20 and 40 percentage points in Western countries over the past 50 years (Butler & Kavanagh, Citation1997).
3. For example, Tedin (Citation1974) finds that young individuals, who are raised in households where parent–child discussions about politics take place, are more likely to become politically informed, which, in turn, increases their likelihood and curiosity for political engagement. Focusing on the socio-economic status Verba, Schlozman, and Burns (Citation2005) argue that children in wealthier families are more likely to have access to newspapers and might more frequently engage in political discussions with their parents, which, in turn, may render those children curious or enthusiastic about politics (see also Anderson & Goodyear-Grant, Citation2008; Gimpel, Lay, & Schuknecht, Citation2003).
4. According to Marshall (Citation1952), adults 25 or younger are still incomplete citizens. As such, they still have to acquire the political and social competences to fully participate in the political life of a society.
5. In support of this claim, Delli Carpini and Keeter (Citation1996) affirm that political knowledge at each formal level of education (i.e. high school, college or graduate) has been declining over the past half century.
6. The coefficients are shown in their standardised form and analogous to and , the age variable is log transformed.
7. We also regress political knowledge (and the control variables) on self-reported turnout for our five-age categories separately to illustrate that political knowledge matters for somebody’s decision to cast his or her ballot. In all five models, we find that political knowledge is a significant predictor of turnout.
8. If we replace the dependent variable, self-reported voting in federal elections, with self-reported voting in either provincial or local elections, the mediating relationship remains the same. The standardised coefficients for the path from age to political knowledge, for the path from political knowledge to self-reported voting and for the direct path from age to self-reported voting have basically the same size as in the model reported here. The coefficients for both the direct and indirect paths from the control variables to political knowledge and voting are also very similar to the model shown here.
9. As robustness check, we use the Canadian Election Studies (Citation2011) and run an equivalent model to our main recursive model. Due to the fact that the CES was conducted before the election in 2011, we include as the dependent variable, somebody’s voting intentions. The mediating variable is political knowledge (i.e. name recognition of the Canadian federal minister of finance). The independent variables are education, income and gender. This additional specification confirms that political knowledge mediates the positive relationship between age and a higher likelihood of voting.
10. In the model, all coefficients are standardised betas. Because political knowledge and voting are endogenous, we include residuals and residual variances into the model. We also correlate income and education, age and income, and gender and income.