Educational Attainment Has a Causal Effect on Economic, But Not Social Ideology: Evidence from Discordant Twins
| Author | Stig Hebbelstrup Rye Rasmussen,Aaron Weinschenk,Asbjørn Sonne Nørgaard,Jacob von Bornemann Hjelmborg,Robert Klemmensen |
| DOI | http://doi.org/10.1177/00323217211008788 |
| Published date | 01 February 2023 |
| Date | 01 February 2023 |
| Subject Matter | Articles |
https://doi.org/10.1177/00323217211008788
Political Studies
2023, Vol. 71(1) 256 –275
© The Author(s) 2021
Article reuse guidelines:
sagepub.com/journals-permissions
DOI: 10.1177/00323217211008788
journals.sagepub.com/home/psx
Educational Attainment
Has a Causal Effect on
Economic, But Not Social
Ideology: Evidence from
Discordant Twins
Stig Hebbelstrup Rye Rasmussen1,
Aaron Weinschenk2, Asbjørn Sonne Nørgaard3,
Jacob von Bornemann Hjelmborg4
and Robert Klemmensen5
Abstract
In this article, we examine the nature of the relationship between educational attainment and
ideology. Some scholars have argued that the effect of education on political variables like ideology
is inflated due to unaccounted-for family factors, such as genetic predispositions and parental
socialization. Using the discordant twin design and data from a large sample of Danish twins, we
find that after accounting for confounders rooted in the family, education has a (quasi)-causal
effect on economic ideology, but not social ideology. We also examine whether the relationship
between education and economic ideology is moderated by levels of economic hardship in the
local context where individuals reside. We find that the (quasi)-causal effect of education on
economic ideology increases in economically challenged areas.
Keywords
education, causality, political ideology
Accepted: 21 March 2021
1Department of Political Science, University of Aarhus, Aarhus, Denmark
2Department of Political Science, University of Wisconsin-Green Bay, Green Bay, WI, USA
3Cevea, København, Denmark
4Department of Health, University of Southern Denmark, Odense, Denmark
5Department of Political Science, University of Southern Denmark, Odense, Denmark
Corresponding author:
Stig Hebbelstrup Rye Rasmussen, Department of Political Science, University of Aarhus, 8000 Aarhus,
Denmark.
Email: stighj@hotmail.com
1008788PSX0010.1177/00323217211008788Political StudiesHebbelstrup Rye Rasmussen et al.
research-article2021
Article
Hebbelstrup Rye Rasmussen et al. 257
A growing and methodologically sophisticated literature has questioned the time-honored
assumption that education is one of the most important causal factors influencing a host
of political behaviors such as political participation, political interest, political sophistica-
tion, and social trust (Berinsky and Lenz, 2011; Dee, 2004; Dinesen et al., 2016; Highton,
2009; Kam and Palmer, 2008; Oskarsson et al., 2017; Rasmussen, 2015; Shani, 2009;
Sondheimer and Green, 2010; Weinschenk and Dawes, 2019). The shared point of criti-
cism in this burgeoning literature is that nobody enters an educational institution as a
blank slate, and that—due to problems of self-selection—the quasi-causal effect of edu-
cation is inflated. Education may be nothing more than a proxy for predispositions that
are influenced by genetics and other pre-adult factors (Gidengil et al., 2019; Kam and
Palmer, 2008). So far, no one has critically examined the effect of education on political
ideology while taking genetic and pre-adult social influences into account. This article
undertakes such an analysis.
Few of the previous studies arguing that the education effect is inflated due to unob-
served confounding have distinguished between the socialization effects and resource
effects of education (cf., however, Rasmussen and Nørgaard, 2017). By failing to distin-
guish clearly between these different effects of education, we risk misunderstanding
when and why education is confounded by unobserved variables. Furthermore, no previ-
ous study has examined the differences between socialization and resource effects using
twin analyses, which has the obvious advantage that family factors can be directly
modeled.
We argue that social and economic ideology are ideal cases if one wants to examine
which educational effects are more likely to be confounded by family factors. Although
economic and social ideology are likely to be affected by both types of effects, the impact
of education on economic ideology is likely more strongly related to resources, whereas
the impact of education on social ideology is likely more strongly related to socialization
effects. We elaborate on these ideas in more detail below.
To examine whether and how education has a quasi-causal effect on social and eco-
nomic ideology, we examine twins who are discordant on educational attainment. The
research design utilizes the fact that monozygotic (MZ) twins share all of their genes,
whereas dizygotic (DZ) twins share, on average, half their separating genes (like other
biological siblings) and that both MZ and DZ twins reared together share their upbring-
ing. Therefore, unobserved confounders related to family factors are also shared, and
within-pair differences in education that are associated with differences in ideology are
less prone to (unobserved) confounding (McGue et al., 2010). Previous discordant twin
studies have examined whether differences in educational attainment between MZ twins
lead to differences in social trust, political participation, and political knowledge (Dinesen
et al., 2016; Oskarsson et al., 2017; Weinschenk and Dawes, 2019). In these studies, the
effect of education was substantially confounded by family factors, mainly genetic
influences.
In the section that follows, we briefly examine the literature on the relationship
between education and ideology and argue that this relationship is likely confounded by
family factors, although more so for social ideology, in which socialization effects are
dominant. We then present two discordant twin studies that take family factors into
account and allow us to more precisely estimate the effect of education on ideology and
the sources of confounding. Study 1 shows that both genetic and environmental pre-adult
influences confound the educational effect on social, but not economic, ideology. While
To continue reading
Request your trial