It is commonly believed that education causes individuals to become more progressive in their political outlook. However, simply measuring years of education, as has been common in the literature, risks glossing over potentially crucial differences in the content of education, i.e. particular fields of study. Additionally, achieving a causal identification of educational effects can be notoriously difficult. We suggest an approach that leverages comparisons between siblings in identical twin pairs, commonly referred to as a discordant twin design. This allows us to distinguish selection effects from (potential) causal effects that arise from differences in field of study between siblings that share their family background, as well as their genetics. The study is based on a sample of twins in the Swedish Twin Registry (STR), which have been linked to register data concerning university fields of study, and a survey containing measures of ideology. The results indicate that there are effects of enrolling in a social science or humanities field as opposed to a STEM field, on economic and social ideology. The predicted difference in economic ideology was estimated at 0.3 standard deviations, with social science/humanities education being associated with more progressive economic attitudes. The corresponding difference concerning social ideology was a slightly larger 0.4 standard deviations. The results have implications for how we should think about the role of higher education in shaping ideological cleavages in society, since academic fields of study may in fact have causal effects on how citizens make sense of political issues.