'What Kind of Bias Do I Want?' How Cross-Pressured Voters Select Ideological Media

Ideological media -- including traditional and online media content -- is an important feature of the ideological landscape, polarization, and elections in many democracies today. Cross-pressured voters (CPVs) are a large and growing demographic group that face trade-offs when selecting this ideological media. These voters defy traditional ideological groups by holding either economically-leftist and culturally-conservative positions (“left-conservatives”) or economically-rightist and culturally-progressive positions (“right-progressives”). Recent evidence suggests that there are clear gaps in the supply of media outlets so that CPVs lack like-minded outlets that share their left-conservative or right-progressive positions. We do not know how CPVs navigate this trade-off when selecting ideological media outlets and content. I argue that issue salience influences how they select media outlets and content. Accordingly, when CPVs have higher economic salience relative to cultural salience, left-conservatives select more left-progressive outlets and right-progressives select more right-conservative outlets, while both groups should select more economic-related content within these outlets. The reverse should be true when cultural salience is higher. To test these theoretical expectations, I administer the first CPV survey of its kind by using filtering questions (N≈3000) to obtain a solely-CPV sample (N≈1200) in the UK. This survey contains an experiment and questions that simulate the process of selecting an online news outlet and article. It allows me to test whether salience influences how CPVs select ideological media outlets. Once they select an outlet, I also test whether their choice of economic or cultural content from that outlet is affected by salience and the outlet's ideology. To increase external validity, I test the same hypotheses using panel data from the US, the UK, and Germany. The preliminary observational analysis supports the theory and the survey experiment will be administered in late January 2025. This research has important implications for how the growing demographic of CPVs use media to develop their political opinions and vote.