As it turns out, I wrote my dissertation (10 years ago, eek!) on how people answer survey questions when they aren't relying on ideological or partisan cues. A brief thread (and link to the publication at the end): https://twitter.com/DKThomp/status/1318931328696451073

I was curious about this exact question that @DKThomp alludes to - how are people with no real preferences forming their opinions?

The obvious answer is partisanship/ideology heuristics, as established in tons of political science literature. But what about the disconnected?

My theory was that they fall back on deeper-rooted concepts of how the world should work - to operationalize that I used a theory from anthropology ("Cultural Theory," Douglas & Wildavsky) that uses a battery of survey questions to establish how people want the world to operate.

(I was low-tech back then, don't judge)

Anyway, for those who skip over/don't pay attention to the institutions and ideologies, they are going to revert back to their core belief in how the world should work (Cultural worldview) to form quick reactions to issues on a survey.

And obviously since this is published and accepted as my dissertation, it worked. I built path models to approximate the relationships using ideology as an intervening variable:

Dual finding:

-Among those who claim lib/cons ideological views, the effects of cultural worldview were only impacting issue positions via ideology. Ideology biggest predictor.

-Among those (quite a few) who say "moderate" or DK to ideology, the effects of cultural worldview were direct predictors of issue positions, not ideology.

Publication (it took me a few years since I went nonacademic and pubs didn't matter): https://onlinelibrary.wiley.com/doi/full/10.1111/ssqu.12136

So to summarize, people can and do form quick opinions on issues in the absence of paying attention to politics using their core values of how they think the world should work.

And it's useful information. Don't assume politics is the only way people can make judgments. /end

PS: You can get the same basic findings using partisanship, but I used ideology b/c it's more theoretically similar to cultural worldview.