The Huffington Post has a new story up on how altering who is said to support a given policy changes how persons of different partisan persuasions register their opinions. So saying that "universal health care" is a Donald Trump-approved position makes Democrats less likely to support it and Republicans more so; attributing it to Barack Obama has the opposite effect.
For someone like me who is interested in how cultural affinities construct political beliefs, this is an important topic. That notwithstanding, I think the HuffPo article is potentially misleading in at least two respects. The first is the title -- "Republicans Like Obama's Ideas Better When They Think They're Donald Trump's" -- which implies this is a Republican problem when in reality (as the article makes clear) it is true of partisans of either party.
The second problem, though, challenges just how far we can take the implications of these findings. The issue is that while saying someone supports "universal health care" or "affirmative action" does tell us something, it doesn't tell us all that much. There are, after all, many different ways one might operationalize support for universal health care or affirmative action. And it is reasonable for a conservative to believe they are more likely to favor a Donald Trump-style instantiation, and are less likely to find Obama's version amenable (and vice versa). So for that reason, it is not entirely odd, or purely a matter of partisan hackery, that party identity affects how one responds to a question like "Do you agree or disagree with [Obama/Trump] about universal health care?"
None of this is to discount the point that cultural identity (here taking the form of party allegiance) plays a substantial role all its own. The literature supporting such an inference is robust, and this fits nicely into that puzzle. But it is worth taking this particular article with a grain of salt -- two grains, since its conclusions line up with those I generally share.