I'm pretty interested in this question. I feel like the study (as much as it is described) would not do much to challenge the Shor school of thinking. The desc

I'm pretty interested in this question. I feel like the study (as much as it is described) would not do much to challenge the Shor school of thinking. The description said they asked people to report what they thought of the ads. It seems like that would be unable to capture turnout effects or even measure true swing voters that well. Shor's main thesis is that things that score well among the base (like focusing on racism) don't do well and driving votes while things that aren't very hip (like talking about protecting social security payments) actually do a lot to bring turnout up even if people filling out an online survey wouldn't mark the ad as interesting.