Yes that’s ridiculous. I’d personally write to the editor first and then withdraw it if it’s not sent out within a reasonable time frame…

Fair enough, it's the widespread leap from that to -> inference about real-world equilibrium behavior/outcomes that perturbs me but I can see how that's an improvement as far as trying to model the individual process of political cognition

I can appreciate that. What do you see as an example of a topic where survey experiments have greatly improved knowledge greatly over the old approach? Asking out it curiosity as I come at the whole thing from a CP/PE as perspective opposed to behavior

Curious to read your reasoning. Mine: 1. informational equivalence effectively makes them observational quite often: 2. credibility revolution was about empirical strategies to overcome real-world endogeneity, survey exps do opposite, identification by removing realism

I agree with this 100%

I wouldn't disagree in principle, it would just seem rare in practice in social science that there aren't plausibly many open backdoor paths paths or reverse causality between X and Y...? especially when it comes to 'big' variables we typically care about