Which may only generalize to a specific group of people, in a specific country, at a specific period of time. All subject to change when we tweak a couple of variables.

That point, while valid, is only against this study; it doesn't speak to studies in general.

If we're speaking in the context of social sciences I would say it applies to nearly all of them.

Possibly. While I'm definitely sympathetic to the idea that social science (at least as it's actually practiced) is relatively soft, I don't know how we'd go about discounting it all. Seems to me like we'd be throwing the baby out with the bathwater.

At the very least, I don't know what we could do differently, short of improving the practice of social science in various ways. The questions it would purport to investigate (i.e. social questions) are of course critical questions, but I'm not sure how we'd better answer them. More specifically, I'm not sure how we'd better answer them now. Sure, maybe if we had some Grand Unified Theory or even sufficiently sophisticated computer models. But we don't have those. (In the case where we did have them, it seems almost tautological that social science as such would have participated in their creation--but I digress.) What we've got now are social questions worth asking, and an admittedly imperfect practice of answering those questions. What do we do now?

We realize that scientific studies that are in the social sciences may only generalize to a specific group of people, in a specific country, at a specific period of time. All subject to change when we tweak a couple of variables.