The thing I have always loved - as in eating massive amounts of popcorn, drinking beer and laughing my ass off - about the social sciences, is how they treat data definitions.
In the hard sciences, you can find books - entire books - compiled by hundreds of scientists all over the world over decades, that drill down on the precise definitions of single word terms that are used for data in scientific studies. Mathematical charts, arguments, diagrams, philosophies. And why? Because the data is what is used to determin the science, so the definition of the data is utterly crucial.
But in social science, these “scientists” wave their hands and talk about “beauty” and “desirability” and other vagueries and then thrown in hard science “genetics” and voila’ - a published paper that AMCNBCCBS vomits out in the news and everyone freaks out about.
Grudgingly, over the years, I have had to acknowledge that statistical analysis does provide a basis for the social sciences - barely. Because there are more ways to abuse statistics then there are to use them properly. And you’re still stuck with deciding what data you are going to statistically crunch.
But hey, ignoring these issues mean that yes, you too can get a multi-million dollar government grant to “prove” that beautiful women attract men more than ugly women. Oops, see? I said ugly. No PC DemonRat grant for me.
Friend, your problem is obvious -- you're making the wrong comparisons.
The exactitudes & precision of analyses in the social sciences should not be compared to natural sciences, but rather to political science -- and now, the light comes on, right?
I mean... compared to political science where, "if you like your plan, you can keep your plan" means precisely... means exactly... means certainly... means conveniently... means amazingly...
Means whatever it is it means, whenever it is meant to mean it.
I mean, compared to that, social sciences are exact, well defined and rigorously practiced, wouldn't you agree?
;-)