1/ Something I wish people outside of #academia understood: You can find a paper with quantitive evidence or sound arguments -- and, sometimes, even both! (har har) -- supporting just about any shit post you want to make.
That doesn't mean it's True.
-
Show this thread
-
2/ Unless you understand how it fits into the rest of the evidence, you risk drawing bad inferences and mistaken conclusions. When people *present* a single study or two as conclusive "proof" of something, they either don't understand this or they're selling you something.
1 reply 1 retweet 6 likesShow this thread -
3/ That's also why I think social science
#SciComm is so much harder than other fields. If you constrict evidence for pedagogical reasons in something mind-bending like quantum mechanics, the harm is very contained. The learner won't break quantum mechanics.1 reply 2 retweets 6 likesShow this thread -
4/ But in the social sciences, people are very prone to using scientific evidence as Scientism to support unsupported prescriptions, either at institutional or policy level, or socially. Here, bad scicomm can induce harmful changes in the objects under study!
2 replies 4 retweets 6 likesShow this thread -
Replying to @generativist
I don't want to toss dirt on an entire family of disciplines, but... Is it even *possible* to obtain statistically rigorous and extrapolable results in the social sciences? Has always seemed the systems are too complex, w/too many moving parts, to achieve that level of rigor.
2 replies 0 retweets 1 like -
Replying to @btskinn
Sometimes it is and sometimes it is not. But, it is almost always easy to do poorly.
1 reply 0 retweets 1 like -
Replying to @generativist
Can you direct me to examples where it was done super-well? I'm really interested in what that sort of methodology would look like.
1 reply 0 retweets 1 like
Hrm. Let me think on some "good" examples.
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.