More on Bad Science

Graham Peterson takes a bold and controversial stance on debunking bad science…he’s against it.  I am sure Mr. Peterson has the right intentions here, and indeed he appeals to a sense of civility and a healthier community of knowledge.  And nasty methodological sniping may not be pretty, it’s necessary.  Here are a few quick thoughts on Peterson’s no-doubt well-intentioned argument.

  1. Good luck convincing academics and journalists to not fight about anything and everything.
  2. When someone is convinced by a flim-flam, collective knowledge takes a little step backwards.
  3. A culture of civility, assuming good intentions, and deference to other’s methods benefits bad actors more than anything else.
  4. If a more congenial attitude of scholarly inquiry will benefit all of us, where’s the evidence? Until I see a well-specified quasi-experimental design (or randomized experiment, even better!) with a bunch of stars next to the regression coefficients.*  I suspect Peterson is hiding something, and may be committing scholarly malpractice.

To be more serious: bad academic research is corrosive to the profession and to society as a whole.  For people who study and try to advance social science, there is a professional obligation to fight against bad science that hurts the public.  Yeah, people might pick on the “other team” more than their own, but there’s no better way to address this by spreading methodological expertise and knowledge more widely.  Nasty econometric fights play a small but helpful role in this.

*: Even though stars are silly and p-values are methodologically suspect, I know.


Tags: , , , , , ,

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s

%d bloggers like this: