Saturday, September 24, 2016

Replicability in Psychology

This is really Bethany's territory, and I should have run it by her first.  Also, I'm only halfway through it myself. But What Has Happened Here Is The Winds Have Changed by statistician Andrew Gelman looks quite good. 
In short, Fiske doesn’t like when people use social media to publish negative comments on published research. She’s implicitly following what I’ve sometimes called the research incumbency rule: that, once an article is published in some approved venue, it should be taken as truth. I’ve written elsewhere on my problems with this attitude—in short, (a) many published papers are clearly in error, which can often be seen just by internal examination of the claims and which becomes even clearer following unsuccessful replication, and (b) publication itself is such a crapshoot that it’s a statistical error to draw a bright line between published and unpublished work.
Much of the article is a timeline on the replicability crisis. Essentially, only a few voices had claimed there was a crisis before 2011. Now, he states, we are already at "the emperor has no clothes." That's a fast cascade.

7 comments:

Sam L. said...

Who's Bethany?
It seems to me that (Miss? Doctor?) Fiske has been well, truly, and thoroughly FISKed.

Assistant Village Idiot said...

Sidebar, Graph Paper Diaries. Signs in as bsking here.

jaed said...

Hmm. The doctrine of stare decisis as applied to peer-reviewed publications? How is that in the least bit scientific?

(Looking up Fiske's CV, I note without much surprise that her appointment is joint psychology and "public affairs", and her research interests are heavy on "discrimination" and social psych. Her degrees are in Social Psychology and Social Relations.

I suppose it shouldn't be startling that someone with such a background has confused the practice of science with the assembly of canon. The mystery, then, is why someone who isn't a scientist is claiming to be a "member of the scientific community", let alone claiming the authority to set norms for science. [Let alone trying to set norms that are frankly anti-scientific. No one should presume to criticize anointed conclusions unless "curated" by... the likes of Susan Fiske. Goodness.] Sprinkling some not-too-competent statistical sparklies on a field makes it a science now?)

jaed said...

(And I am being far too snarky, especially for Sunday. ;-)

james said...

I suppose I will have to worry along somehow without Fiske's approval. I don't go in for nasty letters to tenure committees, though--that seems hard to justify.

jaed said...

That's true.

On the other hand, I wonder whether we can take her characterization of these letters as accurate, particularly since she doesn't seem to provide examples. One prof's "nasty letters" may be another prof's "he pointed out substantive problems with my research, bluntly and without decorum".

RichardJohnson said...

It has occasionally been documented that the divergence in Pub and Demo trust in "science" began around the same time that the incidence of fraud in published research increased.

I for one do not consider Social Psychology to be science.

It must be disheartening for Dr. Fiske to realize that a substantial proportion of the educated portion of the populace consider he field to be full of mumbo-jumbo. For some funny reason, low replicability of results just might have something to do with that opinion.