A few notes on Regnerus, Sherkat, science . . .
At this point I will acknowledge a contradiction found in my previous post on Sherkat and the Regnerus study: despite my rhetoric, I agree with Sherkat, in this case. Regnerus’s study effectively equates to “science,” whatever that is, from the point of view of the editor and publisher, and so retraction isn’t called for.
First, Regnerus’s paper was peer-reviewed, in contrast to the recently discredited Reinhart-Rogoff paper, which wasn’t.
Second, science is wrong all the time. Being ok with being wrong is kind of the point. Studies don’t lead to final knowledge. They lead to re-examination and further re-examination. Being wrong is central to the pursuit.
Finally, and more generally, if I were to take the role of the editor, I could think of few circumstances that would lead me to “retract.”
So, while I would not retract, I would do something else: engage with the study’s flaws. The paper’s critics — including Sherkat’s audit — have shown without a doubt that the paper’s research is highly flawed and the language not supported by the data. The only way to salvage the findings and the conclusions is for a do-over. As editor, I would invite further contentious debate.
To this extent, why Regnerus isn’t personally and publicly acknowledging the flaws and calling for a re-examination of the study’s findings (from the beginning) is the question that intrigues me most.
Is he really not convinced the paper has flaws?
My position has been and still is that Regnerus should pro-actively admit the flaws, and set out to do a better study. Basically, he should interrogate his own findings. He should do this for two reasons. One, it would be better public relations. Two, it would be better science.
The best reason I can think for Regnerus not to admit the flaws and not set himself to work on the problem, is that, for reasons of values or interests, he prefers the audience of the conservative movement to the audience of sociology. I am not saying a skilled operator couldn’t have both audiences. I am saying such poorly executed science can’t have the sociology audience.
Maybe Regnerus has already made his decision. Maybe the paper was quite successful, in his mind.