Monday, May 18, 2015

It's not that they are liberal, its that they are unscientific

It is interesting to treat Fixing the Problem of Liberal Bias in Social Psychology by Piercarlo Valdesolo as a serious argument.

He starts with an agreed premise.
Social psychology is overwhelmingly composed of liberals (around 85%).
He then goes on
The question of why this is the case, and whether it presents a problem for the field, is more controversial.
Where I think Valdesolo goes astray is his framing of proposed solutions. He creates a strawman. He argues as if the only proposal is for political ideological affirmative action. There are certainly a few voices calling for exactly that. But they are rare.

Valdesolo argues that affirmative action for conservatives is an inappropriate response because two wrongs don't make a right and two biases don't average out to the truth. Both statements are true and irrelevant. Most critics don't wish for an affirmative action program for conservatives. What they want is greater adherence to the scientific method in social psychology. Right now 70-90% of all social psychology experiments cannot be replicated, suggesting that ideological biases and assumptions are creeping in to research.

If all social psychologist were to suddenly adopt the scientific method, it wouldn't matter much what their biases might be. That is partly what the scientific method is for. Revealing replicable truth regardless of assumptions and expectations.

By attacking a strawman argument of his own making and not addressing the real issues for critics, Valdesolo tends to undermine his position. It is inescapably true as well that Valdesolo's preferred solution (let's us fix our own profession ourselves without intervening and forcing us to accept more diversity of viewpoint) is the one that is both easiest on the existing practitioners and the one least likely to make any difference. There have been social psychologists of one leaning or another running solid scientific studies for many decades. Their results are not contingent on their beliefs but on robust science.

The fact that most social psychologists have chosen not to follow that established pattern of behavior is a testament that leaving them alone to fix the acknowledged problem is a likely indication that the problem won't actually get solved.

By using strawman arguments and arriving at solutions that are entirely self-serving and yet unconvincing, Valdesolo illustrates what is actually wrong with the profession. It's not that they are liberal, its that they are unscientific. A fact that, in one form or another, keeps coming up in the comments section.

No comments:

Post a Comment