Statistics professor: no evidence for benefits from diversity

What would I do without Steve Sailer finding stuff like this for me?

Although we appreciate several things about the Duarte et al. essay, “Political Diversity Will Improve Social Psychological Science,” including its insistence that social scientists should work to minimize the impact of their political views on their research and its sensitivity to political threats to social science funding, we find their central argument unpersuasive. Contrary to the assertion of the authors, we have seen no evidence that social science fields with more politically diverse workforces have higher evidentiary standards, are better able to avoid replication failures, or generally produce better research…The authors are thus calling for major changes in policy and practice based on sheer speculation. The authors cite some evidence of the benefits of “viewpoint diversity” in collaboration, but there is a scale mismatch between these studies (of small groups) and the field-level generalizations the authors make.

Remarkable.  Intellectual diversity directly related to the field of study would seem to be the case where one would have the strongest expectation of a diversity payoff.  The case is much easier to make that various social commitments will keep social scientists honest than that female or black engineers provide mysteriously new perspectives.  And yet they claim no measurable effect.  I salute the authors’ humility in refraining from stepping beyond the data of their specific case to the vast society-wide implications of this result.

6 Responses

  1. Many historians of science would agree with Michel Foucauld that “The objective features of a phenomenon so little constrain the ways it is classified and theorized that these features can be disregarded in trying to understand why a particular classification system or scientific theory has been adopted.”

  2. No one ever claimed that leftist social scientists have especially low “evidentiary standards” or shocking levels of “replication failure” because neither of these portentous phrases is relevant to the question of research bias. There is a world of difference between fraud and bias, so a researcher who is innocent of fraud can reek to high heaven with bias. Fraud occurs after the data have been collected; bias occurs before. Bias comes in with the presuppositions that determine the questions that are asked, the way those questions are framed, and in the choice of methodology.

    There is a profound irony here. When the Marxists first claimed a place in academic departments, it was with the argument that their findings were not negated by the findings of positivist social scientists, because Marxist research began with different philosophical presuppositions. Basically, they said that methodological criteria such as evidentiary standards and replication were dependent on epistemological criteria, and that they were as rigorous as the positivists when taken on their own terms.

    Of course, once the leftists are in control, all of this postmodern pluralism goes out the window and we are back to arguing methodological questions, with leftist presuppositions taken for granted.

    I really cannot decide whether these people are stupid or dishonest when it comes to this.

  3. Yeah, the first line of defense in academia is burden of proof games. In statistical jargon, the preferred conclusion always gets to be the null hypothesis. The size of the test is always tiny. And, if you get over those two hurdles and reject the null, then the null is modified. It is not modified by re-doing an optimization. It is not modified by reasoning to the best explanation. It is modified by changing it *just enough* so that the null is no longer rejected at the absurdly small size.

    So, indeed, you don’t get to depose the theory that uniform leftism is bad for a discipline without utterly conclusive proof. Also, you don’t get to depose the theory that diversity in race, sex, and religion are gigantic benefits to a discipline without utterly conclusive proof. The fact that there is no even remotely plausible (let alone proved) theory which accommodates both of these beliefs is just irrelevant.

    Well, unless you produce a mathematical proof of the two views’ inconsistency. In that case, the result will be an industrious search for the smallest relaxation of your proof’s axioms which save the conclusion they want, followed by everyone proclaiming that they always believed the modified axiom and that nobody ever mentioned this belief before because it was so damn obvious. You dummy you.

    This is the point flogged by Kuhn, Quine, and Lakatos (though not remotely in their words, obviously). How we manage to learn stuff under so defective an epistemological paradigm is a mystery, at least to me.

  4. Diversity isn’t about diversity. To be fair, even some leftists have been willing to acknowledge this, though mostly because it causes problems for them and their pet causes.

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 )

Twitter picture

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

Facebook photo

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

Google+ photo

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

Connecting to %s

%d bloggers like this: