The Modeler's Toy

Via Slashdot comes an absurd rant from Scott E. Page in which he rails against the very idea of meritocracy:

The multidimensional or layered character of complex problems also undermines the principle of meritocracy: The idea that the ‘best person’ should be hired. There is no best person. When putting together an oncological research team, a biotech company such as Gilead or Genentech would not construct a multiple-choice test and hire the top scorers, or hire people whose resumes score highest according to some performance criteria. Instead, they would seek diversity. They would build a team of people who bring diverse knowledge bases, tools and analytic skills. That team would more likely than not include mathematicians (though not logicians such as Griffeath). And the mathematicians would likely study dynamical systems and differential equations.

Page here actually argues that there can be no such thing as meritocracy when problem domains are so broad and complex; "no test or criteria applied to individuals will produce the best team", he claims.

But the problems with this criticism should be obvious. First, Page injects his own bias ahead of hiring managers, who he claims have a flawed understanding of what it is that needs to be done by a particular employee. Yet, if hiring managers don't know what a given position requires, how does Page, who is removed from the process entirely, get to claim he knows better than they?

The second problem is the bait-and-switch nature of his definition of "meritocracy". Page has gained notoriety and accolades for his allegedly ground-breaking work showing that there can be no such thing, and in fact that diverse teams produce better results than purported "meritocracies".

Q. The term “diversity” has become a code word for inclusion of racial, ethnic and sexual minorities. Is that what you’re talking about?
A. I mean differences in how people think. Two people can look quite different and think similarly. Having said that, there’s certainly a lot of evidence that people’s identity groups — ethnic, racial, sexual, age — matter when it comes to diversity in thinking.

Page quickly shifts gears from claims of a quantitative judgment to a qualitative one (emboldening mine):

Q. In your book, you advocate affirmative action, an unpopular social policy these days. What’s your argument?

A. That it’s a flat-out good because, as I said earlier, it makes everything we do more powerful.

For a while, I chaired admissions in the graduate political science department at the University of Michigan. We didn’t just look at high test scores. We looked at things like whether an applicant had worked with Teach for America. We wanted to bring in people who had experiences and modes of thinking that would improve everyone else.

At a university, people learn from each other as well as their professors. Another suburban kid who was raised to score high on tests doesn’t add all that much to the mix.

Page elsewhere points to a paper he coauthored in 2004 as alleged proof of this theory, using absurdly simplistic models of problem solvers in which diversity trumps competence. Ultimately, it's Page's bait-and-switch that makes this so infuriating: he claims diversity of knowledge is important, but argues for a sort of watered-down quota system based on diversity of identity. Unsurprisingly, Page's paper is cited 753 times, according to Google Scholar. It is a kind of grasping at straws for the identity politics crowd, who have precious little in the way of proof to cling to in their voodoo cult.