The difficulty of measuring ‘isms’


Lew Rockwell, founder of the Mises Institute, has an interesting observation in his recent column, “What Exactly Is Racism?” He invites us to consider how we define and measure “racism” or other isms.

If we locate or define it as a matter of the heart, we’ll find it difficult to measure of course.

And even if we look to behavior — and move beyond obvious examples — it becomes quite difficult to measure well (if that matters to those who use the term). The usual efforts to measure such things are curious and unsatisfying.

First, people often rely on simple aggregate statistics — comparing all members of group X to all members of group Y. (The most common example here is men versus women.) No other variables are held constant, and all of the differences are assumed to be caused by discrimination.

The comparisons are quite selective. For example, nobody uses this method to compare Asian-Americans to the average.

Theory or logic is ignored. Under what contexts would the market tolerate paying X a modest percentage of what they pay Y? And why aren’t labor markets assumed to be reasonably competitive in such cases? Why wouldn’t greedy folks hire a lot of X to max profit?

Second, people measure certain outcomes and not others. Here, we seem to start with a theory or story of where an ism might be and then look for anecdotal or statistical differences.

If one doesn’t imagine that an ism could exist (perhaps mixing in the supposed existence of “good intentions”), we either don’t look or ignore outcomes when they’re presented.

How we measure police violence by race is an example. How we ignore policy outcomes and policy stances against African-Americans on Social Security, minimum wage and K-12 education are other examples.

Eric Schansberg, an adjunct scholar of the Indiana Policy Review Foundation, is a professor of economics at Indiana University Southeast. Send comments to [email protected].

No posts to display