Schansberg: Statistical Discrimination
by Eric Schansberg, Ph.D.
I had an interesting moment with a student this summer. He emailed to ask for a religious accommodation to join his family in celebrating Eid al-Fitr — a Muslim holy day that celebrates the end of Ramadan. I’m happy to grant schedule extensions in these cases.
Until the email, I knew virtually nothing about him. I knew his name: a Western first name and an East-Asian last name. But the course was online, so I had not even met him. And it was early in the semester, so I had seen little from him in terms of performance.
When he invoked religion, culture and family to ask for the accommodation, I learned more about him and noticed that my perceptions of him changed (slightly). For one thing, he had begun to represent his family, his culture and his religion to me. And I understood that his interactions with me and his performance in my class would (slightly) impact the way I saw his name and those three groups.
As a Christian, this reminded me of the Third Commandment: not to “misuse the name of God.” Often, the commandment is reduced to profaning God’s name — for example, by using it to cuss. But the commandment is broader and more important than this one application. If I invoke God’s name and then act like an idiot, I misuse and harm God’s name. (If I invoke His name and represent Him well, then I bolster how others see God’s name and God Himself.)
In the last few years, we’ve seen many unfortunate events in the area of race. In all such cases, the harm is done by and to individuals. But there is also the broader issue of damage to the groups that the person represents.
Consider the case of Derek Chauvin, the policeman who kneeled on the neck of George Floyd in Minneapolis. Chauvin did a reprehensible thing and our perceptions of him are changed forever. But this evil also cuffed the police in general — and we’ve been living with the consequences of that for the past few months.
Or consider the case of Jessie Smollett, the actor from the television show “Empire” who perpetrated a hoax based on race and politics last year. Smollett paid two confederates to act as if he had been attacked by two white guys wearing Trump MAGA hats. After the ruse was unveiled, this evil inevitably cast a bad light on African-Americans, anti-Trumpers, and other accusations about racial discrimination.
As an economist, this reminds me of “statistical discrimination” — the idea that all of us necessarily judge people and moments by their group affiliations. We do this because information about individuals and events is (highly) imperfect and expensive to obtain. In our efforts to make the best decisions we can, with limited knowledge, we grasp at low-cost information that we believe to have predictive power. (Outside of economics, the closest concept to this is “stereotyping.”)
Chauvin and Smollett have done amazing harm to themselves, but indirect damage to the groups to which they belong. We see police differently because of what Chauvin did. We see claims about racial crimes differently because of what Smollett did. If you’re a good policeman, Chauvin has harmed you. If you deal with a true case of racism, Smollett has harmed you. When there are false charges of sexual harassment, it harms those who have valid accusations. And so on.
This is the way life works, because all of us make decisions with limited information. If the last four students you’ve hired from my business school are gold, the next graduate who applies will look relatively good. If the last four have been turkeys, that’s bad news for the next graduate. It might not be fair, but that’s life.
A lesson here is that we should hold such judgments as lightly as possible. At some point, we must make decisions. But when possible, we should try to learn more and question our assumptions as new information becomes available.
What about my student? It didn’t go well for him in the course. I don’t think it’s because he is a male, a Muslim or comes from a bad family. I think it’s because he belongs to another group of people — students who ask for delays and exceptions. They rarely do well.
It can be hoped that we do our best to enhance our knowledge, test our prior beliefs and make effective decisions. And it can be hoped we’re aware that our actions impact the perceptions of others about us and the groups to which we belong.
Eric Schansberg, Ph.D., is professor of economics at Indiana University Southeast, adjunct scholar for the Indiana Policy Review Foundation and author of “Poor Policy: How Government Harms the Poor.”