Skip to content

Misunderstandings of Affirmative Action: Supreme Court Edition

I’m poorly qualified to pronounce on Affirmative Action as a general topic. But I do see misunderstandings that overlap with one of my favorite topics: errors in perceiving probabilities.

Before you say BORING, let me try to convince you that this is at the heart of the AA debate.

The #1 question about Elena Kagan is “did she get nominated because she’s a woman?” There is no way to answer this for one unique case, but we can ask for women in general, do they benefit a lot from AA?

As usual there are two probabilities that are often confused with each other. The first is that IF there is a big promotion because of AA for women, THEN the beneficiary will be a woman. This probability is 100% by definition.

The second probability is that IF you are a woman, THEN you benefit a lot from AA. I leave it to the experts to figure out this exactly, but you don’t have to be an expert to know this is a lot lower than 100%. If we are talking about AA primarily affecting promotions to a high position, then this is intrinsically rare for both men and women. It follows that this 2nd probability would be very low, because few women are affected by promotions to highly selective positions.

Yet because we often confuse reverse conditional probabilities (i.e. the two probabilities above), as discussed in many previous posts, we think a lot more women are benefiting from AA (which is perceived to be unfair by many) than really are. Conversely, very few men are really being affected by reverse discrimination (again because highly selective promotions are rare to begin with).

So bottom line of this post: I get really annoyed to see the accomplishments of my female professional friends and colleagues stigmatized because they are seen, most of the time incorrectly, as Affirmative Action babies.

PS One interesting aside relevant for today’s previous post. If Kagan is confirmed, the Supreme Court will consist of 6 Catholics and 3 Jews, none of whom would have been considered to fit the core definition of “white” before the mid-19th century in America.

This entry was posted in Cognitive biases. Bookmark the permalink. Follow any comments here with the RSS feed for this post. Both comments and trackbacks are currently closed.

11 Comments

  1. What about affirmative action for potential judges who did not go to Ivy League schools? Since 1900, 31 of the 53 Supreme Court Justices appointed attended an Ivy League institution (either as an undergraduate or a post graduate) — that includes Stevens’s eight fellow justices. (Note: I went to an Ivy League school)

    Posted May 10, 2010 at 4:11 pm | Permalink
  2. Comment from Twitter wrote:

    how can you argue only highly selective positions use AA? *ALL* jobs do, which is what pisses ppl off.

    Posted May 10, 2010 at 5:36 pm | Permalink
  3. Bill Easterly wrote:

    Dear Comment from Twitter:

    Well, I’ll let the experts sort out how much affirmative action applies at different levels of selectivity. Another principle this blog and my other writing suggests is that people respond to public pressures (such as for Affirmative Action) with highly visible but token efforts (such as corporations giving tiny amounts for global poverty and then branding themselves as “socially responsible”). Certainly a lot of the public attention given to affirmative action is for high profile jobs that are intrinsically very selective and rare because not that many jobs can be high profile.

    Posted May 10, 2010 at 5:41 pm | Permalink
  4. RJS wrote:

    So long as affirmative action and quotas exist, every member of every group that is a benificiary will automatically be suspect.

    You can engage in all the magical thinking you want that they aren’t, but they will be and in the back of their own minds they will never really know for sure if they made it because of their own hard work and effort or because they receive preferential treatment.

    Posted May 10, 2010 at 5:52 pm | Permalink
  5. Tim Ogden wrote:

    I think you’re way off in your timing of when Catholics and Jews became part of the “white” majority in this country. Keep in mind that one of the major justifications for the Spanish American War was Christianizing the Catholic Spanish colony of the Philippines.

    And keep in mind that we turned away boatloads of Jewish refugees during WWII.

    Posted May 10, 2010 at 11:16 pm | Permalink
  6. Eric Butter wrote:

    Great point re: highly visible but token efforts. I hope someday soon we can justify appointments like this because of the unique perspectives they bring to work, instead of sex/race/filling a quota …

    Posted May 11, 2010 at 1:09 am | Permalink
  7. William Easterly wrote:

    Tim, please read the previous post on “Before I was white”. This is probably closer to the timing you have in mind and differentiates between Catholics and Jews. The one sentence aside is not detailed enough to capture the more complex story. Best, Bill

    Posted May 11, 2010 at 7:54 am | Permalink
  8. Ray Sawhill wrote:

    Judging from the makeup of the Supreme Court, it sounds like we should be starting to consider a little Affirmative Action for white Protestants.

    Seconding Paul’s concern about Ivy and non-Ivy juges too. My own theory is that we’d all be better off if the Ivy schools were nuked into ashes.

    Posted May 11, 2010 at 11:15 am | Permalink
  9. Adam Baker wrote:

    I agree that the two probabilities you identify are distinct, but I don’t think that the latter is what bothers opponents of affirmative action. If men were specifically excluded from the (serious) nomination process, I would be bothered, but not because of personal interest. I don’t know any male judges. (I also happen to live overseas in a strongly masculinist society, so I am really in no danger.) What would be bothersome would be the heavy-handed effort to redress past wrongs by artificially reducing the candidate pool of Supreme Court justices.

    Posted May 11, 2010 at 1:07 pm | Permalink
  10. Steve wrote:

    In some university contexts the conditional P(benefited from affirmative action | minority) is very large.

    Consider that CalTech is only1% black while MIT is about 9% black. CalTech doesn’t have affirmative action while MIT is proud to. That suggests P(AA baby|black) = .89. Now I think we still don’t do enough to recruit smart black people here–there are a TON of good candidates who just never apply–so I’m fine with AA. But the facts are the facts. (The case is similar for girls–most engineering schools are < 30% female but MIT is ~50%, thus P(AA baby | girl at MIT) ~= .4. I've never met a girl who believes AA-for-girls exists though.)

    A lot of people are going to play this probability game to dodge the issue of whether AA is good or not. We should have a better case for why AA is good than that it doesn't effect most people, even minorities. (Indeed, its ineffectiveness is a cause for wondering how valuable it is.)

    Posted May 12, 2010 at 8:45 pm | Permalink
  11. Mark @ Israel wrote:

    I don’t think Kagan was nominated because she is a woman. It could probably be for other reasons. But those reasons are Obama’s reasons. I am just wondering: how come she was nominated to be an associate justice in the Supreme Court when she has not been a courtroom judge ever since?

    Posted May 16, 2010 at 12:12 pm | Permalink

2 Trackbacks

  1. [...] This post was mentioned on Twitter by William Easterly, Eric Butter, Dan, Eric Butter, Thabo Mophiring and others. Thabo Mophiring said: RT @bill_easterly: Misunderstandings of Affirmative Action: Supreme Court Elena Kagan edition http://bit.ly/dlzqSd [...]

  2. By Amusing point on May 11, 2010 at 5:19 am

    [...] 11th, 2010 · No Comments If Kagan is confirmed, the Supreme Court will consist of 6 Catholics and 3 Jews, none of whom would have [...]