Why Affirmative Action is Justified

[redemption]

Tag with skepticism that merits more thought.  Can this all be grounded on the stereotype threat or is that merely a transmission mechanism?  Does grounding this on stereotype threat shield what lies beneath.  Could deeper analysis perhaps lead to a more targeted intervention that could address these issues.  A quick analogy before I do more thinking:

Black male life expectency is much shorter than White male life expectency in America.  In fact, as Sen points out, Black American males live shorter than Chinese and Keralan males (White American males live longer than both).  It would seem silly to me to argue that the best way to address these health shortcomings would be to cite a number of studies on vitamin deficiency to explain the difference.  That's sort of what this feels like.  I don't agree with the principles, I just think that you've giving too much centrality to what is essentially a symptom and not the problem.

Stereotype threat is a symptom of societal stereotypes, and not a symptom of URM insecurities as such. The remedy, therefore, and in the ideal, would be a society without negative stereotypes. That would eliminate the problem for this group - i.e. the highly-motivated, high-achieving URM applicant who has earned a UGPA that is sufficient to earn him/her consideration for a T14/T50 school.

Short of that, one could eliminate tests that do more than is acceptable to highlight these societal faults, and to invert these faults to make them appear to be instead the fault of the people who are negatively stereotyped.

Failing both of these, it seems to me justifiable and fair to remedy this effect via affirmative action.

My point is that there are underlying societal problems, that the current heavy reliance on the LSAT severely amplifies those problems via a well-documented process, and that schools should either not rely on the LSAT or, in the alternative, should use affirmative action as a remedy.

To do otherwise would be grossly unfair, in my view.

[Quintana]

Tag with skepticism that merits more thought.  Can this all be grounded on the stereotype threat or is that merely a transmission mechanism?  Does grounding this on stereotype threat shield what lies beneath.  Could deeper analysis perhaps lead to a more targeted intervention that could address these issues.  A quick analogy before I do more thinking:

Black male life expectency is much shorter than White male life expectency in America.  In fact, as Sen points out, Black American males live shorter than Chinese and Keralan males (White American males live longer than both).  It would seem silly to me to argue that the best way to address these health shortcomings would be to cite a number of studies on vitamin deficiency to explain the difference.  That's sort of what this feels like.  I don't agree with the principles, I just think that you've giving too much centrality to what is essentially a symptom and not the problem.

  But this particular symptom is central to the justification of AA (the topic of this thread),  regardless of the root cause of the problem.

  Incidentally, I'm fairly sure Sen advocates for SES based AA.

[H4CS]

Stereotype threat is a symptom of societal stereotypes, and not a symptom of URM insecurities as such. The remedy, therefore, and in the ideal, would be a society without negative stereotypes by race

I don't think so, Red and you seem to be aritifically stopping your analysis at this point.  Societal stereotypes are not the cause of racial inequality but are narratives constructed to blame the victims of institutional racism for their  low social standing.  The goal is not a society without negative stereotypes by race, but a society without pervasive institutional racism (which is the ultimate cause of the stereotypes).

Short of that, one could eliminate tests that do more than is acceptable to highlight these societal faults, and to invert these faults to make them appear to be instead the fault of the people who are negatively stereotyped.

Short even of that, it seems to me justifiable and fair to remedy this effect via affirmative action.

I agree with these points.  As I said before, I think that you need to ground this on something deeper than stereotype threat.  In my previous analogy, vitamin deficiency cannot be used to ground a comprehensive approach to addressing discrepensies in the expected lifespan.  Vitamin deficiencies are a symptom of nutritional shortcomings that are a symptom of deeper, structural, institutional problems.  Same goes with stereotype threat.

[redemption]

I don't think so, Red and you seem to be aritifically stopping your analysis at this point.  Societal stereotypes are not the cause of racial inequality but are narratives constructed to blame the victims of institutional racism for their  low social standing.  The goal is not a society without negative stereotypes by race, but a society without pervasive institutional racism (which is the ultimate cause of the stereotypes).

I agree with these points.  As I said before, I think that you need to ground this on something deeper than stereotype threat.  In my previous analogy, vitamin deficiency cannot be used to ground a comprehensive approach to addressing discrepensies in the expected lifespan.  Vitamin deficiencies are a symptom of nutritional shortcomings that are a symptom of deeper, structural, institutional problems.  Same goes with stereotype threat.

Okay, I think that we are talking at two different levels.

I am concentrating on the justification for AA as a remedy for the current crop of law school applicants, not as a first-best  correction for the problem of institutional racism in the United States.

Vitamin deficiencies are indeed a symptom of deeper problems, and yet you and I would nevertheless both argue for Vitamin A distribution now while those structural problems are addressed over time. Right?

[H4CS]

I don't think so, Red and you seem to be aritifically stopping your analysis at this point.  Societal stereotypes are not the cause of racial inequality but are narratives constructed to blame the victims of institutional racism for their  low social standing.  The goal is not a society without negative stereotypes by race, but a society without pervasive institutional racism (which is the ultimate cause of the stereotypes).

I agree with these points.  As I said before, I think that you need to ground this on something deeper than stereotype threat.  In my previous analogy, vitamin deficiency cannot be used to ground a comprehensive approach to addressing discrepensies in the expected lifespan.  Vitamin deficiencies are a symptom of nutritional shortcomings that are a symptom of deeper, structural, institutional problems.  Same goes with stereotype threat.

Okay, I think that we are talking at two different levels.

I am concentrating on the justification for AA as a remedy for the current crop of law school applicants, not as a first-best  correction for the problem of institutional racism in the United States.

Vitamin deficiencies are indeed a symptom of deeper problems, and yet you and I would nevertheless both argue for Vitamin A distribution now while those structural problems are addressed over time. Right?

Yes, but then you need something before your first point.

0.  There continues to exist systematic, multifacited, and pervasive racism within American society that is manifested in a myriad of different ways.

Then the first point can become one of a number of quantitative theories by which point 0 can be measured.  My concern is that without point 0, this whole justification seems vulnerable.  What would you say if studies proved that incoming college students who undertook the same experiements 5 years later showed that the stereotype threat was diminishing over time?  Would that alone be a reason to reduce AA?  I should hope not, unless URM performance on the LSAT correspondingly improved.

[magnumalv]

...tag...

[redemption]

What would you say if studies proved that incoming college students who undertook the same experiements 5 years later showed that the stereotype threat was diminishing over time?  Would that alone be a reason to reduce AA?  I should hope not, unless URM performance on the LSAT correspondingly improved.

If the stereotype threat diminishes over time, then this justification will wither with it. LSAT scores (after controlling for SES etc) will be equivalent, GPAs will be equivalent, and the need for affirmative action that systematically attempts to counteract a test-gap score will also vanish.

The weakness, as I see it, is just the opposite: that the test-gap sore and stereotype threat have remained static over a generation. This suggests that affirmative action does not address the root societal problems, and may actually subsidize the continued existence of the systematic racism in our institutions, by not making the latter expensive enough, embarrassing enough, to change.

Nevertheless, in terms of the frame that I am focused on (is it unfair for URMs with lower LSATs to be admitted to law school?) I maintain that it is not unfair, and that the opposite it true -- that it would be very unfair if they weren't.

[H4CS]

What would you say if studies proved that incoming college students who undertook the same experiements 5 years later showed that the stereotype threat was diminishing over time?  Would that alone be a reason to reduce AA?  I should hope not, unless URM performance on the LSAT correspondingly improved.

If the stereotype threat diminishes over time, then this justification will wither with it. LSAT scores (after controlling for SES etc) will be equivalent, GPAs will be equivalent, and the need for affirmative action that systematically attempts to counteract a test-gap score will also vanish.

Re-read the paragraph you quoted, you just fell for the trap that I set and then explained.  I believe in the existence of the stereotype threat as one of a number of transmission mechanisms whereby racial inequality is perpetuated.  It is possible for the stereotype to diminish with inequality diminishing, in fact, it could increase if the other mechanisms were strengthened.  AA should only decrease when racial inequality decreases, not just when one of its transmission mechanisms decreases.

The weakness, as I see it, is just the opposite: that the test-gap sore and stereotype threat have remained static over a generation. This suggests that affirmative action does not address the root societal problems, and may actually subsidize the continued existence of the systematic racism in our institutions, by not making the latter expensive enough, embarrassing enough, to change.

Stereotype threat hasn't changed because it is predicated on the very real existence of a performance gap on the LSAT.  There was a gap thirty years ago and there is still one today, why should we expect stereotype threat to have vanished when there has been no concerted effort to improve the test?  I think the answer is maintaining AA while addressing the structural problems with the LSAT that would work to address both short-term and long-term issues.

Nevertheless, in terms of the frame that I am focused on (is it unfair for URMs with lower LSATs to be admitted to law school?) I maintain that it is not unfair, and that the opposite it true -- that it would be very unfair if they weren't.

Yeah, but you don't need stereotype threat to argue that, it's just elementary.  If there's a performance gap by race on the LSAT after reasonable controlling for other variables, of course corrective action needs to be taken, regardless of the reason for the gap.

[dun]

One question: why do they score lower on standardized tests?  I don't buy the confidence thing.

[redemption]

#1. I believe in the existence of the stereotype threat as one of a number of transmission mechanisms whereby racial inequality is perpetuated.  It is possible for the stereotype to diminish with inequality diminishing, in fact, it could increase if the other mechanisms were strengthened.  AA should only decrease when racial inequality decreases, not just when one of its transmission mechanisms decreases.

#2. Stereotype threat hasn't changed because it is predicated on the very real existence of a performance gap on the LSAT.  There was a gap thirty years ago and there is still one today, why should we expect stereotype threat to have vanished when there has been no concerted effort to improve the test? I think the answer is maintaining AA while addressing the structural problems with the LSAT that would work to address both short-term and long-term issues.

#3. Yeah, but you don't need stereotype threat to argue that, it's just elementary.  If there's a performance gap by race on the LSAT after reasonable controlling for other variables, of course corrective action needs to be taken, regardless of the reason for the gap.

I agree that stereotype threat is a mechanism of transmission. I agree that it is caused by structural racism. I agree that steroetype threat is unlikely to diminish unless structural racism - both in the design of the test and in the wider society - are eliminated. And I concede that the existence of AA actually diminishes the chances that the LSAT will be redesigned -- that is AA's great weakness, in my view.

I take issue only with point #3. It is not obvious to everyone that "if there's a performance gap by race on the LSAT after reasonable controlling for other variables, of course corrective action needs to be taken, regardless of the reason for the gap".

Some/many (perhaps most) people believe that the reasons for the performance gap are genetic, or due to some cultural fault of URMs. This line of reasoning leads to the conclusion that URMs are therefore "mismatched" at he law schools that they attend, and that they are "unqualified", and "lesser".

The main stress of my argument is that this is simply not plausible in the light of the evidence. I am saying that the disparity in index scores (after adjusting for all the usual variables -- SES etc) is due to the stereotype threat, and that the stereotype threat is - on average - felt most by the most able and qualified and motivated individuals within the URM groups.