I read the Bell curve many years ago, and cited it in an article I wrote. It definitely had a particular agenda which was arguably racist. He wanted to claim that affirmative action programs had to recognize that genetic differences in IQs would always ensure that Black people on the average would not do as well as white people on the a…
I read the Bell curve many years ago, and cited it in an article I wrote. It definitely had a particular agenda which was arguably racist. He wanted to claim that affirmative action programs had to recognize that genetic differences in IQs would always ensure that Black people on the average would not do as well as white people on the average, even though there were plenty of blacks who were smarter than most whites. There's lots of other scientific and reasonable looking stuff in the rest of the book, mostly provided by his co-author. But all that is just window dressing to give respectability to his anti-affirmative action agenda. Even the scientific stuff relies heavily on outdated science. Most Modern cognitive scientists deny the existence of so-called general intelligence, which IQ tests are supposedly measuring. Murray's co-author was an old man fighting to preserve an outdated idea, and he died shortly after the book came out. An example of what Thomas Kuhn called scientific progress occurring funeral by funeral. Murphy was trying to apply that outdated idea to justify an attack on affirmative action.
"He wanted to claim that affirmative action programs had to recognize that genetic differences in IQs would always ensure that Black people on the average would not do as well as white people on the average, even though there were plenty of blacks who were smarter than most whites."
Yeah, Murray still has a bee in his bonnet about affirmative action. He talks about it again in the conversation with Coleman Hughes I linked during the conversation. I'm genuinely surprised (and gently amused) he doesn't know that affirmative action has benefited white women more than any other demographic.
> "I'm genuinely surprised (and gently amused) he doesn't know that affirmative action has benefited white women more than any other demographic."
Steve, have you actually looked at a solid scientific analysis which supports that, or are you repeating something we have all heard as common knowledge? (A scientific analysis attempts to identify and control for confounding factors and uses solid statistics rather than just citing a few naive percentages, as a journalist or polemicist might, and qualifies its conclusions so as not to hyperbolically over-reach the data).
Background. It's not easy to tell who has benefitted from AA. Just think about it - there are no records kept about who got a job or promotion due to AA and who got one due to pure merit. This makes it hard to accurately discern the relative benefits - nobody should sound very confident, because even the best estimates are uncertain.
So what is often done instead is just to measure all relative employment "gains" by different groups - which includes people being employed due to AA, along with people whose employment did not so involve AA, without disentangling the two.
So for example, one could say that women went from X% of employees to Y% in some field, during a time period in which Blacks went from Z% to W% - and assume that 100% of that change was due to affirmative action, and that there was 0% hiring of women due to pure merit. That's an extremely questionable assumption, as I'll try to explain.
During the typical timespans used, the society underwent a major shift in regard to women's roles and expectations. There was a major influx of women into the job market - and in particular of women with good educations. During the typical time periods, women came to outnumber men in college (now about 3:2), acquiring many skills in demand. Few can rationally doubt that women's portion of many job markets would have dramatically increased even absent any AA programs; we simply cannot attribute every woman hired during that period as "a diversity hire", but many comparisons do exactly that. (We should also not attribute every Black hired as due to AA, obviously, but according to the stats, women were on average better qualified to get jobs without needing affirmative action, so they benefitted less from the existence of AA).
In some analyses, the most predictive factor in the number of women hired, is the number of women who applied; and the number of women who applied for many jobs and admissions went way up for external reasons, with or without AA.
Some example quotes to illustrate these points:
> "The effects of affirmative action on white women’s success in the job market are difficult to assess because the period of implementation of affirmative action in the 1970s and 1980s coincided with the rapid increase of women in the labor force (Reskin, 1998). Throughout the 1980s and 1990s, white women’s progress in the labor market and increased earnings seemed to be due to better education and more work experience, factors unrelated to affirmative action (O’Neill & O’Neill, 2000)."
> "The results [of this study] indicate that affirmative action was not significantly related to any level of white female employment when controlling for other variables. Similarly, even simple bivariate correlations between the measure of affirmative action and white female employment range from −.03 to −.13, very low and statistically insignificant correlations. Clearly affirmative action has no impact on women’s employment."
> "So why haven’t white women been helped by affirmative action? The success of white
female workers suggests they need no help. White women, compared with most blacks and
Latinos, have greater education credentials and higher levels of required job skills, both of which make them more qualified in today’s job market. Moreover, the emphasis of affirmative action historically has been on blacks more so than white women. The Equal Employment Opportunity Commission (EEOC), the primary federal enforcement agency for affirmative action, has not perceived that sex discrimination is its primary mission, especially in the South"
By contrast, AA did help Blacks, as it was expected to:
> "Affirmative action is one factor that has assisted African Americans in job competition
with women. A previous study supports this finding in that employer support for affirmative
action had a positive, significant relationship with black employment, particularly at higher
* White women got more new jobs than Blacks during the study period
* White women got more new jobs than Blacks *due to AA* during the study period
But many polemical arguments do exactly that.
This is just one study, which I cite to give some perspective on the variables which must be controlled for in a serious study. I do not present this study as definitive, just one example. But any article in, say, VOX, about this topic which uses raw statistics without this kind of nuance is going to be very scientifically weak, aiming to support a pre-determined conclusion by cherry picking numbers intended to support the journalists preferred narrative.
If you have a more definitive study, which compares the number of women or white women who have materially benefitted from Affirmative Action compared to the number of Blacks, while controlling for the confounding variables, I would be genuinely interested. I don't have any vested interest in either answer, I have no position to defend on this one. I'll be glad to agree that AA benefits white women more than Blacks or Latinos - if there is solid evidence. Or to follow evidence in the other direction.
I'm just asking if you have examined the evidence on all sides for yourself about the relative benefits, or just assumed - as I used to - that it must be true because so many people keep telling us that. And if your assertion is evidence based, I would be very interested in some links to the sources you found best analyzed and most persuasive.
To be clear, I am not asserting either way, yet. In my own mind, I put the assertion that white women are the biggest beneficiaries of Affirmative Action into my mental category "widely asserted by activists but academically disputed; and I have not yet seen enough evidence to definitively support or refute". Neither proven nor disproven (within my limited investigation to date), but untrustworthy so I will not repeat that assertion until I have better evidence. Just like I try not to repeat something which strikes me as a likely urban legend, until I have better evidence. But I'd be glad to have more evidence in either direction if you or anyone else has some to offer.
If nobody has any evidence to back up that assertion, then we should not be surprised and gently amused that "he" (Coleman or Murray?) doesn't know something which may not even be true; if it's not true, their not "knowing" it is a good thing. I used to think that I too "knew" that white women were the largest beneficiaries, but it later turned out that all I really *knew* was only that I had often heard it asserted and so provisionally assumed it must be true. Over time, I gradually check out some of those assumptions and sometimes reverse, or at least question, my former beliefs.
I might rephrase it to "causes some well intended strategies for social justice to be ineffective or counterproductive".
"Social justice" could be a good thing(depending on how it's defined), but many of the current strategies nominally aimed at creating more of it, have become dogmatic sacred cows, and have become detached from reality feedback through ideological justification rather than rational merit - and as such may move society into dysfuntionality rather than producing the supposed outcome. Conformity with the groupthink "lest ye be cast out" become more important than real world effectiveness. And you have nailed two of the mechanisms involved in the unanticipated and undesired result.
An example: DEI trainings that have no effect when studied, or which actually increase tensions and decrease trust and cooperation. But they do conform to the latest ideological "shoulds" and framings, and continuing to keep up with the shifting political fads is more important to the purveyors of such trainings, than producing measurable positive results in the real world. So the trainers focus more on things like "go through our materials and change every instance of 'preferred pronouns' to just 'pronouns', because the former has been criticized and is no longer kosher among activists", than on "are we producing more effective and collaborative teams which incorporate diverse viewpoints?"
I'm old enough to bear witness to the good that affirmative action has done. It did have a down side in that even people who were trying to be non-racist often wondered, is this person really competent or an incompetent affirmative action token? People will generally call that racism, but for the years between initial programs and black professionals proving their worth their were such questions? We now call those doubts micro aggressions. That is not trivial. Just as prescription medicine has a long page of possible adverse side effects, good programs also have adverse side effects that cannot be completely ignored.
Actually, the Bell Curve used multiple regression (the appropriate method) and data from the National Longitudinal Survey of z Youth (the best available data). They included all of their data results. That looked good and fished me in. While I have done more number crunching than the average person, I am not a statistician.
What did I miss, they actually told us but I failed to notice the significance. They wrote, "In the text, we do not refer to the usual measure of goodness of fit for multiple regression, R2 (R squared), but they are presented here (an appendix) for the cross sectional analysis." As Stephen Gould points out in "The Mismeasure of Man", putting it in an appendix that most readers won't study obfuscates the weakness of the associations. Squaring a number less than one makes it smaller, in this case so small that even non-statisticians would notice.
They did not lie and they were transparent in their methods. But this was a disingenuous way of making a mountain out of a mole hill case for the premises stated.
I find Gould's critique to be a proper one, better than just calling Murray a racist as a case against his book. I'm not a fan of name calling, one of the things that Steve talks about in this commentary.
I read the Bell curve many years ago, and cited it in an article I wrote. It definitely had a particular agenda which was arguably racist. He wanted to claim that affirmative action programs had to recognize that genetic differences in IQs would always ensure that Black people on the average would not do as well as white people on the average, even though there were plenty of blacks who were smarter than most whites. There's lots of other scientific and reasonable looking stuff in the rest of the book, mostly provided by his co-author. But all that is just window dressing to give respectability to his anti-affirmative action agenda. Even the scientific stuff relies heavily on outdated science. Most Modern cognitive scientists deny the existence of so-called general intelligence, which IQ tests are supposedly measuring. Murray's co-author was an old man fighting to preserve an outdated idea, and he died shortly after the book came out. An example of what Thomas Kuhn called scientific progress occurring funeral by funeral. Murphy was trying to apply that outdated idea to justify an attack on affirmative action.
"He wanted to claim that affirmative action programs had to recognize that genetic differences in IQs would always ensure that Black people on the average would not do as well as white people on the average, even though there were plenty of blacks who were smarter than most whites."
Yeah, Murray still has a bee in his bonnet about affirmative action. He talks about it again in the conversation with Coleman Hughes I linked during the conversation. I'm genuinely surprised (and gently amused) he doesn't know that affirmative action has benefited white women more than any other demographic.
> "I'm genuinely surprised (and gently amused) he doesn't know that affirmative action has benefited white women more than any other demographic."
Steve, have you actually looked at a solid scientific analysis which supports that, or are you repeating something we have all heard as common knowledge? (A scientific analysis attempts to identify and control for confounding factors and uses solid statistics rather than just citing a few naive percentages, as a journalist or polemicist might, and qualifies its conclusions so as not to hyperbolically over-reach the data).
Background. It's not easy to tell who has benefitted from AA. Just think about it - there are no records kept about who got a job or promotion due to AA and who got one due to pure merit. This makes it hard to accurately discern the relative benefits - nobody should sound very confident, because even the best estimates are uncertain.
So what is often done instead is just to measure all relative employment "gains" by different groups - which includes people being employed due to AA, along with people whose employment did not so involve AA, without disentangling the two.
So for example, one could say that women went from X% of employees to Y% in some field, during a time period in which Blacks went from Z% to W% - and assume that 100% of that change was due to affirmative action, and that there was 0% hiring of women due to pure merit. That's an extremely questionable assumption, as I'll try to explain.
During the typical timespans used, the society underwent a major shift in regard to women's roles and expectations. There was a major influx of women into the job market - and in particular of women with good educations. During the typical time periods, women came to outnumber men in college (now about 3:2), acquiring many skills in demand. Few can rationally doubt that women's portion of many job markets would have dramatically increased even absent any AA programs; we simply cannot attribute every woman hired during that period as "a diversity hire", but many comparisons do exactly that. (We should also not attribute every Black hired as due to AA, obviously, but according to the stats, women were on average better qualified to get jobs without needing affirmative action, so they benefitted less from the existence of AA).
In some analyses, the most predictive factor in the number of women hired, is the number of women who applied; and the number of women who applied for many jobs and admissions went way up for external reasons, with or without AA.
Some example quotes to illustrate these points:
> "The effects of affirmative action on white women’s success in the job market are difficult to assess because the period of implementation of affirmative action in the 1970s and 1980s coincided with the rapid increase of women in the labor force (Reskin, 1998). Throughout the 1980s and 1990s, white women’s progress in the labor market and increased earnings seemed to be due to better education and more work experience, factors unrelated to affirmative action (O’Neill & O’Neill, 2000)."
> "The results [of this study] indicate that affirmative action was not significantly related to any level of white female employment when controlling for other variables. Similarly, even simple bivariate correlations between the measure of affirmative action and white female employment range from −.03 to −.13, very low and statistically insignificant correlations. Clearly affirmative action has no impact on women’s employment."
> "So why haven’t white women been helped by affirmative action? The success of white
female workers suggests they need no help. White women, compared with most blacks and
Latinos, have greater education credentials and higher levels of required job skills, both of which make them more qualified in today’s job market. Moreover, the emphasis of affirmative action historically has been on blacks more so than white women. The Equal Employment Opportunity Commission (EEOC), the primary federal enforcement agency for affirmative action, has not perceived that sex discrimination is its primary mission, especially in the South"
By contrast, AA did help Blacks, as it was expected to:
> "Affirmative action is one factor that has assisted African Americans in job competition
with women. A previous study supports this finding in that employer support for affirmative
action had a positive, significant relationship with black employment, particularly at higher
job levels (Button & Rienzo, 2003)."
Source:
https://spia.uga.edu/faculty_pages/rbakker/pdfs/affirmativeAction.pdf
In short - we must not conflate these two:
* White women got more new jobs than Blacks during the study period
* White women got more new jobs than Blacks *due to AA* during the study period
But many polemical arguments do exactly that.
This is just one study, which I cite to give some perspective on the variables which must be controlled for in a serious study. I do not present this study as definitive, just one example. But any article in, say, VOX, about this topic which uses raw statistics without this kind of nuance is going to be very scientifically weak, aiming to support a pre-determined conclusion by cherry picking numbers intended to support the journalists preferred narrative.
If you have a more definitive study, which compares the number of women or white women who have materially benefitted from Affirmative Action compared to the number of Blacks, while controlling for the confounding variables, I would be genuinely interested. I don't have any vested interest in either answer, I have no position to defend on this one. I'll be glad to agree that AA benefits white women more than Blacks or Latinos - if there is solid evidence. Or to follow evidence in the other direction.
I'm just asking if you have examined the evidence on all sides for yourself about the relative benefits, or just assumed - as I used to - that it must be true because so many people keep telling us that. And if your assertion is evidence based, I would be very interested in some links to the sources you found best analyzed and most persuasive.
To be clear, I am not asserting either way, yet. In my own mind, I put the assertion that white women are the biggest beneficiaries of Affirmative Action into my mental category "widely asserted by activists but academically disputed; and I have not yet seen enough evidence to definitively support or refute". Neither proven nor disproven (within my limited investigation to date), but untrustworthy so I will not repeat that assertion until I have better evidence. Just like I try not to repeat something which strikes me as a likely urban legend, until I have better evidence. But I'd be glad to have more evidence in either direction if you or anyone else has some to offer.
If nobody has any evidence to back up that assertion, then we should not be surprised and gently amused that "he" (Coleman or Murray?) doesn't know something which may not even be true; if it's not true, their not "knowing" it is a good thing. I used to think that I too "knew" that white women were the largest beneficiaries, but it later turned out that all I really *knew* was only that I had often heard it asserted and so provisionally assumed it must be true. Over time, I gradually check out some of those assumptions and sometimes reverse, or at least question, my former beliefs.
A collision of statistics and causation vs correlation that brings ruin to social justice efforts.
I get it.
I might rephrase it to "causes some well intended strategies for social justice to be ineffective or counterproductive".
"Social justice" could be a good thing(depending on how it's defined), but many of the current strategies nominally aimed at creating more of it, have become dogmatic sacred cows, and have become detached from reality feedback through ideological justification rather than rational merit - and as such may move society into dysfuntionality rather than producing the supposed outcome. Conformity with the groupthink "lest ye be cast out" become more important than real world effectiveness. And you have nailed two of the mechanisms involved in the unanticipated and undesired result.
An example: DEI trainings that have no effect when studied, or which actually increase tensions and decrease trust and cooperation. But they do conform to the latest ideological "shoulds" and framings, and continuing to keep up with the shifting political fads is more important to the purveyors of such trainings, than producing measurable positive results in the real world. So the trainers focus more on things like "go through our materials and change every instance of 'preferred pronouns' to just 'pronouns', because the former has been criticized and is no longer kosher among activists", than on "are we producing more effective and collaborative teams which incorporate diverse viewpoints?"
You said it in fewer words, tho.
I'm old enough to bear witness to the good that affirmative action has done. It did have a down side in that even people who were trying to be non-racist often wondered, is this person really competent or an incompetent affirmative action token? People will generally call that racism, but for the years between initial programs and black professionals proving their worth their were such questions? We now call those doubts micro aggressions. That is not trivial. Just as prescription medicine has a long page of possible adverse side effects, good programs also have adverse side effects that cannot be completely ignored.
Actually, the Bell Curve used multiple regression (the appropriate method) and data from the National Longitudinal Survey of z Youth (the best available data). They included all of their data results. That looked good and fished me in. While I have done more number crunching than the average person, I am not a statistician.
What did I miss, they actually told us but I failed to notice the significance. They wrote, "In the text, we do not refer to the usual measure of goodness of fit for multiple regression, R2 (R squared), but they are presented here (an appendix) for the cross sectional analysis." As Stephen Gould points out in "The Mismeasure of Man", putting it in an appendix that most readers won't study obfuscates the weakness of the associations. Squaring a number less than one makes it smaller, in this case so small that even non-statisticians would notice.
They did not lie and they were transparent in their methods. But this was a disingenuous way of making a mountain out of a mole hill case for the premises stated.
I find Gould's critique to be a proper one, better than just calling Murray a racist as a case against his book. I'm not a fan of name calling, one of the things that Steve talks about in this commentary.