All this arrant DEI nonsense puts me in mind of a Saul Bellow quote: "A great deal of intelligence can be invested in ignorance when the need for illusion is deep" In these few words he hit the nail on the head about the nature of our wokeified academia in the Western world. Anyone who has read Heather Mac Donald's comprehensive study of campus DEI The Diversity Delusion could be left in no doubt about who is really discriminated against in universities. But of course the kind of people who need to be put straight on all this, won't ever read anything that doesn't indulge their infantile virtue-signalling prejudices. https://grahamcunningham.substack.com/p/how-diversity-narrows-the-mind
I really like that you are taking the time to really analyze the findings of this article. As a former professor, I know how often subpar articles get published if they enhance the ideological agenda of the academic Left. This, unfortunately, gives the professor who published the article enhanced status and encourages other professors to do the same. Worse, it then gets amplified by the mainstream media.
We need to fight this process by showing how fraudulent these studies are. Hopefully, if professors learn that the more Woke their conclusions, the more likely it is to get torn apart in public, then this creates somewhat of a counter-balance in incentives. This will not stop the process, but it might reduce the negative impact on the rest of society.
I would be interested to see a figure that shows promotion rates (Y axis) as a function of h-index scores (X axis), with separate lines for whites, blacks, Hispanics, and Asians. I suspect that we would find higher promotion rates for DEI faculty at all or most levels of research impact, as measured by the h-index. Just a thought.
With just 93 samples in the black/Hispanic bin, it doesn't sound like there is enough data to make reliable estimates of promotion rates with much granularity. The one graph of h-index vs ethnicity implies that at least two of those lines would stop close to the Y axis anyway, because there is no data for candidates with high h-index in those groups.
It is not racist at all. This is how social science is supposed to work. Trash social science articles should get critiqued harshly by other social scientists.
The Manhattan Institute has just issued a report trying to replicate a well-known paper from the Proceedings of the National Academy of Sciences that concluded that racial concordance between infants and the pediatrician providing neonatal care reduced neonatal mortality. The original authors cooperated, as did the agency from which neonatal outcome data was obtained. This was a study cited by KBJ in her dissent in the recent college affirmative action case.
The Manhattan team found that the original authors had included 65 comorbidity measures, but not a key one - low birth weight for the infant. If the low birth weight variable was included, no difference was seen in mortality for black infants delivered by black or white doctors. Further analysis of the data found that white doctors were much more likely tocare for low birth weight babies and other complicated cases. It is well known that survival rates for very low birth weight infants (<1500 grams bw, or about 3.3 pounds, the definition used by the Manhattan analysts) are much worse for larger babies. These are kids who are going to spend some time in a NICU - vulnerable to hypothermia, dehydration, electrolyte imbalances, and hypoglycemia, as well as often being underdeveloped or having been exposed to infections or adverse environmental exposures in utero. Only about 8% of infants are born <5 pounds, and they make up 80% of the cases of neonatal mortality. Given the detail the original team put into including 65 comorbidities into the analysis, it is astounding that this one was omitted.
Basically, white pediatricians were probably more skilled *on average* than black peers, and hence had a more complicated and risky case mix. The perverse result is that they saw more cases with a high risk of neonatal death from low birth weight. Because low birth weight is more common among black infants, this created an apparent disparity when birth weight was not controlled for, a result that made the more skilled doctors look like their quality was worse. Plotting mortality biy race and birthweight found the risk for the infants followed the SAME curve.
This raises the question of WHY peer reviewers did not notice the omission of the lbw variable, as the relationship between lbw and neonatal mortality is common knowledge in medicine and public health (I covered it in lectures given to students in my undergrad Intro to Public Health course as well as an epidemiology class, for example). My guess is that the findings agreed too well with the DEI fad during a time when the pressure on academics to go along with the neo-Marxist ideology was intense. I focus on the reviewers - given the number of comorbidities that the research team included, it likely slipped through the cracks or was omitted thinking that the other comorbidities accounted for its impact, and so I am willing to give them the benefit of the doubt given their cooperation with the Manhattan effort to replicate the study (behavior akin to John Snow, who invited a parson who raised questions about his Broad Street pump study to help him test those questions, eventually firming up Snow's argument that cholera was transmitted through contaminated water).
The original study is Brad N. Greenwood et al., “Physician–Patient Racial Concordance and Disparities in Birthing Mortality for Newborns,” Proceedings of the National Academy of Sciences (PNAS) 117, no. 35 (Aug. 17, 2020)
I have experience with how political beliefs impact publication decisions. Twenty years or so ago, I had a paper showing no impact of funding levels on local health department preparedness activity (dedicated leadership was the key factor behind activity). AJPH and Public Health Reports rejected it without peer review - both had editors who had argued that inadequate funding explained public health system problems. I submitted to a prominent general health policy journal and had it reviewed, accepted, and in print within six weeks.
I would go further and blatantly state that it is entirely possible for them to have completely fabricated the data from whole cloth. This is referred to as "outcomes-based" results. The conclusion is not jumped to, it is preordained, Soviet style via Lysenko. If the methodology cannot be clearly delineated by the authors without others redoing the statistical calculations, then the results are meaningless, all according to my major professor while I was in graduate school. When the NSF controls almost all of the funding, then you have political control over the scientific findings and topics of interest. I was heavily criticized for going to relatively obscure conferences and publishing in journals not widely subscribed to by certain 'libraries'. I did so due to the political nature of those scholarly works, yet published anyway to the consternation of my enemies. So they were relegated to primitive mockery. One of the conferences I attended and presented at had folks from the world over in attendance, including many Eastern Europeans, Asians, and Buddhist monks in full garb. How much more diversity, true diversity, can one be involved in than that? Yet I was still criticized, although I spoke to a standing room only audience with hundreds of people, many of whom approached me later to tell me that they primarily only came for my talk and no others. Censorship is now to be expected or at least ridicule for the most productive members of society. But this is nothing new to me, for I experienced this firsthand in my industrial career as well. It took a little longer to infiltrate other institutions and now has fully engulfed society as a whole.
Your experience mirrors mine. A lot of my research on political institutions and health programs involved data from the federal bioterrorism program or a large study of influenza preparedness my team carried out, and to me it was more relevant to the readers of the Journal of Homeland Security and Emergency Management or Bioterrorism and Biodefense than AJPH or AJPM, where departmental colleagues published. Likewise, work I did on mental health provider dropout in Tri-Care or a theoretical basis for medical and humanitarian civic action in counterinsurgency and stability operations (on contract for JFCOM/JIWC as part of the development of new COIN doctrine) was best disseminated in journals like Military Medicine. I had trouble getting work published in AJPH and Public Health Reports due to politics - one paper rapidly accepted by a prominent general health policy journal was sent back without review by those journals because it contradicted public positions taken by the editors. I also had a CDC employee try to bully me into not publishing my dissertation results (which found government regulators, but not private sector accreditors, ineffective at promoting medical lab quality) and two presentations on flu preparedness embargoed by the state health director who funded them. I was attacked for a paper I gave at a conference in 2011 on problems with the new ACA - the professor attacking me did not disagree with what I pointed out, but argued that I was being "too negative" and was "undermining the new law." Note that none of this was related to DEI issues!
Another problem is an academic mindset that defines the impact of research work in terms of citations of papers tracked by ISI. My chair told me in 2010 that I needed to do work with more "impact." I had developed guidance for the state health department for building surge capacity for flu and other respiratory epidemics (ahem, COVID), been asked to testify to the state EMS commission on work on rural EMS training and to serve on a state advisory board dealing with the issue, and was working with DoD not only on the new COIN doctrine, but was also involved in developing the concept of operations for post-Surge Iraq (on a team that included Petreaus' COIN advisor Dave Kilcullen and future Isreali COS/Defense Secretary Benny Ganz). My work was having am impact on issues in the real world in real-time, but wasn't in the "right" journals. Never mind the fact that even in the "wrong" journals the papers were cited as much as my peers, and across a range of disciplines in addition to my own. I had over $600K of external funding in six years in that department - fourth out of ALL department faculty, and all but $5000 came from contracts agencies asked me to take on instead of the usual grants, so evidently my real impact was sufficient to be sought out by funders instead of having to write proposals begging for funding. I actually had a tenured faculty member suggest that if I stopped taking on those contracts and applied for an *internal* grant my tenure application would look better! OTOH, if they took on external contracts, then maybe I wouldn't have been funding their graduate advisees as research assistants on my grants!
I am in the private sector now, and the greatest benefit is NOT having to deal with academic peers.
Preferential policies in hiring ensure that minority Assistant Professors are less likely to get tenure.
For example, I sat on a search committee for a joint position with the African-American Studies program at Purdue. Members from THAT program made it clear that they would veto any non-black candidate. On top of that, the research by the actual top candidate, who WAS black, raised points inconsistent with a critical theory approach where blacks were by definition oppressed, and they blocked him (our Dean gave us a new slot just so we could hire him for a non-joint position- he was an obvious star prospect). Their preferred candidate was so weak she likely would have never gotten tenure in our department, and we ended up with a compromise candidate.
All three of us hired as assistant professors without preferences when I was at that Big Ten school ended up very productive - publications, funding, teaching evals, service. Beneficiaries of preferential policies were in general weaker in terms of productivity, as to be expected since merit was a secondary consideration and the pool of candidates actually considered was smaller. One would expect the weaker candidates to be weeded out at the tenure/Assistant to Associate transition point. If a candidate survived that evaluation, they are likely to have shown the ability to compete with non-preferential hires, and thus are likely to be competitive with peers at the Associate to Professor transition point.
It is the Leftist paradigm that the world can be fit to an idealist model, described by Sowell in "A Conflict of Visions." The model is upheld despite empirical evidence. An example is the Lysenko model, which was consistent with Marxist ideology. Darwinian-type evolutionary models are rooted in survival as a function of the fitness of the individual, while in Marxist ideology it is a function of the collective class struggle. Lysenko rejected the idea of genetics and argued that an organism possessed an unlimited capacity for transformation through environmental exposure in a manner similar to Marxist dialectical transformation. Science, he argued, was immune to complexity and random action. Adoption of his ideas led to the virtual elimination of cell biology and genetics in the Soviet Union (including the imprisonment or execution of around 3,000 scientists during Stalin's purges). In Mao's China, it was a major contributor to the mass famines of the Great Leap Forward. It played a major role in the development of antibiotic-resistant diseases, as the ideology did not raise concerns about evolution of such strains due to misuse of antibiotics (see Laurie Garret's work for a good discussion of this). Eastern bloc crop yield fell well behind those of the West after the Green Revolution in agriculture, leading to the Soviet dependence on Western grain in the 1970s and 1980s to feed their population.
With respect, and I mean that seriously, "brevity is clarity". In the future you should just post "liberals and liberalism are one fvcked up set of people or notions after another".
All this arrant DEI nonsense puts me in mind of a Saul Bellow quote: "A great deal of intelligence can be invested in ignorance when the need for illusion is deep" In these few words he hit the nail on the head about the nature of our wokeified academia in the Western world. Anyone who has read Heather Mac Donald's comprehensive study of campus DEI The Diversity Delusion could be left in no doubt about who is really discriminated against in universities. But of course the kind of people who need to be put straight on all this, won't ever read anything that doesn't indulge their infantile virtue-signalling prejudices. https://grahamcunningham.substack.com/p/how-diversity-narrows-the-mind
Name the authors and their institutions.....
I really like that you are taking the time to really analyze the findings of this article. As a former professor, I know how often subpar articles get published if they enhance the ideological agenda of the academic Left. This, unfortunately, gives the professor who published the article enhanced status and encourages other professors to do the same. Worse, it then gets amplified by the mainstream media.
We need to fight this process by showing how fraudulent these studies are. Hopefully, if professors learn that the more Woke their conclusions, the more likely it is to get torn apart in public, then this creates somewhat of a counter-balance in incentives. This will not stop the process, but it might reduce the negative impact on the rest of society.
Keep doing this!
I would be interested to see a figure that shows promotion rates (Y axis) as a function of h-index scores (X axis), with separate lines for whites, blacks, Hispanics, and Asians. I suspect that we would find higher promotion rates for DEI faculty at all or most levels of research impact, as measured by the h-index. Just a thought.
With just 93 samples in the black/Hispanic bin, it doesn't sound like there is enough data to make reliable estimates of promotion rates with much granularity. The one graph of h-index vs ethnicity implies that at least two of those lines would stop close to the Y axis anyway, because there is no data for candidates with high h-index in those groups.
How did this racist shit end up in my feed!
It is not racist at all. This is how social science is supposed to work. Trash social science articles should get critiqued harshly by other social scientists.
I stand corrected. Racism masquerading a social science.
The Manhattan Institute has just issued a report trying to replicate a well-known paper from the Proceedings of the National Academy of Sciences that concluded that racial concordance between infants and the pediatrician providing neonatal care reduced neonatal mortality. The original authors cooperated, as did the agency from which neonatal outcome data was obtained. This was a study cited by KBJ in her dissent in the recent college affirmative action case.
The Manhattan team found that the original authors had included 65 comorbidity measures, but not a key one - low birth weight for the infant. If the low birth weight variable was included, no difference was seen in mortality for black infants delivered by black or white doctors. Further analysis of the data found that white doctors were much more likely tocare for low birth weight babies and other complicated cases. It is well known that survival rates for very low birth weight infants (<1500 grams bw, or about 3.3 pounds, the definition used by the Manhattan analysts) are much worse for larger babies. These are kids who are going to spend some time in a NICU - vulnerable to hypothermia, dehydration, electrolyte imbalances, and hypoglycemia, as well as often being underdeveloped or having been exposed to infections or adverse environmental exposures in utero. Only about 8% of infants are born <5 pounds, and they make up 80% of the cases of neonatal mortality. Given the detail the original team put into including 65 comorbidities into the analysis, it is astounding that this one was omitted.
Basically, white pediatricians were probably more skilled *on average* than black peers, and hence had a more complicated and risky case mix. The perverse result is that they saw more cases with a high risk of neonatal death from low birth weight. Because low birth weight is more common among black infants, this created an apparent disparity when birth weight was not controlled for, a result that made the more skilled doctors look like their quality was worse. Plotting mortality biy race and birthweight found the risk for the infants followed the SAME curve.
This raises the question of WHY peer reviewers did not notice the omission of the lbw variable, as the relationship between lbw and neonatal mortality is common knowledge in medicine and public health (I covered it in lectures given to students in my undergrad Intro to Public Health course as well as an epidemiology class, for example). My guess is that the findings agreed too well with the DEI fad during a time when the pressure on academics to go along with the neo-Marxist ideology was intense. I focus on the reviewers - given the number of comorbidities that the research team included, it likely slipped through the cracks or was omitted thinking that the other comorbidities accounted for its impact, and so I am willing to give them the benefit of the doubt given their cooperation with the Manhattan effort to replicate the study (behavior akin to John Snow, who invited a parson who raised questions about his Broad Street pump study to help him test those questions, eventually firming up Snow's argument that cholera was transmitted through contaminated water).
See https://manhattan.institute/article/do-black-newborns-fare-better-with-black-doctor
The original study is Brad N. Greenwood et al., “Physician–Patient Racial Concordance and Disparities in Birthing Mortality for Newborns,” Proceedings of the National Academy of Sciences (PNAS) 117, no. 35 (Aug. 17, 2020)
Please submit a letter to the editor at Science with these findings
As if they would publish it.
I have experience with how political beliefs impact publication decisions. Twenty years or so ago, I had a paper showing no impact of funding levels on local health department preparedness activity (dedicated leadership was the key factor behind activity). AJPH and Public Health Reports rejected it without peer review - both had editors who had argued that inadequate funding explained public health system problems. I submitted to a prominent general health policy journal and had it reviewed, accepted, and in print within six weeks.
I would go further and blatantly state that it is entirely possible for them to have completely fabricated the data from whole cloth. This is referred to as "outcomes-based" results. The conclusion is not jumped to, it is preordained, Soviet style via Lysenko. If the methodology cannot be clearly delineated by the authors without others redoing the statistical calculations, then the results are meaningless, all according to my major professor while I was in graduate school. When the NSF controls almost all of the funding, then you have political control over the scientific findings and topics of interest. I was heavily criticized for going to relatively obscure conferences and publishing in journals not widely subscribed to by certain 'libraries'. I did so due to the political nature of those scholarly works, yet published anyway to the consternation of my enemies. So they were relegated to primitive mockery. One of the conferences I attended and presented at had folks from the world over in attendance, including many Eastern Europeans, Asians, and Buddhist monks in full garb. How much more diversity, true diversity, can one be involved in than that? Yet I was still criticized, although I spoke to a standing room only audience with hundreds of people, many of whom approached me later to tell me that they primarily only came for my talk and no others. Censorship is now to be expected or at least ridicule for the most productive members of society. But this is nothing new to me, for I experienced this firsthand in my industrial career as well. It took a little longer to infiltrate other institutions and now has fully engulfed society as a whole.
Your experience mirrors mine. A lot of my research on political institutions and health programs involved data from the federal bioterrorism program or a large study of influenza preparedness my team carried out, and to me it was more relevant to the readers of the Journal of Homeland Security and Emergency Management or Bioterrorism and Biodefense than AJPH or AJPM, where departmental colleagues published. Likewise, work I did on mental health provider dropout in Tri-Care or a theoretical basis for medical and humanitarian civic action in counterinsurgency and stability operations (on contract for JFCOM/JIWC as part of the development of new COIN doctrine) was best disseminated in journals like Military Medicine. I had trouble getting work published in AJPH and Public Health Reports due to politics - one paper rapidly accepted by a prominent general health policy journal was sent back without review by those journals because it contradicted public positions taken by the editors. I also had a CDC employee try to bully me into not publishing my dissertation results (which found government regulators, but not private sector accreditors, ineffective at promoting medical lab quality) and two presentations on flu preparedness embargoed by the state health director who funded them. I was attacked for a paper I gave at a conference in 2011 on problems with the new ACA - the professor attacking me did not disagree with what I pointed out, but argued that I was being "too negative" and was "undermining the new law." Note that none of this was related to DEI issues!
Another problem is an academic mindset that defines the impact of research work in terms of citations of papers tracked by ISI. My chair told me in 2010 that I needed to do work with more "impact." I had developed guidance for the state health department for building surge capacity for flu and other respiratory epidemics (ahem, COVID), been asked to testify to the state EMS commission on work on rural EMS training and to serve on a state advisory board dealing with the issue, and was working with DoD not only on the new COIN doctrine, but was also involved in developing the concept of operations for post-Surge Iraq (on a team that included Petreaus' COIN advisor Dave Kilcullen and future Isreali COS/Defense Secretary Benny Ganz). My work was having am impact on issues in the real world in real-time, but wasn't in the "right" journals. Never mind the fact that even in the "wrong" journals the papers were cited as much as my peers, and across a range of disciplines in addition to my own. I had over $600K of external funding in six years in that department - fourth out of ALL department faculty, and all but $5000 came from contracts agencies asked me to take on instead of the usual grants, so evidently my real impact was sufficient to be sought out by funders instead of having to write proposals begging for funding. I actually had a tenured faculty member suggest that if I stopped taking on those contracts and applied for an *internal* grant my tenure application would look better! OTOH, if they took on external contracts, then maybe I wouldn't have been funding their graduate advisees as research assistants on my grants!
I am in the private sector now, and the greatest benefit is NOT having to deal with academic peers.
I’m shocked! Shocked!
Next you will tell me there is gambling in Casablanca…
I don't understand how you can be Jewish and against DEI.
I assume this is sarcasm.
Preferential policies in hiring ensure that minority Assistant Professors are less likely to get tenure.
For example, I sat on a search committee for a joint position with the African-American Studies program at Purdue. Members from THAT program made it clear that they would veto any non-black candidate. On top of that, the research by the actual top candidate, who WAS black, raised points inconsistent with a critical theory approach where blacks were by definition oppressed, and they blocked him (our Dean gave us a new slot just so we could hire him for a non-joint position- he was an obvious star prospect). Their preferred candidate was so weak she likely would have never gotten tenure in our department, and we ended up with a compromise candidate.
All three of us hired as assistant professors without preferences when I was at that Big Ten school ended up very productive - publications, funding, teaching evals, service. Beneficiaries of preferential policies were in general weaker in terms of productivity, as to be expected since merit was a secondary consideration and the pool of candidates actually considered was smaller. One would expect the weaker candidates to be weeded out at the tenure/Assistant to Associate transition point. If a candidate survived that evaluation, they are likely to have shown the ability to compete with non-preferential hires, and thus are likely to be competitive with peers at the Associate to Professor transition point.
The author conflates a symptom with a cause ... it's not junk science propping up DEI, it's "liberalism".
It is the Leftist paradigm that the world can be fit to an idealist model, described by Sowell in "A Conflict of Visions." The model is upheld despite empirical evidence. An example is the Lysenko model, which was consistent with Marxist ideology. Darwinian-type evolutionary models are rooted in survival as a function of the fitness of the individual, while in Marxist ideology it is a function of the collective class struggle. Lysenko rejected the idea of genetics and argued that an organism possessed an unlimited capacity for transformation through environmental exposure in a manner similar to Marxist dialectical transformation. Science, he argued, was immune to complexity and random action. Adoption of his ideas led to the virtual elimination of cell biology and genetics in the Soviet Union (including the imprisonment or execution of around 3,000 scientists during Stalin's purges). In Mao's China, it was a major contributor to the mass famines of the Great Leap Forward. It played a major role in the development of antibiotic-resistant diseases, as the ideology did not raise concerns about evolution of such strains due to misuse of antibiotics (see Laurie Garret's work for a good discussion of this). Eastern bloc crop yield fell well behind those of the West after the Green Revolution in agriculture, leading to the Soviet dependence on Western grain in the 1970s and 1980s to feed their population.
With respect, and I mean that seriously, "brevity is clarity". In the future you should just post "liberals and liberalism are one fvcked up set of people or notions after another".
Might work as a Comment for Econ Journal Watch, which I edit. If we published, we would invite the commented-on authors to reply.