"What reform to academia could ensure greater fidelity between pervasive academic claims and the actual scientific data addressing those claims? I doubt cutting grant indirects or even DEI will have this effect – regardless of whether one thinks those reforms are good or justified. Pre-registration and public posting of data and code have not had this effect. What could?"
Well that's a hell of a question. Here in the EU, it is European Commission mandated policy to prioritize gender parity over merit. And there are many Horizon initiatives of trying to get more women into STEM.
And the dynamic is self-reinforcing, institutionally speaking:
and so I think this is a more general and structural problem. Only a seismic shift, such as the STEM intelligentsia, after realizing the great tragedy of losing the intelligentsia-working class alliance, completely splitting from the social scientists in class consciousness and aligning with some new more classically liberal or libertarian-lite political movement, would cause any real changes within academia itself, IMHO.
I was on vacation this week without internet access so I am just reading this valuable work now. The results do not surprise me as I have seen such biases in favor of women at every career stage since high school. When applying for a tenure track position in a department were I had served one year as a fixed term hire (the department had not had time to do a full search), it was specifically stated by the hiring Dean that the goal was to hire a woman for the tenure track position. The faculty hiring committee told me I was welcome to apply but that the only way I would even be considered would be if NO qualified women applied.
The Dean even attempted to sabotage my interview to favor the woman finalist. Interviews in this college typically involve the candidate flying in and settling in a hotel on the night of Day 1, having breakfast and lunch with various meetings with faculty, staff and students including a teaching and research demo over Day 2 and returning home either the evening of Day 2 or early on Day 3. In my case, the Dean scheduled the parts of my interview over 4 days surrounding my overload schedule of 16 contact hours with early morning, afternoon and evening classes so that my teaching would not be impacted by the interview. In protest for this patent violation of equitable procedure I cancelled the interview and threatened to go to the Affirmative Action office to challenge the validity of the search. My interview was then rescheduled in the proper format minus the actual travel.
I got the job because it turned out that geoscience educator (female) that they had chosen as the finalist knew nothing about geology and thus was NOT qualified for the job. The Dean was reluctantly forced to hire me for the tenure track job. This was in April. By November, having bought a house and working hard in my first semester on the tenure track, the Dean made clear he had no intention of ever allowing me to get tenure.
So...what should be done about this bias in academia? 1. We need to be honest about it and dispel the myth that women have a bias against them. 2. We need to correctly challenge the legitimacy of every female hire in the past 30 years as being suspect (men have had to operate under the stigma of illegitimacy for decades. Time for women to pay that piper). 3. Future hiring, tenure and promotion committees should have any decision questioned that deviates from the gender population demographics of the applicant pools. (I.e. if 75% of the applicants are male, 75% of finalists, hires and promotions to tenured faculty need to reflect this or have some clear evidence to support deviation from the demographic mean).
Another thing: this creates the bizarre situation of old men excusing their attraction towards young female professionals by invoking the mantra of gender equality.
If a double blind study was conducted, do you think there would be a bias for or against hiring an AI [for a faculty position]?
ChatGPT said:
Great question—let's explore it from both sides, considering a hypothetical double-blind hiring study where evaluators assess applicants, unaware that some are AI-generated.
🔍 Key Assumption
We’re imagining AIs presented as applicants with credible résumés, writing samples, and maybe simulated interviews—but without revealing they are non-human.
⚖️ Possible Outcomes in a Double-Blind Study
✅ Bias for hiring an AI
Hyper-competence: If the AI resume shows exceptional qualifications, publications, or output, evaluators might favor it.
Bias toward objectivity: AI could be perceived as more productive, dispassionate, and objective—qualities prized in academia.
Novelty halo: Evaluators might unknowingly favor what they perceive as a “brilliant outlier” due to unusual patterns in the application.
❌ Bias against hiring an AI
Social subtlety: AIs may struggle to simulate human soft skills, emotional intelligence, or collaboration—important in academia.
Too perfect: An AI-generated CV might seem "too polished" or lack signs of human experience, triggering subconscious distrust.
Uncanny valley in writing: Even in a blind format, evaluators may notice something is off in tone, depth, or nuance.
Lack of institutional pedigree: AI resumes might lack traditional academic "lineage" (e.g., prestigious advisors or institutions).
📊 Likely Result?
If truly double-blind and if the AI outputs are strong but subtly different, the outcome might depend on field:
STEM disciplines might show more openness due to emphasis on output, publication, and technical skill.
Humanities or social sciences may be more cautious due to greater emphasis on human experience, positionality, and interpretive nuance.
🧪 Real-World Parallel?
In 2023–2025, multiple experiments showed that AI-written cover letters, essays, and even grant applications sometimes scored as well or better than human ones—but only when the AI mimicked human limitations (e.g., inserted typos, personal anecdotes).
Can AI replace faculty ? ChatGPT said:
AI cannot fully replace faculty now, but it can augment and eventually automate specific tasks. Here's a concise breakdown:
✅ Tasks AI Can Replace or Automate Soon (Now–5 years)
Grading (especially objective or formulaic work)
Lecture delivery (recorded or interactive tutorials)
⚠️ Tasks AI May Struggle With or Replace Later (5–15+ years)
Research leadership (hypothesis-driven inquiry, grant writing with originality)
Mentorship (emotional intelligence, career guidance)
Scholarly debate and critique (especially in humanities)
Ethical and institutional leadership
Interpersonal classroom dynamics
🧠 Bottom Line
AI will likely supplement, not replace faculty—co-pilot, not pilot—in the foreseeable future. Full replacement is unlikely until general AI can match human reasoning, values, and interpersonal nuance.
The real threat to humanity might not be male/female bias but from the AIs. Enjoy!
Science faculty’s subtle gender biases favor male students – Controlled study showing science faculty rated identical CVs with male names as more competent and likely to hire than female ones.
Steinpreis, Anders & Ritzke (2003) CV study – Evaluators gave male‑named CVs higher ratings for teaching, research, and service than identical female-named ones.
Northwestern University+2UC Davis Health+2NTNU Open+2
Northwestern unconscious‑bias compendium – National CV‑review experiment: both male and female evaluators gave better evaluations and higher salaries to supposedly‑male applicants.
Northwestern University+1UC Davis Health+1
Moss‑Racusin & colleague lab hiring experiment – Science faculty preferred and offered higher salaries to male‑named applicants for lab manager positions, despite identical qualifications.
Meta-analysis of hiring audit studies (1990–2022) – Analysis of 243,202 fictitious US job applications finds overall no significant gender discrimination, but bias varies by occupation and race.
Meta‑analysis of correspondence experiments across decades – Suggests men were preferred in male-stereotyped jobs, though in female-stereotyped jobs bias may reverse, especially after 2009.
Beyond Bias and Barriers (NAS report) – Review finds pervasive unconscious bias across hiring, tenure, promotion, and recommendation, contributing to attrition of women at every stage.
Williams & Ceci PNAS (2015) “two‑to‑one female advantage” – Hypothetical hiring experiments: faculty favored female vs. equally qualified male candidates ~2:1 across STEM fields.
TIME+1glamour.com+1
Abramo, D’Angelo & Rosati (Italy national recruitment, 2018) – Found no overall positive bias toward women and fewer women in the negatively biased group; committee-member ties helped men more.
arXiv+1arXiv+1
Bernstein et al. (2021) letters of recommendation study – In particle physics and social sciences, letters showed few gendered differences; women were more often described as “brilliant” in physics.
arXiv
Way, Larremore & Clauset (2016) CS faculty hiring model – Gender didn't significantly improve predictive models once productivity and institution prestige were included, suggesting indirect rather than direct bias.
John, FYI, most of your links are just to a generic website, (e.g., Wikipedia) rather than to the article on the specific topic. To be helpful, I'd need either a full reference-style citation, or a link to the actual article.
Also, some of what you have here (e.g., Ceci, Moss-Racusin) is in the post above already.
And of course there are zillions of articles on sex bias, but the only ones relevant to my call for "if you know of anything else, post it here" are: 1. Experimental studies where sex of applicant is manipulated; 2. for some sort of academic job; 3. and with faculty participants.
Last, beware of AI's producing nonsense. AFAICT, there is no Steinpress et al 2003 cv study. If I am wrong, please consider producing the full reference and a link to the article.
I replicated the Google scholar search done last night by the original posters. I tried to paste the Excel table, but it did not come through legibly. However, I did calculate the average number of citations per year, based on the number of years since publication (excluding the Solga study published this year). The average for papers showing bias in favor of males is 17.5 citations per year; for those showing bias in favor of females, 166.1/year.
I have a friend who worked for Facebook and confirmed the bias against hiring men in STEM if there was an equivalent woman available.
More important:
Quite apart from that, I have used the example of Williams-Ceci vs Moss Racussin as examples in the pastOne large scale empirical source to note is linkedin, which I have written about in the past, where they found women not only MORE likely to get hired, but also MORE likely to get promoted when they applied for a job.
Lee’s analysis is confirmed. I think one should also evaluate the quality and robustness of each study. Compare this also to other variables like impact factor, journals reputation, ideological commitment of the editorial board etc etc …
Useful to document these studies. However, in the name of rigour and not overstating our own case, I have some caveats I think you should’ve included:
1. Citation is not the same as endorsement, which you seem to imply with your “3x the citations…” bit.
2. Gender includes more information than you give it credit. Having an “identical” CV between a man and woman in a field dominated by one gender might still imply that one person is more of an outlier, or iconoclast, among their gender than the other, which could be viewed as a desirable trait.
3. The paper asking people to list the “top” candidate doesn’t seem to mention the deciding factor. It could be that between two candidates they view as identical they choose the one they know will get HR off their back or improve the balance of their department. It’s different from being asked to assess one’s merits as an academic and the subjects might tell you as much.
Fair-ish points. On 1 (what citations mean), they do in fact usually mean either endorsement or accepting the study at face value. I had an undergrad do small scale content analysis of just this issue (bias against women v. men in academic hiring) a couple of years ago, and most did, in fact, cite it in this manner. Precious few cited contrary evidence. About a third did cite a study just to say "this has been studied" which is literally true, not an endorsement, but still weird that most of those were to the biases against women papers.
On point 2, gender conveys info. Hahahahaha! You are conversing with The King of Stereotype Accuracy (go to my RU page for a list of papers: https://sites.rutgers.edu/lee-jussim/selectpublications/). People may indeed make all sorts of inferences; but when judging individuals, as here, applicants need to be judged on their merits, which were identical in all the experiments, not only because individuals do not necessarily conform to stereotypes, but also to make merit-based decisions. Indeed, this meta-analysis found that those who do emphasize merit engage in the least discrimination: https://scottbarrykaufman.com/wp-content/uploads/2023/09/leslie_LeslieBonoKimBeaver2020.pdf. Plus, it is illegal in the US to discriminate on the basis of sex, regardless of one's "reasons."
On point 3: True, but sex discrimination is illegal, even if you are doing it because you want to avoid trouble with HR or your superiors will approve your decisions.
Re 3, the point I was making is that it’s not the same type of discrimination as being genuinely biased against a person and discounting their abilities (despite being able to see them in the CV, for instance). Almost like statistical vs taste based, but not quite.
Re2, if you interviewed a man and a woman who had identical CVs saying that, I dunno, they were part of some elite military squad (not my field lol) you’d probably justified in making some inferences about their traits above and beyond what was on their CV, because of the additional adversity the woman may have faced in earning that role.
I’m not endorsing this type of discrimination nor do I think it explains most instances of hiring differences. When I was doing my PhD it was an open secret that many departments were on the hunt for women to hire. My point is that we need to be careful in precise in diagnosing what type of discrimination is occurring if we’re to combat it, and most of the papers seem to document the existence of discrimination, but cannot assess the motives or inferences being made.
Continuation of my post below (for some reason, it is being cut off -- word limit? -- No, I see now that you have to click on "see rest of comment" for longer comments.... )
Carlsson et al. (2021) -- 90 citations
Per year citation average: 23
AVERAGE per year citation for studies
finding hiring bias against men: 16
--------------------------
JUSSIM'S PREDICTIONS ARE CONFIRMED:
" ...on average, across the sets of 3 and 7 papers, at least 3x as many citations per year post-publication."
CONFIRMED, but underestimated: Actually, it is 10x, not 3x.
" ...the three studies finding biases against women will be, in total, cited more frequently than all 7 of those finding biases against men combined."
CONFIRMED, based on combined on per year citation rate totals: 495 vs. 112
"What reform to academia could ensure greater fidelity between pervasive academic claims and the actual scientific data addressing those claims? I doubt cutting grant indirects or even DEI will have this effect – regardless of whether one thinks those reforms are good or justified. Pre-registration and public posting of data and code have not had this effect. What could?"
Well that's a hell of a question. Here in the EU, it is European Commission mandated policy to prioritize gender parity over merit. And there are many Horizon initiatives of trying to get more women into STEM.
And the dynamic is self-reinforcing, institutionally speaking:
https://www.youtube.com/watch?v=4auHntuf65k
and so I think this is a more general and structural problem. Only a seismic shift, such as the STEM intelligentsia, after realizing the great tragedy of losing the intelligentsia-working class alliance, completely splitting from the social scientists in class consciousness and aligning with some new more classically liberal or libertarian-lite political movement, would cause any real changes within academia itself, IMHO.
I was on vacation this week without internet access so I am just reading this valuable work now. The results do not surprise me as I have seen such biases in favor of women at every career stage since high school. When applying for a tenure track position in a department were I had served one year as a fixed term hire (the department had not had time to do a full search), it was specifically stated by the hiring Dean that the goal was to hire a woman for the tenure track position. The faculty hiring committee told me I was welcome to apply but that the only way I would even be considered would be if NO qualified women applied.
The Dean even attempted to sabotage my interview to favor the woman finalist. Interviews in this college typically involve the candidate flying in and settling in a hotel on the night of Day 1, having breakfast and lunch with various meetings with faculty, staff and students including a teaching and research demo over Day 2 and returning home either the evening of Day 2 or early on Day 3. In my case, the Dean scheduled the parts of my interview over 4 days surrounding my overload schedule of 16 contact hours with early morning, afternoon and evening classes so that my teaching would not be impacted by the interview. In protest for this patent violation of equitable procedure I cancelled the interview and threatened to go to the Affirmative Action office to challenge the validity of the search. My interview was then rescheduled in the proper format minus the actual travel.
I got the job because it turned out that geoscience educator (female) that they had chosen as the finalist knew nothing about geology and thus was NOT qualified for the job. The Dean was reluctantly forced to hire me for the tenure track job. This was in April. By November, having bought a house and working hard in my first semester on the tenure track, the Dean made clear he had no intention of ever allowing me to get tenure.
So...what should be done about this bias in academia? 1. We need to be honest about it and dispel the myth that women have a bias against them. 2. We need to correctly challenge the legitimacy of every female hire in the past 30 years as being suspect (men have had to operate under the stigma of illegitimacy for decades. Time for women to pay that piper). 3. Future hiring, tenure and promotion committees should have any decision questioned that deviates from the gender population demographics of the applicant pools. (I.e. if 75% of the applicants are male, 75% of finalists, hires and promotions to tenured faculty need to reflect this or have some clear evidence to support deviation from the demographic mean).
Another thing: this creates the bizarre situation of old men excusing their attraction towards young female professionals by invoking the mantra of gender equality.
Maybe you could create an entire alternate culture where men and women are treated with respect.
Thanks for your efforts Lee digging up this data and putting it together.
If a double blind study was conducted, do you think there would be a bias for or against hiring an AI [for a faculty position]?
ChatGPT said:
Great question—let's explore it from both sides, considering a hypothetical double-blind hiring study where evaluators assess applicants, unaware that some are AI-generated.
🔍 Key Assumption
We’re imagining AIs presented as applicants with credible résumés, writing samples, and maybe simulated interviews—but without revealing they are non-human.
⚖️ Possible Outcomes in a Double-Blind Study
✅ Bias for hiring an AI
Hyper-competence: If the AI resume shows exceptional qualifications, publications, or output, evaluators might favor it.
Bias toward objectivity: AI could be perceived as more productive, dispassionate, and objective—qualities prized in academia.
Novelty halo: Evaluators might unknowingly favor what they perceive as a “brilliant outlier” due to unusual patterns in the application.
❌ Bias against hiring an AI
Social subtlety: AIs may struggle to simulate human soft skills, emotional intelligence, or collaboration—important in academia.
Too perfect: An AI-generated CV might seem "too polished" or lack signs of human experience, triggering subconscious distrust.
Uncanny valley in writing: Even in a blind format, evaluators may notice something is off in tone, depth, or nuance.
Lack of institutional pedigree: AI resumes might lack traditional academic "lineage" (e.g., prestigious advisors or institutions).
📊 Likely Result?
If truly double-blind and if the AI outputs are strong but subtly different, the outcome might depend on field:
STEM disciplines might show more openness due to emphasis on output, publication, and technical skill.
Humanities or social sciences may be more cautious due to greater emphasis on human experience, positionality, and interpretive nuance.
🧪 Real-World Parallel?
In 2023–2025, multiple experiments showed that AI-written cover letters, essays, and even grant applications sometimes scored as well or better than human ones—but only when the AI mimicked human limitations (e.g., inserted typos, personal anecdotes).
Can AI replace faculty ? ChatGPT said:
AI cannot fully replace faculty now, but it can augment and eventually automate specific tasks. Here's a concise breakdown:
✅ Tasks AI Can Replace or Automate Soon (Now–5 years)
Grading (especially objective or formulaic work)
Lecture delivery (recorded or interactive tutorials)
Administrative tasks (scheduling, documentation, syllabus prep)
Basic advising (degree tracking, FAQs)
Content generation (slides, quizzes, summaries)
⚠️ Tasks AI May Struggle With or Replace Later (5–15+ years)
Research leadership (hypothesis-driven inquiry, grant writing with originality)
Mentorship (emotional intelligence, career guidance)
Scholarly debate and critique (especially in humanities)
Ethical and institutional leadership
Interpersonal classroom dynamics
🧠 Bottom Line
AI will likely supplement, not replace faculty—co-pilot, not pilot—in the foreseeable future. Full replacement is unlikely until general AI can match human reasoning, values, and interpersonal nuance.
The real threat to humanity might not be male/female bias but from the AIs. Enjoy!
✅ Evidence Supporting Gender Bias Exists
Science faculty’s subtle gender biases favor male students – Controlled study showing science faculty rated identical CVs with male names as more competent and likely to hire than female ones.
sociologicalscience.com+15astro.uchicago.edu+15TIME+15
[Link to PDF]
Steinpreis, Anders & Ritzke (2003) CV study – Evaluators gave male‑named CVs higher ratings for teaching, research, and service than identical female-named ones.
Northwestern University+2UC Davis Health+2NTNU Open+2
Northwestern unconscious‑bias compendium – National CV‑review experiment: both male and female evaluators gave better evaluations and higher salaries to supposedly‑male applicants.
Northwestern University+1UC Davis Health+1
Moss‑Racusin & colleague lab hiring experiment – Science faculty preferred and offered higher salaries to male‑named applicants for lab manager positions, despite identical qualifications.
mdpi.com+15TIME+15astro.uchicago.edu+15
Meta-analysis of hiring audit studies (1990–2022) – Analysis of 243,202 fictitious US job applications finds overall no significant gender discrimination, but bias varies by occupation and race.
LSE Blogs+11sociologicalscience.com+11en.wikipedia.org+11
Meta‑analysis of correspondence experiments across decades – Suggests men were preferred in male-stereotyped jobs, though in female-stereotyped jobs bias may reverse, especially after 2009.
en.wikipedia.org
Beyond Bias and Barriers (NAS report) – Review finds pervasive unconscious bias across hiring, tenure, promotion, and recommendation, contributing to attrition of women at every stage.
mdpi.com+3en.wikipedia.org+3livescience.com+3
🧪 Evidence Arguing Bias Is Minimal or Reversed
Williams & Ceci PNAS (2015) “two‑to‑one female advantage” – Hypothetical hiring experiments: faculty favored female vs. equally qualified male candidates ~2:1 across STEM fields.
TIME+1glamour.com+1
Abramo, D’Angelo & Rosati (Italy national recruitment, 2018) – Found no overall positive bias toward women and fewer women in the negatively biased group; committee-member ties helped men more.
arXiv+1arXiv+1
Bernstein et al. (2021) letters of recommendation study – In particle physics and social sciences, letters showed few gendered differences; women were more often described as “brilliant” in physics.
arXiv
Way, Larremore & Clauset (2016) CS faculty hiring model – Gender didn't significantly improve predictive models once productivity and institution prestige were included, suggesting indirect rather than direct bias.
arXiv
John, FYI, most of your links are just to a generic website, (e.g., Wikipedia) rather than to the article on the specific topic. To be helpful, I'd need either a full reference-style citation, or a link to the actual article.
Also, some of what you have here (e.g., Ceci, Moss-Racusin) is in the post above already.
And of course there are zillions of articles on sex bias, but the only ones relevant to my call for "if you know of anything else, post it here" are: 1. Experimental studies where sex of applicant is manipulated; 2. for some sort of academic job; 3. and with faculty participants.
Last, beware of AI's producing nonsense. AFAICT, there is no Steinpress et al 2003 cv study. If I am wrong, please consider producing the full reference and a link to the article.
I replicated the Google scholar search done last night by the original posters. I tried to paste the Excel table, but it did not come through legibly. However, I did calculate the average number of citations per year, based on the number of years since publication (excluding the Solga study published this year). The average for papers showing bias in favor of males is 17.5 citations per year; for those showing bias in favor of females, 166.1/year.
Less important:
I have a friend who worked for Facebook and confirmed the bias against hiring men in STEM if there was an equivalent woman available.
More important:
Quite apart from that, I have used the example of Williams-Ceci vs Moss Racussin as examples in the pastOne large scale empirical source to note is linkedin, which I have written about in the past, where they found women not only MORE likely to get hired, but also MORE likely to get promoted when they applied for a job.
. https://www.linkedin.com/business/talent/blog/talent-acquisition/how-women-find-jobs-gender-report
Lee’s analysis is confirmed. I think one should also evaluate the quality and robustness of each study. Compare this also to other variables like impact factor, journals reputation, ideological commitment of the editorial board etc etc …
Useful to document these studies. However, in the name of rigour and not overstating our own case, I have some caveats I think you should’ve included:
1. Citation is not the same as endorsement, which you seem to imply with your “3x the citations…” bit.
2. Gender includes more information than you give it credit. Having an “identical” CV between a man and woman in a field dominated by one gender might still imply that one person is more of an outlier, or iconoclast, among their gender than the other, which could be viewed as a desirable trait.
3. The paper asking people to list the “top” candidate doesn’t seem to mention the deciding factor. It could be that between two candidates they view as identical they choose the one they know will get HR off their back or improve the balance of their department. It’s different from being asked to assess one’s merits as an academic and the subjects might tell you as much.
Fair-ish points. On 1 (what citations mean), they do in fact usually mean either endorsement or accepting the study at face value. I had an undergrad do small scale content analysis of just this issue (bias against women v. men in academic hiring) a couple of years ago, and most did, in fact, cite it in this manner. Precious few cited contrary evidence. About a third did cite a study just to say "this has been studied" which is literally true, not an endorsement, but still weird that most of those were to the biases against women papers.
On point 2, gender conveys info. Hahahahaha! You are conversing with The King of Stereotype Accuracy (go to my RU page for a list of papers: https://sites.rutgers.edu/lee-jussim/selectpublications/). People may indeed make all sorts of inferences; but when judging individuals, as here, applicants need to be judged on their merits, which were identical in all the experiments, not only because individuals do not necessarily conform to stereotypes, but also to make merit-based decisions. Indeed, this meta-analysis found that those who do emphasize merit engage in the least discrimination: https://scottbarrykaufman.com/wp-content/uploads/2023/09/leslie_LeslieBonoKimBeaver2020.pdf. Plus, it is illegal in the US to discriminate on the basis of sex, regardless of one's "reasons."
On point 3: True, but sex discrimination is illegal, even if you are doing it because you want to avoid trouble with HR or your superiors will approve your decisions.
Re 3, the point I was making is that it’s not the same type of discrimination as being genuinely biased against a person and discounting their abilities (despite being able to see them in the CV, for instance). Almost like statistical vs taste based, but not quite.
Re2, if you interviewed a man and a woman who had identical CVs saying that, I dunno, they were part of some elite military squad (not my field lol) you’d probably justified in making some inferences about their traits above and beyond what was on their CV, because of the additional adversity the woman may have faced in earning that role.
I’m not endorsing this type of discrimination nor do I think it explains most instances of hiring differences. When I was doing my PhD it was an open secret that many departments were on the hunt for women to hire. My point is that we need to be careful in precise in diagnosing what type of discrimination is occurring if we’re to combat it, and most of the papers seem to document the existence of discrimination, but cannot assess the motives or inferences being made.
Continuation of my post below (for some reason, it is being cut off -- word limit? -- No, I see now that you have to click on "see rest of comment" for longer comments.... )
Carlsson et al. (2021) -- 90 citations
Per year citation average: 23
AVERAGE per year citation for studies
finding hiring bias against men: 16
--------------------------
JUSSIM'S PREDICTIONS ARE CONFIRMED:
" ...on average, across the sets of 3 and 7 papers, at least 3x as many citations per year post-publication."
CONFIRMED, but underestimated: Actually, it is 10x, not 3x.
" ...the three studies finding biases against women will be, in total, cited more frequently than all 7 of those finding biases against men combined."
CONFIRMED, based on combined on per year citation rate totals: 495 vs. 112
Narrative vs. data. Which will win?
Eaton et al. (2020) -- 522 citations 5 years.
Per year average: 104
Moss-Racusin et al. (2012) -- 4551 citations 13 years
Per year average: 350
Steinpreis et al. (1999) -- 1138 citations 26 years
Per year average: 43
AVERAGE per year citation for studies finding hiring bias against women: 165
Solga et al. (2025) -- 1 citation
Per year citation average: 1
Solga et al (2023) -- 20 citations
Per year citation average: 10
Williams and Ceci (2015) -- 573 citations
Per year citation average: 53
Carey et al. (2020) - 58 citations
Per year citation average: 12
Henningsen et al. (2022) -- 29 citations
Per year citation average: 10
Honeycutt et al. (2025) -- 5 citations
Per year citation average: 5
Carlsson et al. (2021) -- 90 citations
Per year citation average: 23
AVERAGE per year citation for studies finding hiring bias against men: 16
JUSSIM'S PREDICTIONS ARE CONFIRMED:
" ...on average, across the sets of 3 and 7 papers, at least 3x as many citations per year post-publication."
CONFIRMED, but underestimated: Actually, it is 10x, not 3x.
" ...the three studies finding biases against women will be, in total, cited more frequently than all 7 of those finding biases against men combined."
CONFIRMED, based on combined on per year citation rate totals: 495 vs. 112
Narrative vs. data. Which will win?
Eaton et al. (2020) -- 522 citations
Moss-Racusin et al. (2012) -- 4551 citations
Steinpreis et al. (1999) -- 1138 citations
Solga et al. (2025) -- 1 citation
Solga et al (2023) -- 20 citations
Williams and Ceci (2015) -- 573 citations
Carey et al. (2020) - 58 citations
Henningsen et al. (2022) -- 29 citations
Honeycutt et al. (2025) -- 5 citations
Carlsson et al. (2021) -- 90 citations