Aaron Clauset
LightNews LightNews
banner
aaronclauset.bsky.social
Aaron Clauset
@aaronclauset.bsky.social
6K followers 120 following 1.1K posts
NO KINGS. NO FASCISTS. FUND SCIENCE.

Professor of Computer Science @ BioFrontiers Institute at University of Colorado, Boulder and External Faculty @ Santa Fe Institute

orcid: https://orcid.org/0000-0002-3529-8746
Posts Replies Media Videos
aaronclauset.bsky.social
Slides from my @mit.edu IDSS Distinguished Speaker seminar "Networks untangle gender differences in productivity and prominence among scientists" this week

I argue that collaboration networks act like unequally distributed (and gendered) social capital

aaronclauset.github.io/slides/Claus...
Slide title: "network effects in scientific labor" networks mediate most scientific activities: 1. scientific training, hiring, collaboration, teaching, attention, peer review, etc. 2. networks act like a form of unequally distributed social capital a productive collaborator --> increases your productivity a prominent collaborator --> increases your prominence How much does who you work with impact your productivity and prominence? Slide title: "model checking". Shows the results of applying our probabilistic generative from Li et al. Nature Communications (2022) for estimating individual productivity and prominence parameters from observed collaboration network data. This "nets out" the network, and estimates individual levels of activity. Left figure is joint distribution of productivity lambda and prominence theta, showing productivity is tightly concentrated around 0.42 papers/year, while prominence has a long tail, with most mid-career scientists (we studied 200,000 from 6 STEM fields) have very low prominence. Right figure shows pairwise correlation matrix of measures like number of papers, number of citations, lambda, theta, having a high-lambda coauthor, and having a high-theta coauthor. Strong correlation between your own number of papers and having a high-lambda (very productive) coauthor, etc. Good sanity checks for the model. Slide title "gender vs. productivity & prominence" Notes decades of past work showing a "productivity gap" between men and women, in which men publish more papers and receive more citations over time. But, after "netting out" their collaboration networks, we find no gender difference at all between individual productivity and individual prominence, implying that it's difference in who men and women work with (the size and composition of their collaboration networks) that drives the observed gap in productivity, not differences at the individual level. Slide title "how important is who you work with?" This is a wrap-up slide from the end of the talk: networks act like unequally distributed social capital in science they mediate our scientific attention, evaluation, and collaboration differences in collaboration networks can explain gendered differences in productivity & prominence early-career productivity & prominence what else? can we intervene in these networks to mitigate inequalities? funds for new collaborations, eg, after parenthood? early-career fellowships to work with elite senior coauthors?
November 6, 2025 at 6:05 PM
aaronclauset.bsky.social
Our 2017 @pnas.org study of computer science shows how academic productivity varies over 20 years. There's definitely a pre-tenure surge, with productivity decaying slowly after, and then stabilizing (at a level we showed in a 2022 paper is determined by lab group size)
www.pnas.org/doi/full/10....
Fig 2. Five average productivity curves (publications per year) are shown, one for each strata of university prestige, showing a rapid rise to an early peak (around expected tenure time) followed by a slow decline to a stable, prestige-specific value. Averages over researchers at all levels of institutional prestige follow similar productivity trajectories, in agreement with the conventional narrative, but at differing scales of output.
July 30, 2025 at 7:10 PM
aaronclauset.bsky.social
Slides from my keynote lecture "What drives the productivity of scientific labor?" at the 2025 Oxford Summer School on Economic #Networks. Part 2 of 2: how does who you collaborate with shape your own productivity, and how are networks like social capital? aaronclauset.github.io/slides/Claus...
Slide titled 'network effects in scientific labor', arguing that your own productivity is a network effect, because it includes the productivity of your coauthors too Slide showing a worked example of a collaboration network, and using a probabilistic model to marginalize the network effects of collaboration to estimate individual productivity (without network effects) Slides showing results of applying the network effect model showing that collaborating with a highly productive or highly prominent coauthor in your early career has a lasting effect on your own productivity and prominence, suggesting that social capital (network effects) are a kind of semi-transferrable form of wealth Slides describing a network analysis of homophily in collaboration networks, showing that diversity in collaborators is a learned tendency, and mid-career researchers who trained in highly diverse research groups go on to lead relatively more diverse groups themselves
June 24, 2025 at 9:35 PM
aaronclauset.bsky.social
Slides from my keynote lecture "What drives the productivity of scientific labor?" at the 2025 Oxford Summer School on Economic #Networks. Part 1 of 2: why and how do elite scientists dominate scientific discourse?
aaronclauset.github.io/slides/Claus...
Title slide: "What drives the productivity of scientific labor?" a talk by Aaron Clauset at the Oxford Summer School on Economic Networks 2025 Summary slide arguing that prestige is a structural variable in the scientific ecosystem, and that it is differences in working environment (not pedigree) that explains why elite researchers dominate scientific discourse. Ends with a question: how exactly does environment do this? Summary slide arguing that working environment (specifically, institutional prestige) explains differences in scientific productivity because there's just more available academic labor at elite places, and researchers use this labor to write more papers. Ends with a question: how much does it matter who you collaborate with? That leads into part 2 of the slides
June 24, 2025 at 9:35 PM
aaronclauset.bsky.social
Hurray! 🎉 @arash.network (left) is the 24th recipient of the Zachary's Karate Club CLUB prize. For his talk @netsciconf.bsky.social NetSci 2025, he was entrusted with the illustrious trophy 🏆 from @nwlandry.bsky.social #networkscience networkkarate.tumblr.com
A delightfully nerdy photo of @arash.network receiving a small karate club trophy from @nwlandry.bsky.social An "action shot" of @arash.network during his award winning talk, with a TV in the background showing the famous karate club network
June 4, 2025 at 4:27 PM
aaronclauset.bsky.social
Next, we consider race and define a similar null model of relative diversity. Here, we find that the same dynamic holds for racial diversity 🎯: diverse early collaborations are strongly associated with more diverse junior coauthors as an established researcher /7
Multi-panel figure from the paper. Caption: Racial diversity-association effects in the co-authorship networks of researchers. Racial diversity-association effects in co-authorship networks relative to the randomized null model, and how the racial diversity of junior co- authors in the established period is correlated with that of their early co-authors for individual researchers. The cohort of junior co-authors for a researcher is defined to be racially diverse if the racial entropy score is greater than the baseline rate predicted by the null model. a) The interplay between the racial diversity of junior co-authors in the early-career and established periods of researchers. b) Racial diversity-association effects in co-authorship networks over time, from 1960 to 2012, according to the first career years of individual researchers. c) Racial diversity-association effects across countries, where we select the top-20 countries with the most researchers and arrange them according to the continents they belong to. d) Statistical distribution of racial diversity-association effect Δc of countries with at least 100 researchers. Bars in a represent mean values and error bars indicate 95% confidence intervals. Lines denote mean values and shaded areas represent 95% confidence intervals in panel b. We report effect sizes using the Cohen’s d statistic and use two-sided t-tests for comparisons. P values reported in the plot are adjusted by the Benjamini– Hochberg procedure for multiple comparisons. ***P < 0.001.
April 21, 2025 at 8:13 PM
aaronclauset.bsky.social
Coauthorship data shows similar results 👏: more gender diverse coauthors early in a career correlates with more gender diverse junior coauthors later in a career, even in gender diverse fields! This effect is stable across countries, and men exhibit stronger effects than do women /6
Multi-panel figure from the paper. Caption: Gender diversity association in the co-authorship networks of researchers. Results of gender diversity-association analysis in the co-authorship networks relative to the randomized null model, showing how gender diversity of junior co-authors in the established period correlates with that of their early co-authors for individual researchers. The cohort of junior co-authors for a researcher is defined to have a high proportion of women if the percentage of women is greater than the baseline rate predicted by the null model. a) The interplay between the gender diversity of junior co-authors in the early-career and established periods of researchers, for men and women. b) Gender diversity- association effects in co-authorship networks over time, from 1960 to 2012, according to the first career years of individual researchers. c) Gender diversity-association effects by the number of junior co-authors in the early-career period. d,f), Gender diversity association across countries, for men (d) and women (f). e,g) Statistical distribution of gender diversity-association effect Δc of countries with at least 100 researchers, for men (e) and women (g). We select the top-20 countries with the largest number of established researchers and arrange them according to the continents they belong to. Bars in a and c represent mean values and error bars indicate 95% confidence intervals; the error bars in c are too small to be graphically visible. Lines denote mean values and shaded areas represent 95% confidence intervals in panel b. We report effect sizes using the Cohen’s d statistic and use two-sided t-tests for comparisons. P values reported in the plot are adjusted by the Benjamini–Hochberg procedure for multiple comparisons. ***P < 0.001.
April 21, 2025 at 8:13 PM
aaronclauset.bsky.social
Relative to our gender diversity null model, we find that researchers trained in gender diverse groups in the early period is strongly associated with working with gender diverse junior colleagues as established researchers 💪. The effect is particularly strong for men researchers /5
Multi-panel Figure from paper. Caption: Comparing gender diversity of advisors’ groups and researchers’ own groups. The association effects of gender diversity relative to the null model using the mentorship data. a) The interplay between the gender diversity of advisees in the group of a researcher’s advisor and that of his/her own group, for men and women. The cohort of advisees in a research group is defined to have a high proportion of women if the real percentage of women is greater than the baseline rate predicted by the null model. b) The gender diversity- association effects over time from 1970 to 2010, based on the researchers’ first year as advisees. As data are less available for researcher groups in the early period, we merge data points for every 5 years instead of displaying yearly data. c) The interplay between the research group size of researchers’ advisors and the gender diversity-association effects. Bars represent mean values and error bars indicate 95% confidence intervals in panels a and c. Lines denote mean values and shaded areas represent 95% confidence intervals in b. We report effect sizes using the Cohen’s d statistic and use two-sided t-tests for comparisons. ***P < 0.001.
April 21, 2025 at 8:13 PM
aaronclauset.bsky.social
First, we consider gender diverse research groups (see paper for defn). From detailed longitudinal data on collaborations, we find that early collaborations with gender diverse groups is strongly associated with working with a gender diverse group of junior researchers as an established scholar 📈 /3
Multi-panel figure from the paper. Caption: Diversity-association effects in the social networks of researchers in the early-career and established periods. a) An illustrative example of the gender diversity-association effect for two researchers during the early-career and established periods. The early social network of the red author comprises mostly women, and his network in the established period continues to be dominated by women. The green author interacted more often with men in their early career, and her connections in the established period are mostly men researchers as well. b) The group size distribution of advisors for researchers in the mentorship data. c) The fraction of an advisor’s advisees who are women as a function of advisor gender and the gender composition of their own early-career environment (n = 17, 917). d) Number of early junior co-authors of researchers in the collaboration network. e) The correlation between the proportion of women junior co-authors for researchers in the early-career and established periods (n = 562, 494). Bars represent mean values and error bars indicate 95% confidence intervals in panels c and e. We report effect sizes using the Cohen’s d statistic and use two-sided t-tests for comparisons. ***P < 0.001.
April 21, 2025 at 8:13 PM
aaronclauset.bsky.social
🎉 Our new paper "Gender and racial diversity socialization in science" @natcomputsci.nature.com with @weihuali.bsky.social, H Zheng and @jenniebrand.bsky.social studies how early-career experiences with diversity drive more diverse teams later in a career 👉 /1
www.nature.com/articles/s43...
"Gender and racial diversity socialization in science", a new published article by Weihua Li, Hongwei Zheng, Jennie E. Brand, and Aaron Clauset. Abstract: Scientific collaboration networks are a form of unequally distributed social capital that shapes both researcher job placement and long-term research productivity and prominence. However, the role of collaboration networks in shaping the gender and racial diversity of the scientific workforce remains unclear. Here we propose a computational null model to investigate the degree to which early-career scientific collaborators with representationally diverse cohorts of scholars are associated with forming or participating in more diverse research groups as established researchers. When testing this hypothesis using two large-scale, longitudinal datasets on scientific collaborations, we find that the gender and racial diversity in a researcher’s early-career collaboration environment is strongly associated with the diversity of their collaborators in their established period. This diversity-association effect is particularly prominent for men. Coupled with gender and racial homophily between advisors and advisees, collaborator diversity represents a generational effect that partly explains why changes in representation within the scientific workforce tend to happen very slowly.
April 21, 2025 at 8:13 PM
aaronclauset.bsky.social
Great turnout at the #StandUpForScience #StandUpForScience2025 rally in Denver! ✊

My sign read “I’d rather be PEER REVIEWING!” 🧪🔬
March 7, 2025 at 10:52 PM
aaronclauset.bsky.social
The end of 2024 is in sight! That makes it time for my annual "Year in Review" post 📈, which is to remind myself that yes, I did get some things done this year 😅. Here's to closing out 2024 and hoping for good 2025! 🥳 aaronclauset.github.io/blog/2024_YiR
Screenshot of a 2024 Year in Review blog post, dated 28 December 2024. "This is it for the year. Here's a look back at my 2024, by the numbers:" [Click the link to read the rest!]
December 28, 2024 at 2:06 PM
aaronclauset.bsky.social
Yes to all of this. Fwiw, our survey respondents had A LOT to say about the particular reasons. We wrote up the results here: "Gendered devaluation underlies faculty retention", also led by @kspoon.bsky.social, with @danlarremore.bsky.social, @larivera.bsky.social and others. osf.io/preprints/so...
Screenshot of paper title, authors, and abstract: "Gendered devaluation underlies faculty retention" (2024), by Katie Spoon, Joanna Mendy, Maria Martinez, Mirta Galesic, Daniel B Larremore, Aaron Clauset, and Lauren A Rivera. Abstract: "Women faculty experience academia differently from men in many ways, which can lead them to consider leaving their positions. Using a large-scale survey of 10,071 current and former tenure- track and tenured faculty, representing nearly all U.S. PhD-granting institutions and 29 diverse fields of study, we show that perceived workplace climate is the most gendered aspect of faculty life, compared to stress related to research pressures, work-life balance, and departmental support. Further, analyzing 6,615 free-text responses from the same respondents reveals that devaluation, both in formal evaluations and informal interactions, is the most gendered workplace climate factor. These patterns are especially salient among women of color and tenured women. Women report that devaluation is exacerbated when institutional leadership fails to respond to devaluation, leading many to leave their jobs. Our results highlight that successful remedies must involve organizational change rather than solely individual solutions."
December 1, 2024 at 4:43 AM
aaronclauset.bsky.social
Since launching in 2016, ICON (icon.colorado.edu) has supported a pile of fun papers from my own group on #networkscience, including network "corpora" data sets and systematic comparative studies. I also use it in my classes. ICON would not exist without the many students who contributed to it ❤️ /3
November 11, 2024 at 6:17 PM
aaronclauset.bsky.social
ICON (icon.colorado.edu) supports search by domain, subdomain, graph properties, graph size, and by keywords. The new site does some dynamic loading to make it more responsive, and retains the original "mini page" functionality for easy browsing /2 #networkscience
Screenshot of the Index of Complex Networks (icon.colorado.edu) search interface, allowing users to search among the 699 network data sets by domain, subdomain, graph properties, and graph size, or by keyword.
November 11, 2024 at 6:17 PM
aaronclauset.bsky.social
🚨 The Colorado Index of Complex Networks (icon.colorado.edu) is reborn!

ICON 2.0 indexes 699 distinct research-quality network data sets, classified into 6 domains (Social, Biological, Information, Economic, Technology, Transportation) and 53 subdomains /1 #networkscience
Screenshot of the Index of Complex networks (icon.colorado.edu) website, showing histograms of the 699 network data sets it indexes, sizes of the networks, and their representation across subdomains. It's a nice, slick interface.
November 11, 2024 at 6:17 PM
aaronclauset.bsky.social
You are correct! 🤣 Since most people know that reference via Harry Potter, here, I fixed it for them
Screenshot of a scientific preprint titled "The Philosopher's Stone for Science -- The Catalyst Change of AI for Scientific Creativity" but with the philosopher's stone crossed out and replaced with "The Mirror of Erised", a Harry Potter reference to a mirror that showed the "deepest, most desperate desire of our hearts."
November 8, 2024 at 10:29 PM
aaronclauset.bsky.social
This is cute. Pretty tremendous growth in a year. The science community is really coming along nicely
User number 220,156 reporting for duty
September 17, 2024 at 3:27 AM
aaronclauset.bsky.social
It was a true honor to give the Gordon Rausser Keynote @AAEA_Economics in New Orleans, discussing how prestige acts like a demographic filter on the composition of the scientific workforce and how it shapes the spread of scientific ideas. Slides: aaronclauset.github.io/slides/Claus...
Photo of Acadia meeting room, full of people in seats, watching Aaron Clauset on stage give the Gordon Rausser Keynote address.
July 30, 2024 at 12:56 PM
aaronclauset.bsky.social
Hurray! 🎉 @nwlandry.bsky.social (left) is the 23rd winner of the Zachary's Karate Club CLUB prize. For his talk @netsciconf.bsky.social NetSci 2024, he received the coveted trophy 🏆 from Elsa Andres (on behalf of @lgajo.bsky.social) #networkscience networkkarate.tumblr.com
Image of Nicholas Landry receiving the Zachary's Karate Club CLUB trophy from Elsa Andres
June 27, 2024 at 8:07 PM
aaronclauset.bsky.social
It's a reasonable idea, but research suggests published rankings have significant anchoring effects, so residuals aren't independent of the rankings themselves. See Bastedo & Bowman (2010) www.journals.uchicago.edu/doi/full/10....
Bastedo & Bowman (2010) USNews & World Report College Rankings: Modeling Institutional Effects on Organizational Reputation. Abstract: Processes of certification and evaluation are some of the most powerful institutional forces in organizational fields, and in the higher education field, rankings are a primary factor in assessing organizational performance. This article explores the institutional effects of the USNWR undergraduate rankings on the reputational assessments made by senior administrators at peer universities and liberal arts colleges. In the estimation of structural equation models, we found that published college rankings have a significant impact on future peer assessments, independent of changes in organizational quality and performance and even of prior peer assessments of reputation.
April 30, 2024 at 4:32 PM
aaronclauset.bsky.social
Love this: "Complaints about too many books predate printing..." and then after it was invented: "...worrying about the end of civilization from too many people writing (bad) books..." #ht Misha Teplitskiy
Book page saying "Complaints about too many books predate printing, including the biblical lament 'of making many books there is no end' (Ecclesiastes 12:12) or the thirteenth-century prologue in which Vincent of Beauvais complained of 'the multitude of books, the shortness of time, and the slipperiness of memory' that motivated him to write a compendium of all knowledge. While only a narrow elite of scholars had experienced it in earlier periods, after the first century of printing a broad cross section of the educated articulated a concern about what we would call 'information overload'-the sense that there were too many books for an individual to read and to master. Complaints about an overabundance of books became a well-worn refrain throughout early modern Europe and were used to justify any number of postures: railing about the base commercial motives of printers who produced whatever sold without a care for quality or intellectual merit, mocking the vanity of bookish..."
March 16, 2024 at 1:11 PM
aaronclauset.bsky.social
Apropos nothing: my generic "how to be a good reviewer" advice that I give my students when we co-review papers. (Reposted from the bad place)
A referee report should begin with a one paragraph summary of the paper (written in your own words, from your own perspective, not a paraphrasing of the paper's abstract), its main goals and its approach. It should then have a paragraph briefly summarizing the main successes or the main problems (1-2 sentences each) of the paper. Here you can say whether and why (at a high level) you think the paper makes a substantial or very useful scientific contribution, and whether it rises to the level of a scientific publication in the venue being considered. (Sometimes, a paper is fine, but its contribution will be of more interest to a more specialized audience than the venue supports.) Novelty can be a part of a paper's contribution, but all research lacks novelty in some way, so think more deeply; utility is often more important, eg, will other scientists find these results useful to build on? Do they clarify or quantify our understanding of some phenomenon in a way that facilitates bett...
February 20, 2024 at 10:41 PM
aaronclauset.bsky.social
The end of 2023 is finally in sight, making it time for my annual "Year in Review" post 📈, to remind myself that yes, I did get some things done this year 😅. Here's to closing out 2023 and hoping for good 2024! 🥳 aaronclauset.github.io/blog/2023_YiR
Screenshot of the beginning of my 2023 Year in Review. Click through for more
December 29, 2023 at 11:01 PM
aaronclauset.bsky.social
Slides from my talk "Gender & Retention Patterns Among US Faculty" earlier this week @UMass ADVANCE, on our paper with the same name, led by @kspoon.bsky.social. What an amazing team at UMass, doing critical work to improve science and academia 👏
aaronclauset.github.io/slides/Claus...
slide title: "attrition over a career." shows data indicating that for "all-cause attrition", at every career age, women are more likely to leave than men slide title: "study design." highlights the survey portion of the study, intended to elicit the reasons faculty leave their jobs, noting that gendered reasons for leaving are more important than gendered rates of leaving slide title: "push & pull." shows data indicating that women feel pushed out of their academic jobs at substantially higher rates than men, and that gender+career age are the strongest predictors of whether a survey respondent reported feeling pushed out of their faculty job slide title: "conclusions." 1. women faculty universally leave academia at higher rates than men, but the rates vary in magnitude by domain and career age, mainly effecting tenured women, in non-STEM fields, at lower-prestige universities. but regardless of the rates, women leave for different, more worrisome reasons than men: they feel pushed out of their jobs. retention efforts have to focus on these reasons, not the rates
November 9, 2023 at 7:22 PM