iiinova — where your digital self keeps evolving

Kate Barnes

@k-barnes.bsky.social

19 followers 10 following 12 posts

Graduate student at CU Boulder: Ph.D. Computer Science & MA Philosophy.

Posts Replies Media

Reposted by Kate Barnes

Dan Larremore @danlarremore.bsky.social · 25/08/2026

🎉 New paper (and data) on editorial & peer review at Science and Science Advances, two elite general science journals. With a wonderful team! @samzhang.bsky.social, @nicklaberge.bsky.social, Quinten and @aaronclauset.bsky.social. 1/ www.science.org/doi/full/10....

Screenshot of the title page:

Elite general science journals shape scientific discourse, public policy, and scientific careers. However, expectations of confidentiality in most editorial proceedings has limited efforts to understand and improve the review process. Here, we describe and release deidentified data on 110,303 manuscript submissions over 5 years at Science and Science Advances, two elite general science journals. Analyzing evaluation dynamics across the initial editorial review and subsequent peer review stages, we find strong selective effects associated with higher institutional prestige, larger team size, and certain topics and countries. Corresponding authors who are men exhibit a small but significant advantage at Science, while authors based in China have a significant disadvantage. These associations are generally stronger in editorial review than in peer review, even as final editorial decisions correlate strongly with reviewer advice. These patterns highlight the complexity of multistage evaluations at elite journals and the importance of open data to better understand them.

39439

Kate Barnes @k-barnes.bsky.social · 23/06/2026

12\ Check it out in FAccT 2026 @facct.bsky.social, “Edge interventions mitigate disparities in the CS coauthorship network” with Mia Ellis-Einhorn, @cchavezru.bsky.social, Nayera Hasan, Mohammad Fanous, Blair D. Sullivan, @friedler.net, @aaronclauset.bsky.social 🥳

030

Kate Barnes @k-barnes.bsky.social · 23/06/2026

11\ Our results are not causal but they do suggest mechanisms which could be tested experimentally in future research by actually implementing fellowship programs following our principled selection of target scholars to maximize program benefits ❣️🎓

140

Kate Barnes @k-barnes.bsky.social · 23/06/2026

10\ A simple linear model shows that coauthorship centrality is predictive placement institution rank on the academic job market, with influence over and above degree. When applied to Ph.D. students, our intervention improves predicted placement according to this model.

Bar chart showing coefficients for of linear model. Controlling for Ph.D. rank, closeness centrality has predictive power over and above coauthorship degree. Scatter chart shows that targeted Ph.D. student scholars from the lowest rank institutions can be expected to see a 5 rank improvement in their placement outcomes.

153

Kate Barnes @k-barnes.bsky.social · 23/06/2026

9\ Increases in coauthorship centrality may be associated with more access to information and opportunities, improvements in citations and other academic success metrics. We conjecture that centrality plays a practical role in improving individuals’ career prospects.

140

Kate Barnes @k-barnes.bsky.social · 23/06/2026

8\ We consider how one might address these disparities with a simulated intervention that adds a single edge between target and sponsor individuals. The intervention improves targets’ centrality more than would be expected from random edge additions.

Schematic of our simulated intervention: 1. Select target individual with minoritized racial identity from low ranked institution, 2. Connect with sponsor individual from top institution, 3. Measure new degree and closeness centrality. Scatter plot shows improvement in closeness centrality positively correlated with targets institutional rank.

121

Kate Barnes @k-barnes.bsky.social · 23/06/2026

7\ In the computer science coauthorship network, we observe disparities in coauthorship centrality by gender, race, academic parent status, and most prominantly by institutional prestige 📈

Bar chart showing that women are less centrally located than men, minority races less central than majority races, those at low prestige institutions less central than those at high prestige institutions and those whose parents held academia jobs are less central than those with academic parents.

125

Kate Barnes @k-barnes.bsky.social · 23/06/2026

6\ Key finding: computer scientists from top institutions are located more centrally in the coauthorship network. The network is built from all DBLP-indexed publications, and also exhibits significant racial but not gender-based homophily.

Figure shows a strong linear correlation between the institutional prestige scores and coauthorhsip centrality of U.S. computer scientists. Bar chart displays homophilic collaboration rates significantly higher than expected for White, East Asian, Indian and Middle Eastern scholars, as well as generally for scholars from majority and minority racial groups.

142

Kate Barnes @k-barnes.bsky.social · 23/06/2026

5\ With survey self-reports to validate against, we developed 2 meta-labeling algorithms that combine perceived and name-based demographic labels for more accurate estimates. Our algorithms are a methodological contribution, improving demographic annotation practices in computational social science.

Heat maps showing algorithm performance. The gender and race meta-labelers respectively achieve 98% and 92% agreement with survey self-reports. Bar plots showing demographic distributions in our data. Most (87%) CS faculty are from majority racial groups (White, East Asian or Indian), and most (80%) are men.

130

Kate Barnes @k-barnes.bsky.social · 23/06/2026

4\ We hired 7 undergraduate coders to conduct a manual census of tenure-track faculty at all U.S. universities that grant PhD’s in computer science. The census was augmented with a survey requesting self-report demographic information and parental education levels, with a response rate of 15%.

Four questions from our survey of all U.S. tenure-track faculty in computer science: How do you identify your gender? How do you identify your race / ethnicity? What is the highest educational attainment of any of your parents? Have any of your parents ever been a tenured of tenure-track professor at a PhD-granting institution?

120

Kate Barnes @k-barnes.bsky.social · 23/06/2026

3\ Previous work shows systematic devaluation of scientists from gender & racial minorities. Prestige structures scientific outcomes - reflecting a combination of merit and various non-meritocratic factors. Our ameliorative approach empirically expands on this literature and presents a solution ⭐

120

Kate Barnes @k-barnes.bsky.social · 23/06/2026

2\ We first show that coauthorship in computer science exhibits disparities by prestige, demographics and having a parent in academia. We explore interventions which mitigate these disparities. Think: early career fellowships to work with elite senior scientists.

The image shows the landing web pages of the NSF Research Experience for Undergraduates and Saturday Academy ASE Research Internships for high schools. Our research suggests that research programs such as these could also be beneficial at the Ph.D. level.

142

Kate Barnes @k-barnes.bsky.social · 23/06/2026

1\ New paper out in FAccT 2026 @facct.bsky.social #FAccT2026 ! 🥳 Grateful to have worked with this team, and had the opportunity to combine my interests in fairness and the science of science. Link: dl.acm.org/doi/10.1145/....

Screenshot of the first page of our FAccT publication. Title: "Edge interventions can mitigate demographic and prestige disparities in the computer science coauthorship network." Authors: Kate Barnes, Mia Ellis-Einhorn, Carolina Chavez, Nayera Hasan, Mohammad Fanous, Blair D. Sullivan, Sorelle A. Friedler and Aaron Clauset. Affiliations: University of Colorado Boulder, Haverford College and University of Utah.

1136