@chadtopaz
Data science + math for social justice. Violist, yogi, husband, dad to human & dogs. 🏳️🌈 Views do not represent my employers. Author of Unlocking Justice: The Power of Data to Confront Inequity and Create Change. Preorder at: https://bit.ly/4nT7qUh
Academic friends, friendly (honestly) reminder not to use words like "outstanding" as a descriptor for what you are looking for in your job ads. It doesn't help at all, and it *does* privilege the demographically biased audience of potential candidates who in fact self-perceive as outstanding.
The number of times per week we say “nuts and gum, together at last” in my household
Definitely a Geneva Conventions violation.
This scoping review synthesizes recently published research on news media coverage of gun violence in the United States. Following PRISMA-ScR guidelines, a systematic search of Web of Science and Google Scholar databases identified 76 peer-reviewed empirical studies published since 2000. Analysis reveals several recurring themes. First, news media coverage tends to be greater for incidents involving numerous victims, particularly women and/or children; occurring in schools, religious sites, or government buildings; or carried out by perpetrators who are young, ideologically driven, and/or show signs of severe mental illness. Second, race plays a significant role in shaping narrative frames, with media portrayals often depicting white individuals more sympathetically compared to racially minoritized individuals, regardless of whether they are victim or perpetrator. Third, media narratives display dynamic evolution over decades, transitioning from framing gun violence as isolated and episodic to addressing it as a part of broader social contexts and societal concerns. Finally, news coverage of gun violence is associated with negative audience emotions, especially fear, as well as with increased gun background checks suggestive of accelerated gun purchasing. Regarding media coverage's role in inciting further violence, however, studies present mixed findings. Limitations include the exclusion of social media discourse, which warrants study in future research. Future studies should also examine how news media coverage represents individuals across diverse gender identities, LGBTQIA+ communities, and disability statuses, while adopting intersectional perspectives. Three interrelated conclusions emerge from this analysis: the different types of gun violence that occur, the level of news media attention each receives, and the academic research that has investigated these media portrayals to date.
Personal academic news: the paper that I submitted 520 days ago has been accepted! It's on a topic that is meaningful to me bc of gun violence in this country and bc of the shooting that happened in my town when I was a kid.
Forthcoming: "News Media Coverage of Gun Violence: A Scoping Review"
🙏
This leaves only <checks notes> 13 submitted mansucripts languishing in editorial limbo.
This scoping review synthesizes recently published research on news media coverage of gun violence in the United States. Following PRISMA-ScR guidelines, a systematic search of Web of Science and Google Scholar databases identified 76 peer-reviewed empirical studies published since 2000. Analysis reveals several recurring themes. First, news media coverage tends to be greater for incidents involving numerous victims, particularly women and/or children; occurring in schools, religious sites, or government buildings; or carried out by perpetrators who are young, ideologically driven, and/or show signs of severe mental illness. Second, race plays a significant role in shaping narrative frames, with media portrayals often depicting white individuals more sympathetically compared to racially minoritized individuals, regardless of whether they are victim or perpetrator. Third, media narratives display dynamic evolution over decades, transitioning from framing gun violence as isolated and episodic to addressing it as a part of broader social contexts and societal concerns. Finally, news coverage of gun violence is associated with negative audience emotions, especially fear, as well as with increased gun background checks suggestive of accelerated gun purchasing. Regarding media coverage's role in inciting further violence, however, studies present mixed findings. Limitations include the exclusion of social media discourse, which warrants study in future research. Future studies should also examine how news media coverage represents individuals across diverse gender identities, LGBTQIA+ communities, and disability statuses, while adopting intersectional perspectives. Three interrelated conclusions emerge from this analysis: the different types of gun violence that occur, the level of news media attention each receives, and the academic research that has investigated these media portrayals to date.
Personal academic news: the paper that I submitted 520 days ago has been accepted! It's on a topic that is meaningful to me bc of gun violence in this country and bc of the shooting that happened in my town when I was a kid.
Forthcoming: "News Media Coverage of Gun Violence: A Scoping Review"
Better than waxing season, that’s for damn sure. www.berkshireeagle.com/business/wil...
Ugh, "SocArXiv"
Inefficiency and Inequity of the Law Review Submission System Chad M. Topaz1,2,3,* 1Williams College, Williamstown, MA, USA 2University of Colorado–Boulder, Boulder, CO, USA 3QSIDE Institute, Williamstown, MA, USA *Corresponding author: cmt6@williams.edu Abstract Where a legal scholar works shapes publication outcomes nearly as much as what they write. In the law review submission system—the primary publication market for legal scholarship in the United States—student editors face thousands of submissions for a handful of slots and rely heavily on institutional prestige as a proxy for article quality. We build a calibrated agent- based simulation of this market and benchmark it against deferred acceptance, a centralized matching algorithm used in markets like medical residencies. The simulation predicts severe misallocation: more than 60% of top-tier placements differ from what centralized signal- based matching would produce, and the rank correlation between article quality and journal prestige is 0.45 versus 0.79 under centralized matching. Which system produces better placements overall depends on how many authors are competing for how many slots. As competition intensifies—a trend already underway—the current system’s disadvantage grows, with the model predicting up to 13.4% loss in match quality. Partial reforms like extending deadlines have negligible effects; in the simulation, the primary source of inefficiency is the decentralized structure of the market itself. The simulation also reveals that credential dependence produces inequity that persists even among articles of comparable quality: authors from prestigious institutions receive markedly better placements regardless of the matching mechanism. Centralized matching fixes the sorting problem but not this equity problem— prestige bias is embedded in editorial signals and would require changes to how articles are evaluated, not just how they are assigned.
This'll be my last post on this (unless/until publication) but the fruits of my rage are now "officially" posted on SoxArXiv and have been submitted for publication, yay!
"Inefficiency and inequity of the law review submission system"
Link: osf.io/preprints/so...
Ah, the two genders
Model assumes that articles have "quality" which we cannot know perfectly and which is noisily perceived by editors. These are all parameters that are varied in the model. See results.
My extra radical far loony left position is that no one should ever be allowed to make a profit from academic publishing. It is a public good. See also: education, healthcare (on a good day, public transport…)
Anyone in my orbit in the exclusive club of 9,000 handling editors for PLOS One? Asking for real.
Paper title and abstract: Inefficiency and Inequity of the Law Review Submission System (author: Chad M. Topaz) Where a legal scholar works shapes publication outcomes nearly as much as what they write. In the law review submission system---the primary publication market for legal scholarship in the United States---student editors face thousands of submissions for a handful of slots and rely heavily on institutional prestige as a proxy for article quality. We build a calibrated agent-based simulation of this market and benchmark it against deferred acceptance, a centralized matching algorithm used in markets like medical residencies. The simulation predicts severe misallocation: more than 60\% of top-tier placements differ from what centralized signal-based matching would produce, and the rank correlation between article quality and journal prestige is 0.45 versus 0.79 under centralized matching. Which system produces better placements overall depends on how many authors are competing for how many slots. As competition intensifies---a trend already underway---the current system's disadvantage grows, with the model predicting up to 13.4\% loss in match quality. Partial reforms like extending deadlines have negligible effects; in the simulation, the primary source of inefficiency is the decentralized structure of the market itself. The simulation also reveals that credential dependence produces inequity that persists even among articles of comparable quality: authors from prestigious institutions receive markedly better placements regardless of the matching mechanism. Centralized matching fixes the sorting problem but not this equity problem---prestige bias is embedded in editorial signals and would require changes to how articles are evaluated, not just how they are assigned.
🚨What if some bitches simulated the bonkers law review submission market to show exactly how it's wildly inefficient and deeply inequitable? It's me, I'm bitches!
(Here's a *much* refined version of yesterday's preprint.)
Sharing is caring!
#LawSky #AcademicSky
drive.google.com/file/d/1dsDm...
Ok but for real, asking for advice. Where should I send work like this? I actually have no idea.
Told my kid that with St Patrick's Day coming up, he should get ready for a week of me making Irish-themed dad jokes. He glared at me. "Glare all you want, I'm Dublin' down on it."
🎯
Said as a very quanty person: some dude I don’t know on here shitting on academic fields for having theoretical frameworks and not using causal inference… earns an instablock.
In other news, trans rights are human rights.
We are a three-electric-car family and it’s never been a better time.
Yes exactly!!
omg I don't know how I missed this three months ago. I'm ☠️.
It's interesting for me to ponder this. I have been using some Claude Code. I'm at a tiny undergrad-only school, where I only get, at best, a semester to work on research with a student. The LLM hasn't replaced any person's labor but it has helped finish projects that I would otherwise leave undone.
^^^This.
Lol I have not submitted anywhere because I am honestly not sure where to send it but I am SHAMELESSLY solicit social media here. Tell all your friends or students or whomever. First one to solicit this article from me gets it. 🙃
For some of the folx who chimed in yesterday: there's a much-refined version posted now. Just click through to skeet below. @hoffprof.bsky.social @bdgesq.bsky.social @davidasimon.bsky.social @msmith750.bsky.social @profferguson.bsky.social @narosenblum.bsky.social
oh p.s., I have not yet submitted this anywhere and I literally dare any legal publication to solicit this from me. 🙃
Take note, @abovethelaw.com !
