Mark Igra

Mark Igra

@markigra.bsky.social

Sociology PhD at the University of Washington, studying altruism and morality. Former software guy. Started sciences.social.

405 followers651 following274 posts

“If I could understand myself better, I'd probably understand a lot more about the world and certainly about my country, in which so many people seem to revere their persecutors…” This bit from Arundhati Roy on the mother who tormented Roy and her brother gets to the heart of much sociology.

Mark Igra reposted

Highlighting the (automatic) Cost of Living Adjustment in a *headline* about inflation is a choice by NPR. Maybe they know all their listeners are on social security, but even then, a COLA is not news.

Mark Igra reposted

Amazon thinks I have an “ongoing academic reading habit”. I wish. Sadly browsing isn’t acquiring & acquiring isn’t reading. Plus, local bookshop can order most academic books.

This post is unavailable.

Mark Igra reposted

The Capabilibara project seeks to understand how a language model learns to interpret people's beliefs, emotions, intentions, and everyday moral choices. We trace that ability back to the training data using influence functions and unlearning hcai-lab-gt.github.io/capabilibara/ Find us at COLM!

Mark Igra reposted

+++ PUBLICATION ALERT +++ My first solo-authored PhD paper is finally out in Network Science! I am so happy that it finally happened! TL;DR: Representatives in Congress do not just disagree on issues, they oftentimes do not even talk about the same things at all. www.cambridge.org/core/journal...

cambridge.orgMeasuring issue-level polarization in congressional speech using community detection-algorithms | Network Science | Cambridge CoreMeasuring issue-level polarization in congressional speech using community detection-algorithms - Volume 14

Mark Igra reposted

"The Department of Jewish Studies and the Department of Sociology invite applications for a tenure-track Assistant Professor position (Segal Chair in Jewish Studies), commencing August 1, 2027." Come work with us at McGill! More information can be found here: www.mcgill.ca/jewishstudie...

mcgill.caJob PostingsFor course lectureship opportunities in Winter 2027, please see the Course lectureships tab. Assistant Professor Tenure-Track Faculty Position in Jewish Studies, with a Second Specialization in Socio...

For other nerds who do what @kjhealy.co calls Modern Plain Text computing, how do you keep quarto/markdown/obsidian files synced across devices (e.g. iPad/laptop)? In a pinch I used a file sync service & it messed up my git history. But constant git commits are painful, not iPad friendly.

Mark Igra reposted

"Replication crisis" reforms like preregistration & pre-analysis plans were developed to control false positives in experimental research. Does that work in observational studies? Our new paper in Sociological Methods & Research argues for a different approach. 🧵 journals.sagepub.com/doi/10.1177/...

Mark Igra reposted

I'll be in New York on Oct 20 to talk about The Truth Tellers, a new film by Pamela Yates and Paco de Onís, and about documenting immigration detention right now. With @documentedny.bsky.social. Tue, Oct 20, 6 p.m., Columbia SIPA, NYC. Get tickets: givebutter.com/Documented-T...

givebutter.comPrivate Screening: Borderland – The Line Within. A Conversation on Immigration Detention, Accountability, and Who Profitsimmigration, media, New York, Borderland

Mark Igra reposted

Replying to @catacalypto.bsky.social

it's kind of the dominant culture of text based social media. People who could read critically but prefer to offload the work of interpretation by skipping the article and going straight to the comments so they have a quick digest of what they should think about something

Lower percentage of daily requirements for sodium than I thought. Then again, 1942 servings per container. “INGREDIENT” (singular) is a flex

Seeking courage in getting my book proposal together I came across this set of rejections for what became the Immortal Life of Henrietta Lacks by Rebecca Skloot. (Of course nobody proudly posts the wall of rejections from a proposal that never amounted to anything.)

I am confused by this. Isn’t it Congress’s job to set taxes? Are there limits to fees that the executive branch can set? Is this a “major question”? (Not a fan of major questions doctrine, but if it exists it’s worth establishing boundaries).

Michael Clemens@mclem.org

The US Administration is poised to announce a new $100,000 tax on international students who work even briefly after graduation. This is not the H-1B tax, it's new and additional. Today @iza.org released my new estimates of the economic effects of that tax —>

Mark Igra reposted

This is APPALLING. The Trump administration is detaining multiple random Somali immigrants with no criminal record at Guantanamo Bay, seemingly because the White House ordered ICE specifically to detain Somalis at Gitmo as part of its effort to punish an ethnic group Trump has attacked.

amanda moore 🐢@noturtlesoup17.bsky.social

“The worst place ever I’ve been in my life," Sakariye Abdi Mohamed, a Somali asylum seeker, told me over the phone from Guantanamo Bay last week. Sakariye is one of 7 or 8 Somalis who are being held at Gitmo. They say guards tell them the order to keep Somalis at the facility comes from DC.

Appreciated the metaphor as a conversation starter. In short, soc focuses on -do (internal fights by known rules & inculcation of shared culture) rather than -jutsu (fighting to win against external forces). But discussion/conclusion not sociological enough for audience (author rolls eyes here)...

Torbjørn Skardhamar@skardhamar.bsky.social

Some guy said sociology is a martial art. What if we take that metaphor seriously? I was not sure how serious this paper would be when I started. Turned pretty serious. This is about disciplinary survival. link.springer.com/article/10.1...

View full thread

Certainly SOC needs to be much better about data sharing. Causal modeling is nice when possible (and used in SOC!), but rarely broadly convincing. Not sure that computation critique is valid: there's lots! Used right AI can increase productivity, but likely a drag on credibility (training bias...).

Appreciated the metaphor as a conversation starter. In short, soc focuses on -do (internal fights by known rules & inculcation of shared culture) rather than -jutsu (fighting to win against external forces). But discussion/conclusion not sociological enough for audience (author rolls eyes here)...

Torbjørn Skardhamar@skardhamar.bsky.social

Some guy said sociology is a martial art. What if we take that metaphor seriously? I was not sure how serious this paper would be when I started. Turned pretty serious. This is about disciplinary survival. link.springer.com/article/10.1...

In Skardhamar's view the relevant battle is among disciplines, and soc loses by being short on credibility (anti-positivist, dismissive of causal modeling, data sharing), and behind on AI/computation. Some valid critiques, but is that really the battle? What are the rules of the game? Who decides?