Tamkinat Rauf

@tsrauf.bsky.social

Asst. professor of Sociology Interests: happiness, depression, inequality, social psych, genomics, quant methods, open science www.tamkinatrauf.com

In the early 2000s, modest weekday social interaction built toward big peaks on Friday & Saturday nights. Friday night now looks like a typical weeknight & Saturday is less elevated than before. Overall, time with friends has fallen by more than half. journals.sagepub.com/doi/full/10....

The End of Friday Nights with Friends - Neal Caren, 2026

This study examines how Americans’ time with friends has changed over the week, focusing not on how much social interaction has declined but on when it has. Usi...

journals.sagepub.com

As part of our guidance for the upcoming changes to OSF Projects, we developed a help guide on choosing a repository and a list of repository options to help researchers identify potential places to store their research outputs. ℹ️ Guidance on choosing a repository:

Where should I share my Data, Materials, Code now? - OSF Support

This Article Is Licensed Under CCO For Maximum Reuse. The following is a Table of Contents that links directly to specific sections within the guide: Table of R

help.osf.io

New WP!📄📣 For anyone working with quasi-experiments (especially UESD) and survey data, this is for you. Treatment events may affect not only opinions, but also survey participation and thereby bias causal estimates. I develop a framework to assess and tackle this bias. osf.io/preprints/so...

Bild

Interesting comment on using p-values in analyses of non-random samples/pop census data. We typically operate with a “superpopulation” assumption in such cases. But is that assumption relevant when investigating a historical question: i.e. did x lead to y in place z? (Not: does x lead to y?)

David Bartram@dbartram.bsky.social · 4mo ago

I worry about adopting a “no effect” conclusion, in particular. Here’s a critique of a similar conclusion, using a data structure with some affinities (different units, though). osf.io/preprints/so...

🚨New in PNAS: www.pnas.org/doi/abs/10.1... Genetic principal components (PCs) are commonly used to adjust for population stratification in genetic studies. In this paper, we quantify how much direct genetic signal is lost when using these “catch-all” adjustments. 🧵

Quantifying direct genetic signal captured by principal component adjustment | PNAS

Contemporary genomic studies of complex traits, such as genome-wide association studies and polygenic index (PGI) analyses, frequently include the ...

pnas.org

I’m very surprised by these comments. I genuinely enjoyed this book. Even if one was skeptical about the proportion of variance explained by genes (no one’s reasonably going to argue that it’s zero), I thought she poses interesting questions that we all should be thinking about.

Jeff Lockhart@jwlockhart.bsky.social · 4mo ago

"None of this deepens our understanding of the scientific or religious questions posed, and again, that may not be the point. The book reads like an Eat Pray Love of biotech grifting" - @jonothingeb.bsky.social demonstrating criticism is an art form.

Is social media dying? How much did Twitter change as it became X? Which party now dominates the conversation? Using nationally representative ANES data from 2020 & 2024, I map how the U.S. social media landscape has changed Here are the key take-aways 🧵 Full paper out now in in JQD:DM!

Shifts in U.S. Social Media Use, 2020–2024: Decline, Fragmentation, and Enduring Polarization | Journal of Quantitative Description: Digital Media

journalqd.org