BLUE
Profile banner
SK
Sebastian Karcher
@adam42smith.bsky.social
Associate Director Qualitative Data Repository (personal account). Data, Zotero, Social Science Methods sebastiankarcher.com
336 followers321 following706 posts
SKadam42smith.bsky.social

Also, IMO, much more interesting intellectually to work on. The Bak-Coleman/Devezer critique of the Protzko et al piece is a good ex. They really wanted to say "the design is bad & the estimator can't yield the estimand" -- but had to do a bunch of tedious work bc all ppl cared about was cheating

2
SKadam42smith.bsky.social

Used to be. Didn't work this way under Trump and w Biden, since 2023 you need to apply through an app. Details (and the many problems w it) e.g. here: www.hrw.org/news/2023/05... (additional restrictions were added this summer).

US: Biden ‘Asylum Ban’ Endangers Lives at the Border
US: Biden ‘Asylum Ban’ Endangers Lives at the Border

President Joe Biden’s new border plan focused on deterrence will almost certainly lead to a rise in the already record number of migrants dying at the United States southern border, enrich criminal ca...

0
Reposted by Sebastian Karcher
AMawmercer.bsky.social

Look, polls have biases. But slso, the MOE for a subgroup of n=96 is about +/- 10 points. Add in the design effect, it’s +/- 13 points. For the vote margin it’s +/- 26. For the change in the margin since the last poll, it’s about +/- 50 points. That’s just accounting for random sanpling.

5
SKadam42smith.bsky.social

Are you surprised Gemini is so high? Everyone I know uses Claude or ChatGPT

1
SKadam42smith.bsky.social

As I said, hard to know, but since we _do_ know that ~80% of people use google every day, we can say that the vast majority of respondents doesn't

1
SKadam42smith.bsky.social

It's hard to interpret self-reportes usage of anything, but I'm pretty sure folks using it accidentally or seeing Google AI summaries won't report that (not least bc definitely more than ~25% of folks uses google & sees the summaries

1
SKadam42smith.bsky.social

After a small (genuinely unpleasant) uptick in property crime, half of our 'in this house...' neighborhood is ready to either roll back (very moderate) crim justice reform to 1996 or institute mob justice. It's dispiriting.

1
SKadam42smith.bsky.social

The basic model is not so unlike OLH: Institutions pay a fee for read& publish, richer institutions pay more, researchers from poorer institutions & countries get waivers. Main differences I see is that OLH is lower cost & transformative agreements will get $ from funders via APCs.

0
Reposted by Sebastian Karcher
RTretrotechdreams.bsky.social

Counter-Strike LAN party in Germany (2003)

1
Profile banner
SK
Sebastian Karcher
@adam42smith.bsky.social
Associate Director Qualitative Data Repository (personal account). Data, Zotero, Social Science Methods sebastiankarcher.com
336 followers321 following706 posts