Tired: Dos a Cero
Wired: 13-10
๐
@bradytwest.bsky.social
Research Professor @ Univ. of MI Institute for Social Research (ISR) and Department of Biostatistics. Usually discussing statistics, surveys, sports, music, and leadership. Personal website: www.umich.edu/~bwest
Tired: Dos a Cero
Wired: 13-10
๐
blog post:
probability sample = known nonzero probability
epsem = equal individual probabilities
SRS = equal entire-sample probabilities
OPINION TODAY featured podcast โ
Dealing with Selection Bias in Surveys, with Rod Little, @randridge.bsky.social and @bradytwest.bsky.social
(@aapor.bsky.social)
opiniontoday.substack.com/i/180586526/...
come work with us! @bluelabs.bsky.social is hiring a survey scientist. we're looking for someone who loves working with survey data and wants to solve tricky problems facing the polling industry. please share widely!
job-boards.greenhouse.io/bluelabsanal...
Hard disagree. The official BLS jobs numbers are still the most reliable data we have, and it's not close. And there's no way the Administration has (yet) fiddled with its data. I promise to let you know at the first whiff of interference.
03.12.2025 13:47 โ ๐ 737 ๐ 134 ๐ฌ 29 ๐ 7I had fun hosting this podcast for @aapor.bsky.social with Rod Little and @randridge.bsky.social, where we discussed modern approaches to dealing with selection bias in surveys (and why weights may not work). I hope that AAPOR will continue to make the P.O.P. available!
aapor.org/media/public...
Yet another excellent example of why it is important to always think critically about from where (or whom!) exactly the data are coming.
01.12.2025 15:34 โ ๐ 4 ๐ 0 ๐ฌ 0 ๐ 0For the avoidance of doubt: does svyglm() use robust standard errors? Yes!
notstatschat.rbind.io/2025/12/01/d...
Needed - larger samples, more realism about (the lack of) heterogeneous treatment effects:
-"less than a third of proposed hypotheses were supported... the largest predictor of positive exp. results was sample size"
-"moderation hypotheses were rarely significant"
academic.oup.com/poq/advance-...
Just a miserable failure by the Lions coaching staff today. Hasnโt been great for multiple games, but this is a nadir. And now the playoffs are in jeopardy.
27.11.2025 21:00 โ ๐ 2 ๐ 0 ๐ฌ 0 ๐ 0Really bad time for one of your better defensive players to just glitch outโฆWTF?
27.11.2025 18:50 โ ๐ 1 ๐ 0 ๐ฌ 0 ๐ 0Iโve never seen anything like what Michigan basketball is doing right now. Bonkers. If they hit shots at this kind of clip, look out, because the D should be there every night. #HAIL
27.11.2025 04:08 โ ๐ 1 ๐ 0 ๐ฌ 0 ๐ 0Our new #rstats package. This one implements and extends Lee et al. (2022) maximum entropy classification so you can automate your record linkage pipeline. Let us know what you think! cc @tslumley.bsky.social @vincentab.bsky.social @bradytwest.bsky.social @bschneidr.bsky.social
26.11.2025 19:09 โ ๐ 4 ๐ 1 ๐ฌ 1 ๐ 0This was amazingโฆ!
All the makings of a pretty special team. #HAIL
Hanyu Sun & Fred Conrad examine the differences in interviewer and respondent paralinguistic behaviors (such as โuhโ and โumโ or laughter) between video and in-person interviews and whether these difference influence respondents' answers myumi.ch/rAPZ7
24.11.2025 15:40 โ ๐ 1 ๐ 1 ๐ฌ 0 ๐ 0PDHP workshop
Mark your calendar for the next, hybrid half-day PDHP Workshop with @d3center.bsky.social experts, covering Adaptive Interventions: Dec. 12! Register at pdhp.isr.umich.edu/workshops/ @leb112358.bsky.social
24.11.2025 15:48 โ ๐ 3 ๐ 4 ๐ฌ 0 ๐ 0The Lions defense has been beyond awful all day. Like, what is this? You are letting Jameis Winston do whatever he wants. Just absolutely inexcusable.
23.11.2025 20:52 โ ๐ 0 ๐ 0 ๐ฌ 0 ๐ 0Heads up! If you are interested in pursuing graduate studies in survey and data science (read: all things quantitative social science) at the University of Michigan, please check out these upcoming info sessions!
21.11.2025 21:24 โ ๐ 8 ๐ 4 ๐ฌ 0 ๐ 0So great to see BLS back to work measuring and informing us about our large and dynamic economy.
Itโs a great institution full of dedicated civil servants deeply committed their mission. They have had a tough year, and everyone is covering for someone who left. I thank them all for their service.
Hey authors! Check to see if Anthropic stole your book to train their slop generator on. Youโre entitled to $1500 per stolen Work.
Look up your work, and if youโre in the database, file a claim
secure.anthropiccopyrightsettlement.com/lookup/
How about these Pistons??? ๐ฎโ๐จ๐ค๐ฏ
19.11.2025 03:01 โ ๐ 0 ๐ 0 ๐ฌ 0 ๐ 0Webinar recording now available: myumi.ch/kPRXk
ICYMI: watch the #sbeccc webinar and view the presentation slides from @bradytwest.bsky.social on ICPSR's YouTube channel: myumi.ch/kPRXk
18.11.2025 20:26 โ ๐ 2 ๐ 1 ๐ฌ 0 ๐ 0Meet Zachary Winograd, student at the Michigan Program for Survey and Data Science and research assistant at @d3center.bsky.social at the Survey Research Center (SRC)!
Learn more about the MPSDS and students here: surveydatascience.isr.umich.edu
#masters #research #datascience
Chat GPT tells me that what happened in the Michigan football game today has never happened before (Div. 1, playing on road, 5 turnovers, 2 missed field goals, still win):
15.11.2025 20:57 โ ๐ 5 ๐ 0 ๐ฌ 0 ๐ 0interesting little "need to get your shit together asap" moment here for ol' Sherrone Moore's Michigan
15.11.2025 20:10 โ ๐ 57 ๐ 5 ๐ฌ 6 ๐ 1OK, itโs clear to everyone and their brother that Michigan cannot trust Zvada. Like, enough. Sherrone needs to go for it on 4th and short, every time.
15.11.2025 17:22 โ ๐ 0 ๐ 0 ๐ฌ 0 ๐ 0A road win is a road win, but man Michigan has got to fix the careless turnover problem. Been an issue for two straight years with completely different playersโฆ๐คจ
15.11.2025 04:28 โ ๐ 0 ๐ 0 ๐ฌ 0 ๐ 0Survey Statistics: weights and MRP for voters
statmodeling.stat.columbia.edu/2025/11/11/s...
Making connections at #GSA2025? The @umisr.bsky.social Institute for Social Research is an extraordinary hub for studying aging. Follow us all with starter pack tabs from @um-psc.bsky.social @um-src.bsky.social and @rcgd-isr.bsky.social, and comment to be added! go.bsky.app/JcQbemD
11.11.2025 14:44 โ ๐ 9 ๐ 6 ๐ฌ 0 ๐ 0