Raphael Nishimura
@rnishimura.bsky.social
Survey methodologist and statistician at the Survey Research Center, Institute for Social Research, University of Michigan #survey | #sampling | #statistics | #RStats
@jackmiller.bsky.social and i went to #AAPOR in may and wrote up some of our findings and others' interesting research for the @bluelabs.bsky.social blog! disengaged voters - how to reach them & why they matter: bluelabs.com/dynamic-elec... AI fraud in online panels - bluelabs.com/when-ai-come...
Em pleno 2026 e tem gente que ainda leva AtlasIntel a sério... Se você realmente acha que não tem nada de problemático com a metodologia deles, eu tenho um princípe nigeriano aqui para te apresentar... 😅
C'mon folks, what are we even doing here? Pollster self-proclaims themselves as the most accurate pollster in the world, according to analyses conducted by the pollster themselves using metrics defined by the very same pollster. Shameless self-promotion with a pinch of conflict of intererest 😅
Glad to see that the NYT (Nate Cohn) has finally decided to join the past vote weighting team! :) Their "synthetic past vote" approach is quite similar to what I've been advocating: using the voter file to reduce turnout overreporting. www.nytimes.com/2026/05/18/u...
A Few Changes for the Times/Siena Poll
nytimes.com
É frustante ver gente tomando isso pelo valor de face, sem o menor senso crítico, e depois sai por aí repetindo que esse foi o instituto que mais "acertou" as eleições. Por conta disso, coloco nesse fio alguns dos problemas e limitações dessa comparação: 🧵+
The methodology of publicly published pre-election polls shouldn't be a black box!
Bolsonaristas surtados com a AtlasIntel no Brasil, enquanto nos EUA eles veem sofrendo hate dos Trumpistas há meses, e eu só assim...
Michael Jackson Comendo Picoca
ALT: Michael Jackson Comendo Picoca
static.klipy.com
Our @um-src.bsky.social super-star, no introductions needed, @bradytwest.bsky.social delivering his presentation on how to improve survey data integration between probability and non-probability samples at #AAPOR #AAPOR26 #AAPOR2026
Back to LA for #AAPOR! Many places to visit, lots of people to see and not enough meals for all the restaurants we need to go! Oh, and of course, brand new things to share and to learn in survey research!
If you think we had already talked enough about weighting on "recalled vote", think again! This time, we will be presenting some interesting results of the effects of this approach on measures of association under a missing not at random mechanism.
Center for Political Studies@umisrcps.bsky.social · 3mo agoUsing the 2024 American National Election Studies, Matthew DeBell et al use simulations to investigate when weighting on “recalled vote” (asking survey participants who they voted for in a previous election and weighting voters to match the election outcome) can correct biased associations.
Celebrating 80 years of the world-renowned Survey Research Center of the University of Michigan! 🥳🎉 What a huge honor to be part of this history!
The contrast of me giving an interview and @joywilke.bsky.social giving an interview, despite both of us being under the same roof!
👀👀👀👀
G Elliott Morris@gelliottmorris.com · 4mo agoNew podcast with @joywilke.bsky.social of the data analytics company BlueLabs covering what new focus group interviews in Michigan tell us about economic anxiety, anti-system attitudes, and faith (or lack thereof) in the parties today. www.gelliottmorris.com/p/podcast-jo...
This is an interesting illustration of potential bias due to non-ignorable selection: respondents from a political poll recruited by ads on Instagram reporting that their main source of information about politics is Instagram.
Partisan nonresponse bias in a nutshell
Polling USA@usapolling.bsky.social · 4mo agoGallup's Quarterly Partisan Affiliation Poll Since 2020 Q2
A couple of examples of what could happen to the betting odds on these types of platforms if there were no election polls:
Às vezes me perguntam quais considero os melhores institutos de pesquisa no Brasil. Como metodologista e estatístico, prefiro me concentrar nas melhores práticas metodológicas. Então fica aqui uma colinha para o que considero as melhores prática metodológicas em pesquisas de opinião no Brasil: 🧵+
Me at looking at the logos of Michigan, Oklahoma and Eastern during the gymnastics meet: "Margin Of Error" 😜
Love surveys? Work with the best!
Joy Wilke@joywilke.bsky.social · 6mo agoLove surveys? We're hiring! We're looking for new members of the polling team at @bluelabs.bsky.social. Two roles are focused on execution of day-to-day polling operations (Polling Manager, Lead Polling Manager) while the Survey Scientist is an R&D role! Share & apply! bluelabs.com/careers/
Love surveys? We're hiring! We're looking for new members of the polling team at @bluelabs.bsky.social. Two roles are focused on execution of day-to-day polling operations (Polling Manager, Lead Polling Manager) while the Survey Scientist is an R&D role! Share & apply! bluelabs.com/careers/
Careers - BlueLabs
We’re always looking to add new talent to our growing team of strategists, data scientists, and software engineers who operate at the forefront of data
bluelabs.com
Huge loss to the public opinion field!
Scott Clement@sfcpoll.bsky.social · 6mo agoNews: Gallup will stop measuring presidential approval ratings after 89 years. Read the firm's reasons for making the change and what topics the Gallup will continue conducting in today's Early Brief s2.washingtonpost.com/camp-rw/?tra...
One of my many pet peeves is people who denote survey sample sizes by N. In surveys statistics, because we are dealing with finite population inferences, we reserve the notation N for population size and use n for sample size.
The Shades of ISR Pantone ™️ palette 💙 💛
I always thought this rule that the total error is 2x reported MoE nonsensical, because bias in surveys doesn’t typically scale with sample size. In fact, we usually see the opposite: smaller, more carefully designed prob samples tend to present smaller total error than large nonprob samples.
shira mitchell@shiraamitchell.bsky.social · 7mo agoblog post: Total Margin of Error margin of error = 2 x (reported margin of error) and how much of this error is "bias" vs "variance" ?
The Survey Research Center at the Institute for Social Research of the University of Michigan turns 80 years old this year. Very proud to be part of this history! youtu.be/giSsP7o4PZg?...
We Are SRC
YouTube video by Survey Research Center, University of Michigan
youtu.be
Resultados da pesquisa Meio Ideia, primeira pesquisa pré-eleitoral nacional de 2026 registrada no TSE (BR-06731/2026). Pesquisa telefônica com 2 mil entrevistas realizadas entre 08/01 e 12/01. Perguntas espotânea e estimuladas com cenários de 1° turno com Tarcísio e Flávio): +
I remember that when I would urge media to talk more about winning probabilities derived by forecasters/aggregators in their polling coverage, people would tell me the general public would not understand it. I guess the media decided that this is not a problem anymore? www.cnbc.com/2025/12/04/c...
CNBC and Kalshi Strike Exclusive Partnership
cnbc.com
What would once be the dream of any subjectivity Bayesian -- the popularization of using wagering odds as probabilities, is now becoming a real dystopia... 😫
