iiinova — where your digital self keeps evolving

shira mitchell @shiraamitchell.bsky.social · 19h

blog post: ANOVA follow-up The classical ANalysis Of VAriance table (Sum Sq, Mean Sq, F statistic, p-value) tests whether batches of effects are zero V(beta_j) = 0. But it doesn't directly show estimated variances V(beta_j).

030

shira mitchell @shiraamitchell.bsky.social · 23/09/2026

blog post: Fat Bear Week 2026 Happy Fat Bear Week to all who celebrate 🐻🎉 If estimating the bears' total size: under which outcome model does a Horvitz-Thompson (inverse probability) estimator do well ?

020

shira mitchell @shiraamitchell.bsky.social · 16/09/2026

blog post: ANOVA Andrew says "Anova is still important; it’s just been subsumed by hierarchical models." How does ANOVA relate to Survey Statistics ?

020

shira mitchell @shiraamitchell.bsky.social · 09/09/2026

blog post: @pewresearch.org finds “No Easy Fix for Bogus Respondents in Online Opt-In Polls” They try 3 screening methods and compare them with 3 data quality measures. How does screening affect estimates of 2024 vote choice ? How do opt-in and probability samples compare ?

020

shira mitchell @shiraamitchell.bsky.social · 02/09/2026

blog post: logit shift and raking a short post to see the logit shift as a special case of raking

000

shira mitchell @shiraamitchell.bsky.social · 26/08/2026

blog post: more on SynthMargins and Bayes-Raking We want to poststratify (MRP) but only have partial information (margins) about the poststratification variables in the population. More ideas from Bob Carpenter (in #mcmc_stan), @yajuansi.bsky.social, and @shirokuriwaki.bsky.social

070

shira mitchell @shiraamitchell.bsky.social · 19/08/2026

blog post: Modeling Complex Contingency Tables We want to poststratify (MRP) but only have partial information about the poststratification variables in the population. Max Goplerud, @shirokuriwaki.bsky.social, Jens Wiederspohn, @adamcs.bsky.social, and Philip Greengard have ideas !

031

shira mitchell @shiraamitchell.bsky.social · 12/08/2026

blog post: wanting workflow We review a covid survey case study from the new Bayesian Workflow book. We model: - measurement: test sensitivity and specificity - representation: differences between sample and population (MRP) How can workflow help us here ?

000

shira mitchell @shiraamitchell.bsky.social · 05/08/2026

blog post: structured MRP to smooth survey weights Adjusting for lots of variables can lead to very large weights. So @yajuansi.bsky.social, @trangucc.bsky.social, Jonah Sol Gabry, and Andrew Gelman turned to a structured MRP and its equivalent weights.

050

shira mitchell @shiraamitchell.bsky.social · 28/07/2026

blog post: equivalent models, equivalent weights (locally) with survey-weighting methods, we can ask: under which outcome models do they do well ? with outcome-model methods (like MRP), we can ask: what are the (locally) equivalent weights ?

020

shira mitchell @shiraamitchell.bsky.social · 22/07/2026

blog post: poststratification without population level information Poststratification uses population data on X to help estimate a population mean E(Y). But sometimes population data on X isn't available: In 2016 Andrew asked pollsters to poststratify on party ID, but how ?

010

shira mitchell @shiraamitchell.bsky.social · 14/07/2026

blog post: quantifying uncertainty in ranked choice voting polls RCV uses rankings to get a winner by instant runoff. Polls estimate rank probabilities with uncertainty. Unlike with non-RCV, even in random samples a plurality of uncertainty mass can get an incorrect winner.

240

shira mitchell @shiraamitchell.bsky.social · 08/07/2026

blog post: toy example for energy balancing weights How do energy balancing weights (used now by the NYT/Siena Poll) handle unsampled population groups ? Let's work thru a toy example and compare to Poststratification, Raking, and MRP.

140

shira mitchell @shiraamitchell.bsky.social · 02/07/2026

blog post: Big Changes in the Times/Siena Poll 2 changes to their survey weights: 1. new weighting variable: support score 2. new weighting method: energy balancing

021

shira mitchell @shiraamitchell.bsky.social · 24/06/2026

blog post: perfect collinearity in the sample but not in the population Two variables are perfectly collinear in your sample, so you drop one. You use your model to predict in the population. What can go wrong ? Let's talk thru a Census Bureau toy example from BDA2.

000

shira mitchell @shiraamitchell.bsky.social · 17/06/2026

blog post: using MRP in later analyses (pride edition) happy pride ! 🌈 @jeffreylax.bsky.social & Phillips 2009 used MRP to estimate state-level public opinion about policies affecting gays and lesbians. They then use this as a predictor of whether the state adopts the policies.

021

shira mitchell @shiraamitchell.bsky.social · 10/06/2026

blog post: should MRP workflow include LOCO-CV ? Individual-level loss orders models differently than the population-level loss we want judging MRP. To get population-level loss, use out-of-sample classical poststratification to compare with MRP. How to split data ? LOCO = leave one cell out.

000

shira mitchell @shiraamitchell.bsky.social · 03/06/2026

blog post: it is (still) the people Survey Statistics blog series' 1st birthday 🥳 Andrew Gelman's 60-ish Birthday 🥳 and NYT weights with synthetic past vote 🗳️

110

shira mitchell @shiraamitchell.bsky.social · 27/05/2026

blog post: double-plus robustness Meng (2022): GREG is not only “double robust” (consistent if either the outcome model or response model are correct), but “double-plus robust” (consistent if what is left of the outcome model and response model are uncorrelated).

010

shira mitchell @shiraamitchell.bsky.social · 20/05/2026

blog post: GREG GREG is Generalized REGression estimator. We can think of it either as: 1. Adjusting an estimate based on the model with a Horvitz-Thompson estimate of the error, or 2. On the flip side, adjusting the Horvitz-Thompson estimate with the model.

030

shira mitchell @shiraamitchell.bsky.social · 15/05/2026

blog post: relevant alternatives ? We saw that the multinomial logit model implies independence from irrelevant alternatives (IIA). Let’s expand the model to include choice set C within the logits f(X_ic,C), allowing for non-IIA.

021

Reposted by shira mitchell

shira mitchell @shiraamitchell.bsky.social · 10/03/2026

blog post: work with us at Blue Rose ! use cutting edge statistics, machine learning, and engineering to study public opinion, forecast elections, and advise Democrats.

012

shira mitchell @shiraamitchell.bsky.social · 29/04/2026

blog post: exploded logit ! a common choice model is multinomial logit. this model implies that rankings follow an exploded logit !

001

shira mitchell @shiraamitchell.bsky.social · 15/04/2026

blog post: irrelevant alternatives ? a common choice model is multinomial logit. this model implies Independence of Irrelevant Alternatives (IIA), e.g. the ratio of Left-vs-Right preference is the same in round 1 as in the runoff.

020

shira mitchell @shiraamitchell.bsky.social · 08/04/2026

blog post: improving with structure We’ve met Mr. P (Multilevel Regression and Poststratification). We’ve met Mrs. P (Multilevel Regression with Synthetic Poststratification). Now let’s meet Ms. P (Multilevel Structured regression with Poststratification).

121

shira mitchell @shiraamitchell.bsky.social · 01/04/2026

blog post: design-based cross validation how to split train and test sets to respect survey design ? what lessons carry over to nonprobability samples ?

2133

shira mitchell @shiraamitchell.bsky.social · 25/03/2026

blog post: Individualism and the CV Noise Problem Politically meaningful differences among models can be swamped by cross-validation noise.

010

shira mitchell @shiraamitchell.bsky.social · 18/03/2026

blog post: individualism doesn't work (even when weighted) individual-level loss (even weighted to the population) orders models differently than the population-level loss of interest to folks using MRP

010

shira mitchell @shiraamitchell.bsky.social · 10/03/2026

blog post: work with us at Blue Rose ! use cutting edge statistics, machine learning, and engineering to study public opinion, forecast elections, and advise Democrats.

012

shira mitchell @shiraamitchell.bsky.social · 03/03/2026

blog post: sampling-weighted loss we use sampling weights to estimate a population mean E(Y). what about to estimate a conditional mean E(Y|X) ? the best-fit model in the sample may not be the best-fit model in the population.

030

shira mitchell @shiraamitchell.bsky.social · 25/02/2026

blog post: sampling to assess data quality @bhedtgauthier.bsky.social et al. (2012) used sampling to assess and improve data quality in Malawi

021

shira mitchell @shiraamitchell.bsky.social · 17/02/2026

blog post: Gallup's Presidential Approval Ratings Gallup will no longer track presidential approval after 88 years Let's look at their sampling, mode, and weighting (still used for other survey questions)

010

shira mitchell @shiraamitchell.bsky.social · 11/02/2026

blog post: more on recalled vote we've talked about measurement error in recalled vote in the US. how does this change in multiparty states ?

000

Reposted by shira mitchell

David Shor @davidshor.bsky.social · 15/01/2026

We recently tested ~ a dozen public statements from a diverse set of Democratic elected officials on the murder of Renee Good and this was the top testing one

733410791

shira mitchell @shiraamitchell.bsky.social · 03/02/2026

blog post: 5 flavors of calibration 2 from survey statistics 1 from machine learning 2 from Gelman et al.'s workflow article

151

shira mitchell @shiraamitchell.bsky.social · 27/01/2026

blog post: Total Margin of Error (Part II) For election polls from 1998 to 2014 Shirani-Mehr et al. found: margin of error = 2 x (reported margin of error) Let's revisit Meng's “Statistical Paradises and Paradoxes” to understand this more generally.

031

shira mitchell @shiraamitchell.bsky.social · 21/01/2026

blog post: Total Margin of Error margin of error = 2 x (reported margin of error) and how much of this error is "bias" vs "variance" ?

193

shira mitchell @shiraamitchell.bsky.social · 18/01/2026

blog post: Margin of Error how can we get a poll's margin of error ? let's start with MRP and some simplifying assumptions.

020

Reposted by shira mitchell

Will Marble @wpmarble.bsky.social · 08/01/2026

This post has some more discussion of other methods for incorporating known ground-truth margins in an MRP framework, based on some validation exercises in osf.io/preprints/so...

osf.io

OSF

153

shira mitchell @shiraamitchell.bsky.social · 08/01/2026

blog post: 4th helpings of the logit shift y_1 = governor vote choice y_2 = abortion proposition vote choice x = demographics You want E(y_2 | county). You have y_1, y_2, x in a survey, x in the population, and E(y_1 | county). @wpmarble.bsky.social and Josh Clinton have ideas !

162

shira mitchell @shiraamitchell.bsky.social · 30/12/2025

blog posts: should we adjust for a mismeasured X ? You know the population distribution for X (e.g. vote choice in 2024). But you only have a reported X* in your survey. Should you adjust for it ? Later today: exploring toy examples to see.

151

shira mitchell @shiraamitchell.bsky.social · 16/12/2025

blog post: 3rd helpings of the logit shift You have multiple outcomes, but only some have aggregate truth to shift to. How can we calibrate our estimates of p(y_1, y_2 | X) to aggregate data about E[y_1] ? @wpmarble.bsky.social and Josh Clinton have ideas !

042

shira mitchell @shiraamitchell.bsky.social · 10/12/2025

blog post: 3 probabilities in Meng 2022 1. (human) design probabilities, e.g. P[R = 1 | stratum] in stratified sampling 2. divine probabilities, e.g. P[R = 1 | anything about a person] where responders follow laws of nature 3. device probabilities, e.g. P[R = 1 | X] modeled

040

shira mitchell @shiraamitchell.bsky.social · 05/12/2025

blog post: probability sample = known nonzero probability epsem = equal individual probabilities SRS = equal entire-sample probabilities

062

Reposted by shira mitchell

Anita Gohdes @argohdes.bsky.social · 06/08/2025

Coming soon: our introduction to the politics of human rights 🥳📚 Preorder available here: www.cambridge.org/highereducat... @sabinecarey.bsky.social

Cover of book, titled „The Politics of Human Rights“ by Sabine Carey, Mark Gibney, and Anita Gohdes. Picture shows a woman kneeling in front riot police during a BLM protest in the US.

25813

shira mitchell @shiraamitchell.bsky.social · 25/11/2025

blog post: quantity vs quality compare 2 surveys: 1. 100% coverage, but response probability P[R = 1 | Y] differs a lot by Y 2. Only 5% coverage, but P[R = 1 | Y] is roughly constant across Y which would you use ? both ?

011

shira mitchell @shiraamitchell.bsky.social · 19/11/2025

new blog post: sampling the sample we’ve focused on estimating means E[Y]. but say Y are openends ("describe how you feel about the candidate") and you want to read thru a few draws from the population, not only survey responders. what should you do ?

022

shira mitchell @shiraamitchell.bsky.social · 12/11/2025

blog post: weights and MRP for voters so far we've talked about weights and MRP for E[Y], vote choice in the population overall. but what if you want E[Y | V = 1], vote choice in the population of voters. what are the weights and how do you modify MRP ?

053