Bayesian Multilevel Estimation with Poststratification: State-Level Estimates from National Polls
The DOI matches its Crossref record; 14 of the 15 references it cites were found in a registry or an index.
Abstract
We fit a multilevel logistic regression model for the mean of a binary response variable conditional on poststratification cells. This approach combines the modeling approach often used in small-area estimation with the population information used in poststratification (see Gelman and Little 1997, Survey Methodology 23:127–135). To validate the method, we apply it to U.S. preelection polls for 1988 and 1992, poststratified by state, region, and the usual demographic variables. We evaluate the model by comparing it to state-level election outcomes. The multilevel model outperforms more commonly used models in political science. We envision the most important usage of this method to be not forecasting elections but estimating public opinion on a variety of issues at the state level.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
The source backs part of the claim, not all of it — or backs it for a different population, condition or comparison.
In the citing paper
“small-area methods that borrow information across subgroups 48 can perform better with higher-quality-albeit few-data”
From Unrepresentative big surveys significantly overestimated US vaccine uptake (2021) · cites this paper as [48]
Experimental — accuracy not yet measuredThe cited source says what the claim attributes to it.
In the citing paper
“The multilevel model allows more accurate inferences-the usual partial pooling or Bayesian rationale [see Park, Gelman and Bafumi (2004)]”
From Analysis of variance—why it is more important than ever (2005) · cites this paper as Park, Gelman and Bafumi (2004)
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Gelman, A., Carlin, J. B., Stern, H. S., & Rubin, D. B. (2003) · doi:10.1201/9780429258480 · Cited by 27,276
Self-citationHierarchical Linear Models: Applications and Data Analysis Methods.
Longford, N. T., Bryk, A. S., & Raudenbush, S. W. (1993) · Contemporary Sociology · doi:10.2307/2075823 · Cited by 18,724 · cited 2×
“See Kreft and de Leeuw (1998), Snijders and Bosker (1999), Bryk and Raudenbush (2001), and Gelman et al. (2003) for introductions to multilevel models from classical and Bayesian perspectives.”
BackgroundMethodMultilevel analysis : an introduction to basic and advanced multilevel modeling
Snijders, T. A. B., & Bosker, R. (1999) · Cited by 13,786
Hierarchical Linear Models: Applications and Data Analysis Methods
Davidian, M. (2003) · Journal of the American Statistical Association · doi:10.1198/jasa.2003.s288 · Cited by 13,723
Introducing Multilevel Modeling
Kreft, I., & de Leeuw, J. (1998) · doi:10.4135/9781849209366 · Cited by 2,997
Bayesian Data Analysis, 2nd edn
Brix, A. (2005) · Journal of the Royal Statistical Society Series A: Statistics in Society · doi:10.1111/j.1467-985x.2004.00347_4.x · Cited by 1,767
Prior distributions for variance parameters in hierarchical modelsPreprint
Gelman, A. (2004) · RePEc: Research Papers in Economics · Cited by 1,632
Self-citationMeasuring Citizen and Government Ideology in the American States, 1960-93
Berry, W. D., Ringquist, E. J., Fording, R. C., & Hanson, R. L. (1998) · American Journal of Political Science · doi:10.2307/2991759 · Cited by 1,592 · cited 2×
“In the early 1970s, Weber et al. (1972–1973) expanded the number of voter types from 480 to 960 categories and were able to estimate state-level opinions on a variety of issues (for recent research on state-level estimates, see Erikson et al. 1993, Berry et al. 1998, and Brace et al. 2002).”
BackgroundInfluentialStatehouse democracy: public opinion and policy in the American states
(1995) · Choice Reviews Online · doi:10.5860/choice.32-3008 · Cited by 1,216
Why Are American Presidential Election Campaign Polls So Variable When Votes Are So Predictable?
Gelman, A., & King, G. (1993) · British Journal of Political Science · doi:10.1017/s0007123400006682 · Cited by 846
Self-citation
Research works printed but not linked: 5
- Burdick, 1964, The 480
- Spiegelhalter, 1999, WinBugs Version 1.4
- Park D. 2004. “Multilevel Models of Representation in the U.S. States.” Ph.D. dissertation. Department of Political Science, Columbia University.
- Gelman, 1997, Postratification into Many Categories Using Hierarchical Logistic Regression, Survey Methodology, 23, 127
- 2003, R: A Language and Environment for Statistical Computing
Cited by
Items 501–560 of 560
Diversity-aware population modeling
(2025) · Nature Computational Science · doi:10.1038/s43588-025-00787-9
The Big, The Bad, and The Ugly: Geographic estimation with flawed psychological dataPreprint
Hoover, J., & Dehghani, M. (2018) · PsyArXiv (OSF Preprints) · doi:10.31234/osf.io/bthqc_v1
How Responsive Is Local Immigration Enforcement to Constituency Preferences?Preprint
Dagonel, A., & Magazinnik, A. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5433734
Forecasting Election Outcomes: A Bibliometric Analysis
Tsiligkaridis, G. C., Mallidis, I., & Giannarakis, G. (2025) · IGI Global eBooks · doi:10.4018/979-8-3373-1037-4.ch001
Zhang, Y., Gill, T. M., Bandeen‐Roche, K., Becher, R. D., Davis‐Plourde, K. L., & Zang, E. X. (2025) · American Journal of Epidemiology · doi:10.1093/aje/kwaf209
Affective Polarization across Time and Place in the United StatesPreprint
Warner, S. (2025) · SSRN Electronic Journal · doi:10.2139/ssrn.5451614
Yu, Y., Chen, J., Knight, M. J., O’Brien, S. F., Buckeridge, D. L., Charlton, C. L., & Russell, W. A. (2025) · Canadian Journal of Public Health · doi:10.17269/s41997-025-01128-z
“By borrowing strength across similar groups, MRP generates more stable and precise estimates than rak-ing, especially for small or sparse subgroups, which was the case for the early pandemic period when few infections were detected (Park et al., 2017).”
Futility Analysis under ScrutinyPreprint
Jin, R., Wu, C., & Mesenbrink, P. (2025) · arXiv (Cornell University) · doi:10.48550/arxiv.2507.22770
“Bayesian multilevel regression [7], also known as hierarchical modeling, is a flexible statistical framework for analyzing data with nested or grouped structures.”
Estimating death rates in complex humanitarian emergencies using the network survival method
Breen, C. F. (2024) · Oxford University Research Archive (ORA) (University of Oxford) · doi:10.17605/osf.io/y6f4q
Additional file 1 of Quantifying participation biases on social media
Pokhriyal, N., Valentino, B. A., & Vosoughi, S. (2024) · Figshare · doi:10.6084/m9.figshare.26607154
Wright, O. (2026) · Warwick Research Archive Portal (University of Warwick) · doi:10.82444/warw.33901504
Essays on Sharing Economy and Platforms
Zeng, Y. (2025) · eScholarship (California Digital Library)
La saillance des promesses électorales : ses outils, ses mesures et sa validité
Tremblay‐Antoine, C. (2023) · Corpus Université Laval (Université Laval)
The Primary Connection: How Early Donors Influence Lawmaking in Congress
Chun, A. (2025) · eScholarship (California Digital Library)
Angry and Afraid: Race, Public Opinion, and the Politics of Punishment in the States
Duxbury, S. W. (2020) · OhioLink ETD Center (Ohio Library and Information Network)
Regressió multinivell amb postestratificació per fer estimacions territorials a partir d'enquestes
Ardiaca Arnau, J. (2024) · UPCommons institutional repository (Universitat Politècnica de Catalunya)
Economic analysis of beef cattle and groundwater
Ha, S. S. (2022) · K-State Research Exchange (Kansas State University)
Chen, J. (2023) · eScholarship (California Digital Library)
Osayande, N. (2024) · eScholarship@McGill (McGill) · doi:10.82308/7057
Poirier, W. (2023) · Corpus Université Laval (Université Laval)
Modeling issue competence over time: a Bayesian framework for estimating dynamic issue ownership
Nyhuis, D., Baumert, J.-F., Choi, J., Block, S., & Harmening, M. (2025) · Political Science Research and Methods · doi:10.1017/psrm.2025.10074
Lee, J.-H., & Hooper, A. L. (2026) · Mathematics · doi:10.3390/math14030512
Non-probability sampling: Theory, methods, and practical guidance
Borualogo, I. S. (2026) · Elsevier eBooks · doi:10.1016/b978-0-443-26629-4.00197-0
Lloyd-Smith, P. R. (2026) · Journal of Environmental Economics and Management · doi:10.1016/j.jeem.2026.103315
Estimating Residential Displacement in the Central Puget Sound Region using Household Survey DataPreprint
Dharamshi, A., Richards, M., Childress, S., Lee, B. W., & Casey, D. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2603.06437
“…of categorical variables, estimates the target indicator for each subpopulation using potentially non-representative data, and then aggregates these estimates back to the target population using external population weights derived from a high-quality source (Gelman, 1997; Park et al. , 2004).”
Baumert, J.-F., & Nyhuis, D. (2026) · Quality & Quantity · doi:10.1007/s11135-026-02704-2
Benchmarking Chlamydia Screening Rates: A Multilevel Regression and Post-stratification Approach
Arul, K. (2026) · Lecture notes in networks and systems · doi:10.1007/978-3-032-19179-3_14
Ferres, L., & Gauvin, L. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2604.16193
“MRP builds on earlier work on poststratification and hierarchical regression, with its modern formulation usually traced to Gelman and Little (1997) and its influential application to small-area public-opinion estimation developed by Park et al. (2004).”
Das, S., & Lahiri, P. (2026) · Research Square · doi:10.21203/rs.3.rs-9683522/v1
Farr, K. B. (2026) · Carolina Digital Repository (University of North Carolina at Chapel Hill) · doi:10.17615/nqh4-sn42
Assessing Intraurban Variation in Heat‐Attributable Health Burden to Inform Local Interventions
Farr, K. B., Mcinroe, E. M., Tsai, W.-L., Hsu, A., & Rappold, A. G. (2026) · GeoHealth · doi:10.1029/2025gh001791
Educational Gaps and Communication Priorities for the Healthfulness of Carbohydrate Foods
Erndt‐Marino, J. D., & Ghirardelli, A. (2026) · Journal of the American Nutrition Association · doi:10.1080/27697061.2026.2687436
Generalized reparametrized variational Bayes with skew-symmetric normalizationPreprint
Chen, A., & Tan, L. S. L. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2607.16698
“They are widely used in political science (Park et al. 2004), epidemiology (Blangiardo et al. 2013) and ecology (Royle & Dorazio 2008), but posterior inference becomes challenging when latent variables are high-dimensional and strongly correlated.”
Palm, E. C., Williamson, M. A., Snow, N. P., & Christensen, S. A. (2026) · Journal of Wildlife Management · doi:10.1002/jwmg.70254
da Silva, A. A., & Treleaven, P. C. (2026) · SSRN Electronic Journal · doi:10.2139/ssrn.7143999
Representation in Municipal Government: Evidence from LGBTQ+ Policy
Elwell, M., & Hamel, B. T. (2026) · Urban Affairs Review · doi:10.1177/10780874261471472
Mind the Gaps: Mixture-of-Minds for Human SimulationPreprint
Dahiya, P. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2608.06115
“They weight the respondents by post-stratification, and by its multilevel-regression extension, so that the reweighted sample matches known population margins [23].”
Yörük, E., Hürriyetoğlu, A., KINA, M. F., Duruşan, F., Yardı, M. C., Atsızelti, Ş., Etgü, T., Delen, E., Gürsoy, M. E., Yıldız, Ö., et al. (2026) · EPJ Data Science · doi:10.1140/epjds/s13688-026-00682-x
Niitsoo, M., Rebane, R., & Jüristo, T. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2608.14062
Koc, P., & Kołczyńska, M. (2026) · doi:10.33774/apsa-2026-dwm7b
Beltrán-Sánchez, M. Á., Martínez-Beneito, M. Á., Eikemo, T., Martino, S., & Riebler, A. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2609.29893
“…approaches are formulated at the individual level and are combined with poststrati-fication, following the multilevel regression and poststratification approach (Gelman and Little, 1997; Park et al., 2004, 2006), to obtain finite population estimates for geographical areas and population subgroups.”
Essays on the Material Origins of Political Change
Agnolin, P. (n.d.) · Archivio istituzionale della Ricerca - Bocconi
Ahlquist, J. S., & Thai, E. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2610.06774
Using AI Predictions to Augment Rather Than Replace SurveysPreprint
Wei, L., & Zhu, Y. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2611.06743
Population-Aware Bayesian Fairness in AI under Complex Survey DesignsPreprint
Chatterjee, D. (2026) · Research Square · doi:10.21203/rs.3.rs-11335320/v1
Bounding MrP: Subnational public opinion estimation without detailed census dataPreprint
Marbach, M. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2610.16726
“…has become the dominant method for estimating public opinion in small subnational units from survey data (e.g., US states, UK constituencies, Swiss cantons) where direct estimates are too noisy to be useful (Gelman and Little, 1997; Park, Gelman and Bafumi, 2004; Leemann and Wasserfallen, 2020).”
Jonaitis, D., Corcoran, B., POUDYAL, B., Roskam, A., Ali, A., Agrawal, S., Duch, R. M., & Cerina, R. (2026) · PS Political Science & Politics · doi:10.1017/s1049096526103084
Lee, S. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2610.20321
United States Jewish Population, 2024
Sheskin, I. M., Dashefsky, A., & Comenetz, J. (2026) · American Jewish year book · doi:10.1007/978-3-032-15750-8_6
Data Governance for Generative and Agentic AI : Technology, Law, Economics, and SecurityPreprint
Darapaneni, N., Padmaja, K., & Gnana Chaithra, D. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2610.33605
The uneven rise of AI-generated web contentPreprint
Visokay, A., Pan, S., Griston, M., Iasakova, P., Nepf, M., Peskoff, D., Sinha, R., Sun, J., Thiara, I., Wang, L., et al. (2026) · arXiv (Cornell University) · doi:10.48550/arxiv.2610.36631
Bayesian multinomial small area estimation with post-stratification and stochastic benchmarking
Nikolaidis, I. B., & Giakoumatos, S. (2026) · Statistical Journal of the IAOS · doi:10.1177/18747655261504627
Understanding AI Adoption in Local Government: Capacity Constraints or Political Caution?
Tsai, Y.-Y., Keiser, L. R., & Jabbari, N. (2026) · The American Review of Public Administration · doi:10.1177/02750740261506489
Brown, Z., Martens, N. J., & Bolter, K. (2026) · Regional Science Policy & Practice · doi:10.1016/j.rspp.2026.100361
Breen, C. F., Mahmud, A. S., & Feehan, D. M. (2022) · doi:10.1371/journal.pcbi.1010742.r002
On their own initiative: how politicians use direct democracy in the United States
Lubbock, T. (2012) · Oxford University Research Archive (ORA) (University of Oxford) · doi:10.5287/ora-8rbny0zxz
Measuring Public Opinion toward Social Welfare in Taiwan
Tsai, C. (2015) · doi:10.6683/tpsr.201506.19(1).241-279
“…(MRP) has consistently proved to be successful when using national survey data to estimate the influence of demographic and geographic predictors on the dependent variable (Kastellec, Lax and Phillips 2014; Lax and Phillips 2009; Park, Gelman, and Bafumi 2004; Warshaw and Rodden 2012).”
BackgroundA Third Mechanism of Policy Diffusion: The Social Contagion ModelPreprint
Pacheco, J. (2010) · SSRN Electronic Journal
Metadata
- Authors
- Park, D. K., Gelman, A., & Bafumi, J.
- Affiliations
- Washington University in St. Louis (US)
- Columbia University (US)
- Venue
- Political Analysis
- Publisher
- Cambridge University Press (CUP)
- Published
- 2004
- Volume
- 12
- Issue
- 4
- Pages
- 375-385
- ISSN
- 1047-1987 (ISSN-L)
- 1476-4989
- OpenAlex
- W2105602542
- Cited by
- 480
- References
- 15
- Citations by year
- 2026: 19 · 2025: 20 · 2024: 26 · 2023: 19 · 2022: 28 · 2021: 48 · 2020: 85 · 2019: 45 · 2018: 48 · 2017: 16 · 2016: 20 · 2015: 21 · 2014: 20 · 2013: 12 · 2012: 10
- FWCI
- 5.74
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Political Analysis (publishedVersion)
- RePEc: Research Papers in Economics (submittedVersion)
- Topics
- Physical Sciences › Mathematics › Statistics and Probability › Survey Sampling and Estimation Techniques
- Social Sciences › Political Science and International Relations › Electoral Systems and Political Participation
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Bayesian Inference
- Keywords
- Bayesian hierarchical model, post-stratification, small area estimation, state-level estimates, national polls, election forecasting, multilevel logistic regression
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.