How public opinion is measured, by us and by the field we work in.
16 items · newest 5 Oct 2026
Pew asked a language model to stand in for American survey respondents and missed a real result by 21 points. The structural reason that happened is far worse here than it is there.
A UN official warns that extreme heat from climate change is a fatal threat to pregnant women and newborns, citing a survey of healthcare professionals. Read More: https://punchng.com/extreme-heat-threatens-pregnancy-birth-safety-un-official/
Survey experiments with Opus 4.6 and GPT-5.1 each paint a very different picture of U.S. public opinion, neither of which reflect reality.
AI survey respondents are more likely than humans to agree on one answer choice or avoid a response option entirely.
Synthetic survey respondents tend to exaggerate differences between demographic subgroups and portray some responses as nearly universal.
Synthetic surveys using digital twins do a poor job estimating the public mood and are especially error-prone for some subgroups, including Republicans and Black adults.
Public Opinion Quarterly, AAPOR’s flagship publication, has published a new special issue, “Advancing Public Opinion Research Across Countries and Contexts,” examining how public opinion is shaped by the political, cultural, historical, and social environments in which it develops. Edited by Hsuan-Ting Chen, Hernando Rojas, and Sebastián Valenzuela, the special issue addresses a central challenge in […] The post POQ Releases Special Issue on Advancing Public Opinion Research Across Countries and Contexts first appeared on AAPOR .
There are good arguments against weighting surveys to match how people voted in past elections or an independent estimate of political party affiliation. We know because we’ve made many of those arguments over the years! However, the steep decline in survey response rates and the rise in differential partisan nonresponse have made those arguments less […]
Pollsters vary in how they weight their survey samples on political characteristics, if at all. Their choices tend to be informed by factors such as the population being surveyed and the sample source (e.g., a list of all addresses in the United States or a list of registered voters), the focus of the poll (preelection […]
One challenge pollsters face when weighting surveys based on people’s self-reported political party affiliation is that the population target can change fairly quickly. One year, Americans may be evenly divided between Republicans and Democrats. The next year, politics may shift and one party clearly outnumbers the other in supporters. Weighting to the shares of Republicans […]
This report explains how political weighting works, how Pew Research Center does it, and some of its limitations.
