Can AI Stand In for Human Survey-Takers? Not Really
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.
Numbers, Facts and Trends Shaping Your World
Read our research on:
Methodological Research
Sign up for our Methods newsletter
The latest on survey methods, data science and more, delivered quarterly.
All
Methodological Research
Publications
-
How many same-sex married couples in the U.S.? Maybe 170,000
A new research paper suggests that the number of married same-sex couples in the United States in 2013 may have been much lower than the Census Bureau’s initial estimate for that year.
-
Tips for Creating Web Surveys for Completion on a Mobile Device
With so many respondents taking Web surveys on smartphones, creating surveys with smartphone respondents in mind is critical.
-
How Pew Research Center conducted its survey of multiracial Americans
We released our first report on American multiracial adults, a group that comprises an estimated 6.9% of the adult population, or nearly 17 million adults. The report looks at who they are demographically, their attitudes and experiences, and the spectrum of their racial identity.
-
Ukraine 2015: How We Approached Our Sample Design in Light of Insecurity in Eastern Ukraine
The uneasy ceasefire in eastern Ukraine, punctuated by almost daily fighting between separatists and government forces, posed a major challenge to the Pew Research Center as we set about conducting a new public opinion survey in that country this past spring. As always, our first priority was the safety of interviewers and respondents, who can both be at risk when it comes to face-to-face surveys in a conflict zone.
-
Q&A: A look at what’s driving the changes seen in our Religious Landscape Study
Fact Tank sat down with David Campbell, a professor of political science at the University of Notre Dame, to explore what the new findings mean.
-
Methods can matter: Where Web surveys produce different results than phone interviews
A Pew Research Center experiment found several key areas where Web surveys produced different results than those conducted by phone.
-
Q&A: How we projected the future of world religions
Demographer Conrad Hackett explains how he and his team put together our major new report and why it differs from past efforts to predict religious change.
-
Building Pew Research Center’s American Trends Panel
In early 2014, Pew Research Center set out to build a probability-based panel – the American Trends Panel – to supplement our traditional method of data collection in the U.S. – the random digit dial (RDD) telephone survey. Here’s how we built and managed the panel, and what we learned from it in 2014.
-
What we learned about surveying with mobile apps
No research has compared app-based surveys with polls administered via Web browsers. Our new, experimental work compares the results of these two modes.
REfine Your Selection
Date
Years
Formats
Regions & Countries
Research Teams
Bylines
Signature Reports
A Field Guide to Polling: Election 2020 Edition
While survey research in the United States is a year-round undertaking, the public’s focus on polling is never more intense than during the run-up to a presidential election.
Comparing Two Types of Online Survey Samples
Pew Research Center conducted a study to compare the accuracy of six online surveys of U.S. adults – three from probability-based panels and three from opt-in sources. On average, the absolute error on opt-in samples was about twice that of probability-based panels.
Q&A: What is the American Trends Panel, and how has it changed?
Members of the American Trends Panel can now take our surveys online or over the phone with an interviewer.
How Public Polling Has Changed in the 21st Century
A new study found that 61% of national pollsters used different methods in 2022 than in 2016. And last year, 17% of pollsters used multiple methods to sample or interview people – up from 2% in 2016.
