Cited by 2Electoral Systems and Political ParticipationSurvey Methodology and NonresponseSurvey Sampling and Estimation Techniques

The impact of survey mode on the measurement of polarization in the public: 2016–2020 U.S. panel evidence

Open access
Thornton, Judd R.·2025·American Politics Research

Abstract

Over the last 40 years, affective polarization has noticeably increased, with recent elections showing especially large gains. Moreover, these increases have surpassed other indicators of polarization, such as issue attitudes and partisan-ideological sorting. Could the recent surge in affective polarization be influenced by the growing use of internet-based, self-administered surveys? I argue that measures of partisan affect are more vulnerable to mode effects than other indicators of polarization. In this paper, I employ American National Election Studies panel data from 2016 to 2020. Crucially, in 2016 the survey was conducted with a dual mode in which a portion of respondents were interviewed in the traditional face-to-face format while others completed a self-administered online survey. Both sets of respondents were re-interviewed in 2020 where all respondents completed the survey online. I find that differences in party affective polarization and candidate trait ratings between the two modes, substantial in 2016, largely disappear when social desirability effects are removed. In contrast, issue extremity and partisan-ideological sorting show no noticeable mode effects.

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

Integrity signals

Registry

Crossref · 10.1177/1532673x251347623

Cites

73 of 74 references verified0 citation errors · 0 not checkable

Statistics

Claims the paper makes

The claims and statistics of this paper haven’t been checked yet.

Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.

A few free checks a day · papers already verified open instantly · results are public

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?

  • There are important mode effects in ANES affective polarization measurement: online respondents feel warmer toward their own party and colder toward the opposing party than face-to-face respondents.

    “Figure 2A clearly indicates that there are important mode effects.Those in the online sample feel warmer toward their own party and colder toward the opposing party (also see Thornton, 2025).”

    From Measuring affective polarization: Matching concepts to measures (2026) · cites this paper as Thornton, 2025

Self-check

Percentages, sums and reported differences re-computed from the paper’s own numbers.

The self-check needs the paper’s full text.

Cites

Flagged first, then by citations

Research works printed but not linked: 1

  1. Campbell, A., Converse, P. E., Miller, W. E., & Stokes, D. E. (1960). The American voter: Wiley.

Cited by

Sorted by citations

Authors
Thornton, J. R.
Affiliations
Georgia State University (US)
Corresponding author
Thornton, Judd R.
Venue
American Politics Research
Published
2025-05-30
Volume
53
Issue
5
Pages
531-543
ISSN
  • 0044-7803 (ISSN-L)
  • 1532-673X
  • 1552-3373
OpenAlex
W4410888668
Cited by
2
References
73
Citations by year
2026: 2
FWCI
7.44
Type
journal article
Language
en
Open access
Locations
American Politics Research (publishedVersion)
Topics
  • Social Sciences › Political Science and International Relations › Electoral Systems and Political Participation
  • Social Sciences › Sociology and Political Science › Survey Methodology and Nonresponse
  • Physical Sciences › Mathematics › Statistics and Probability › Survey Sampling and Estimation Techniques
Keywords
affective polarization, survey mode effects, face-to-face interviews, social desirability bias, American National Election Studies

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