Validating the “Genuine Pipeline” to Limit Social Desirability Bias in Survey Estimates of Voter Turnout
- 17 April 2024
- journal article
- research article
- Published by Oxford University Press (OUP) in Public Opinion Quarterly
- Vol. 88 (2) , 268-290
Abstract
It is well documented that survey overreporting of voter turnout due to social desirability bias threatens inference about political behavior. This paper reports four studies that contained question wording experiments to test questions designed to minimize that bias using a “pipeline” approach. The “pipeline” informs survey participants that researchers can perform vote validation to verify turnout self-reports. This approach reduced self-reported turnout by 5.7 points in the 2020 American National Election Study, which represents a majority of the estimated overreporting bias. It reduced reported turnout by 4 points in two nonprobability samples. No effect was found in a third nonprobability study with Amazon Mechanical Turk workers. Validated vote data also confirm that the pipeline approach reduced overreporting. We tested heterogeneous effects for sophistication and several other variables, but results were inconclusive. The pipeline approach reduces overreporting of voter turnout and produces more accurate estimates of voters’ characteristics.Keywords
Funding Information
- National Science Foundation (SES-1835022, SES-1835721)
This publication has 29 references indexed in Scilit:
- Logistic or linear? Estimating causal effects of experimental treatments on binary outcomes using regression analysis.Journal of Experimental Psychology: General, 2021
- Using a Probabilistic Model to Assist Merging of Large-Scale Administrative RecordsAmerican Political Science Review, 2019
- Why Does the American National Election Study Overestimate Voter Turnout?Political Analysis, 2018
- Developing Standards for Post-Hoc Weighting in Population-Based Survey ExperimentsJournal of Experimental Political Science, 2017
- Voter overrepresentation, vote misreporting, and turnout bias in postelection surveysElectoral Studies, 2013
- Linear versus logistic regression when the dependent variable is a dichotomyQuality & Quantity, 2007
- Sensitive questions in surveys.Psychological Bulletin, 2007
- An experimental comparison of question forms used to reduce vote overreportingElectoral Studies, 2006
- The Presence of Others and Overreporting of Voting in American National ElectionsPublic Opinion Quarterly, 1986
- The bogus pipeline: A new paradigm for measuring affect and attitude.Psychological Bulletin, 1971