Using screeners to measure respondent attention on self-administered surveys: Which items and how many?
Name
screener scale_190215.pdf
Description
Submitted version
Size
418.26 KB
Format
Adobe PDF
Checksum (MD5)
2b6dd6edfb546a302824b75a025222f3
Author(s)
Berinsky, Adam
Margolis, Michele F.
Sances, Michael W.
Warshaw, Christopher
Date Issued
November 2019
Journal
Political Science Research and Methods
Publisher
Cambridge University Press (CUP)
Citation
Berinsky, Adam et al. "Using screeners to measure respondent attention on self-administered surveys: Which items and how many?" Political Science Research and Methods (November 2019): dx.doi.org/10.1017/psrm.2019.53 © 2019 European Political Science Association
Version
Original manuscript
Abstract
Inattentive respondents introduce noise into data sets, weakening correlations between items and increasing the likelihood of null findings. “Screeners” have been proposed as a way to identify inattentive respondents, but questions remain regarding their implementation. First, what is the optimal number of Screeners for identifying inattentive respondents? Second, what types of Screener questions best capture inattention? In this paper, we address both of these questions. Using item-response theory to aggregate individual Screeners we find that four Screeners are sufficient to identify inattentive respondents. Moreover, two grid and two multiple choice questions work well. Our findings have relevance for applied survey research in political science and other disciplines. Most importantly, our recommendations enable the standardization of Screeners on future surveys.
MIT Department
Massachusetts Institute of Technology. Department of Political Science
Terms of Use
Creative Commons Attribution-Noncommercial-Share Alike
Persistent DSpace Link
DOI of Published Version
https://doi.org/10.1017/psrm.2019.53
