Does asking about citizenship increase labor survey non-response?
询问公民身份是否会增加劳工调查的无回应？
Robert Bernhardt 1Phanindra V. Wunnava2
作者信息
- 1. University of Chicago
- 2. Middlebury College
- 折叠
摘要
在2020年美国人口普查中增加公民身份问题的尝试失败，引起了其他调查中公民身份问题的关注。同时，研究人员注意到，目前人口调查(CPS)无应答的长期增长。我们结合这些主题，研究1994年CPS重新设计中增加的CPS公民身份问题对拒绝的影响。直接面板回归显示，非公民身份率较高的州有较高的拒绝率。事件研究回归不连续性显示，由于重新设计，拒绝的增加20-40%。此外，差异研究设计显示，非公民和西班牙裔人口较多的州受到重新设计的影响更大。这些结果意味着，这个问题导致非公民和拉美裔美国人不成比例地拒绝参加调查。鉴于这个问题似乎威胁到调查的代表性，我们建议进行随机对照试验，以准确确定问题的效果。
Abstract
Abstract The unsuccessful attempt to add a citizenship question to the 2020 US Census has drawn attention to citizenship questions on other surveys. Simultaneously, researchers have noted a secular increase in Current Population Survey (CPS) non-response. We combine these topics, studying the effect of the CPS citizenship question, added in the 1994 CPS redesign, on refusals. Direct panel regressions show states with higher rates of non-citizenship have higher refusal rates. An event-study regression discontinuity shows a 20-40% increase in refusals attributable to the redesign. Moreover, a difference-in-differences research design shows states with larger non-citizen and Hispanic populations were more affected by the redesign. These results imply the question causes non-citizens and Hispanics to refuse to participate in the survey disproportionately. Given the question appears to threaten the representativeness of the survey, we recommend there be a randomized controlled trial to precisely determine the question’s effects.
Key words
Current population survey/Non-response/Survey refusal/Citizenship status/Immigration/Event study/Regression discontinuity/Panel data引用本文复制引用
出版年
2023
NETL
NSTL
Springer Nature