Skip to main content
Log in

Does asking about citizenship increase labor survey non-response?

  • Original Paper
  • Published:
Journal of Population Economics Aims and scope Submit manuscript

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.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2

Similar content being viewed by others

Notes

  1. The most recent official documentation is found in Technical Paper 77, while the most time-period accurate documentation is contained in Technical Paper 63RV.

  2. The official documentation uses both the terms non-interview and noninterview (e.g. Type-A non-interview), and makes a subtle distinction between a non-interview and non-response. For simplicity, we use the latter term to refer to both concepts.

  3. This capitalization is in the questionnaire listing and in an image of the interviewing software in the interviews’ manual.

  4. See Jasso and Rosenzweig (2020, pp. 646-647) for a clear explanation of why there could be the potential that the current citizenship question may have an adverse effect on response from derivative citizens. To make this point clear, we included the citizenship question/instructions to fill in Section 2.2.3 for January 1994 CPS.

  5. Certain variables are allowed to remain unweighted. This is either because (a) they correspond to an variable directly pertaining to non-response (i.e. refusals) that the weights are meant to correct for but not adjust, or (b) there is a discontinuous shift in 1990 and/or 2000 unrelated to the CPS itself resulting from the weights being updated after new census results were applied.

  6. Alternative versions based on the other dataset (either national or state-level) may be made available upon request.

  7. Consider a scenario with 1000 surveyed individuals, 5% of whom are non-citizens. If 96% of citizens and 92% of non-citizens respond, we would observe 912 citizens and 46 non-citizens, yielding an observed non-citizen percentage of 4.8%.

  8. Results of these simulations are available upon request.

  9. When possible, we use data prior to the redesign to minimize the chance of non-response influencing treatment group assignment. This is not possible for non-citizen population, which was unobservable before the question was added.

  10. The other three are Born in the United States population (1994), born in Mexico population (1994), and citizenship question revisions (1994).

  11. For each of the 5 categories, the same amount of states (20, 15, 10) are used to assigned each treatments/control status. However, it need not be the case that 10 states are in the top (bottom) 10 across all characteristics.

  12. With 5 characteristics, it is impossible to be in 3 or more treatment groups and also be in 3+ control groups.

  13. Results for Type-A non-response are included in the Appendix.

  14. Controls include average household size & income, educational attainment, labor force status, and race/ethnicity variables.

  15. The direct panel regression controls include slightly different education variables than previous regressions.

References

Download references

Acknowledgements

This paper has been shaped through the contributions of numerous individuals. We are thankful for the five anonymous referees, editor-in-chief (Klaus Zimmermann), and managing editor (Madeline Zavody) of the Journal of Population Economics for their constructive comments and recommendations. We thank Ryan Moise, for proposing important clarifications, Liam Puknys for reviewing our simulation methods, Yashar Barardehi for key methodological recommendations, David Munro & Erin Wolcott for introducing one of the authors to the CPS microdata, and Dan Bernhardt for review of previous drafts. An earlier version of this paper was presented at the June 2020 virtual sessions of the Western Economic Association Annual Conference, and benefited immensely by the discussant comments of Nicole Simpson of Colgate University. A previous version of this paper was circulated as IZA Discussion Paper #13350 (June 2020). The usual caveats apply.

Author information

Authors and Affiliations

Authors

Corresponding author

Correspondence to Phanindra V. Wunnava.

Ethics declarations

Ethical approval

No human participants were involved in this research.

Conflict of interest

The authors have no relevant financial or non-financial interests to disclose. The authors have no conflicts of interest to declare that are relevant to the content of this article. All authors certify that they have no affiliations with or involvement in any organization or entity with any financial interest or non-financial interest in the subject matter or materials discussed in this manuscript. The authors have no financial or proprietary interests in any material discussed in this article.

Additional information

Responsible editor: Klaus F. Zimmermann.

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Supplementary Information

Below is the link to the electronic supplementary material.

Supplementary file 1 (pdf 253 KB)

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Bernhardt, R., Wunnava, P.V. Does asking about citizenship increase labor survey non-response?. J Popul Econ 36, 2457–2481 (2023). https://doi.org/10.1007/s00148-023-00945-1

Download citation

  • Received:

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s00148-023-00945-1

Keywords

JEL Classification

Navigation