Asking about citizenship in the US Census would be a costly exercise resulting in inaccuracies, finds new study
As protests continue to simmer across the nation, one target of the backlash has been the government’s National Population Register (NPR). The NPR asks respondents to declare their nationality and birthplace of their parents and is seen by many as a precursor to the National Register of Citizens . The government has set aside ₹13,000 crores for the next census and the NPR but research from the US suggests that the inclusion of citizenship questions could both increase costs and dilute the quality of the survey.
In the study, researchers J. David Brown and others compare the 2010 American Community Survey, that included a question on the respondent’s citizenship, with the 2010 census that did not include any such question. Specifically, they sought to estimate how willing respondents were to accurately disclose their citizenship. To test the veracity of answers, researchers match the community survey data with data from administrative records such as social security numbers or taxpayer data.
They find an 8 percentage point drop in response rates of non-citizen households compared with citizen households. Due to confidentiality concerns and the potential legal consequences of this question, non-citizen households are more sensitive and inconsistent in answering. As a result, they are more frequently found to misreport, abstain, skip, or incorrectly answer the question about citizenship compared to citizen households.
These inconsistencies drive up non-response follow-ups, increasing census costs. They also depress participation rate among non-citizens in the survey. This results in proxy estimates which lowers the quality of population count. As India gears up for its decennial survey and updating its NPR which includes similar questions on citizenship, there may be lessons to be learnt from the US experience.