Department of Labor Logo United States Department of Labor
Dot gov

The .gov means it's official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you're on a federal government site.


The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Assessing the Feasibility of Asking About Sexual Orientation and Gender Identity in the Current Population Survey: Executive Summary

PDF version


In 2016, the Department of Labor sponsored research to explore “the feasibility of adding sexual orientation and gender identity (SOGI) questions to the Current Population Survey (CPS).” The emphasis of the research was on asking SOGI questions in the context of an employment survey and via proxy reporting, in which one person generally responds for all eligible members of the household. Researchers at the Bureau of Labor Statistics (BLS) and U.S. Census Bureau conducted a total of 132 cognitive interviews and four focus groups to evaluate respondents’ ability and willingness to answer SOGI questions about themselves and others in their household in the context of a Federal government survey on employment.1 The cognitive interviews were conducted both with respondents who are lesbian, gay, bisexual, and transgender (LGBT) and with those who were not, while the focus groups were conducted with transgender respondents. Respondents in both the cognitive interviews and focus groups were recruited to include a range of demographic characteristics, and data were collected in four locations in order to explore geographic differences. The overarching goal of the project was to determine whether asking SOGI questions in the CPS is feasible — in other words, whether resources should be committed to pursuing additional research.

Extensive analysis of the data collected was performed, resulting in two independent reports documenting the focus group and cognitive interview. This document serves as a high level summary of the findings. Readers are encouraged to read the full reports because, as with all exploratory research, the results can be complex.

1 Specific research questions were identified for both the cognitive interviews and focus groups. These research questions, and the results for each, are detailed in the full reports.


Question Difficulty

Results indicate that, in general, respondents understood the sexual orientation and gender identity questions as intended and were able to provide an answer without much difficulty. Comparisons with other CPS questions suggest that there were not more issues observed for the SOGI questions than the others; in fact, more difficulty was observed with the income, employment and disability questions than the SOGI questions. LGBT respondents had more difficulty with the SOGI questions than non-LGBT respondents, often being unable to align their self-identity with the response options provided; this was especially true for transgender respondents.

Question Sensitivity

As with difficulty, results indicated that while there was some sensitivity associated with the SOGI questions, a majority of respondents did not indicate sensitivity concerns. Sexual orientation, along with the disability questions, was found to be the most sensitive item, both for self and proxy response. Both LGBT and non-LGBT respondents found the sexual orientation question sensitive, with the former finding their own identity personal to talk about, and the latter finding the sexual orientation question culturally sensitive and more private. Fewer respondents indicated that they found the gender identity question sensitive, for either self or proxy reporting. Those that did indicate gender identity sensitivity had similar reasons as sexual orientation, with the topic being private generally.

Proxy Reporting

Cognitive interviews suggest that proxy response may be effective for sexual orientation and gender identity questions, with respondents generally being willing and able to answer the questions about other household members. The match rates between paired interviews (interviews with two members of a household interviewed separately) were higher for SOGI questions than for income, educational attainment, and some employment questions. Results from the transgender focus groups suggest the opposite conclusion; transgender respondents were opposed to having someone report gender identity on their behalf, and did not want to do so for others in their household. Additionally, focus group respondents suggested that not everyone in their household would necessarily know or be willing to report their true gender identity status.

Demographic Differences

While there were not many consistent demographic trends, we consistently found that most of the respondents who had difficulty and sensitivity with the SOGI questions were lesbian, gay, or bisexual. Furthermore, the few transgender respondents in the study almost universally expressed some difficulty and sensitivity with the SOGI questions. There were some differences by household size, age, urbanicity, educational attainment, and race, but these were inconsistent across all the comparisons (e.g., difficulty for proxy reporting, sensitivity for self reporting), and so we cannot determine if they were meaningful or not.

Federal Employment Survey Context

In this study with paid volunteers, all but two respondents answered the SOGI questions both for themselves and for all other household members (over the age of 15). During the debriefings, respondents indicated that they generally understood the objective of these questions being included on a Federal survey about employment, with only a few saying they did not see the relevance. Many of the transgender focus group respondents and a few cognitive interview respondents had some concerns about how the government would use the SOGI information and who would have access to it.

Study Limitations

This qualitative study was carefully designed to evaluate the feasibility of asking SOGI questions by proxy in the employment survey context of the CPS. The qualitative data collected is not meant to produce estimates or represent the population as a whole. While the qualitative sample was large, some individual cells (e.g., low levels of educational attainment, Hispanics, younger age groups) were small, the study was only conducted in select locations, and cognitive interviews excluded single-person households. Additionally, all participants were paid volunteers willing to respond to a recruitment ad. Finally, the cognitive interviews only evaluated one set of SOGI questions in one place in the CPS instrument. Question wording and placement are known to have strong impacts on how respondents react to questions, so additional testing is needed to determine the ideal wording and placement of the SOGI questions. Caution should be taken when trying to apply the findings to make broad conclusions.


Adding questions to a prominent Federal survey such as the CPS requires careful consideration and significant research. This study did not identify any significant issues that would make collecting SOGI information in the CPS infeasible, though there are many outstanding issues identified in the full study reports that must be studied and addressed prior to any implementation efforts.


Last Modified Date: October 18, 2018