Can They and Will They? Exploring Proxy Response of Sexual Orientation and Gender Identity in the Current Population Survey

• Jessica Holzberg ^[1] ; Renee Ellis ^[1] ; Robin Kaplan ^[2] ; Matt Virgile ^[1] ; Jennifer Edgar ^[2]
  1. [1] US. Census Bureau
  2. [2] US. Bureau of Labor Statistics
• Localización: Journal of official statistics, ISSN 0282-423X, Vol. 35, Nº. 4, 2019, págs. 885-911
• Idioma: inglés
• DOI: 10.2478/jos-2019-0037
• Enlaces
  • Texto completo
• Resumen

  • Within the United States Federal Statistical System, there has been interest in capturing sexual orientation (SO) and gender identity (GI), collectively known as SOGI, on surveys to allow researchers to estimate the size and distribution of sexual and gender minority populations. SOGI measurement in federal surveys may also help to identify disparities between people who identify as lesbian, gay, bisexual, and transgender (LGBT) and those who do not in domains such as health, crime, or employment. Although research has been conducted on best practices for SOGI measurement in surveys, it has largely been limited to examination of self-reports. Many federal surveys use proxy reports, when one person generally responds for all household members. This research used cognitive interviews and focus groups to explore proxy response to SOGI questions. We explored potential sources of measurement error in proxy responses to SOGI questions, including sensitivity, difficulty, as well as the willingness and ability of respondents to answer SOGI questions about other household members. We also conducted paired interviews with members of the same household to assess level of agreement for SOGI questions. Findings suggest that measuring SOGI by proxy may be feasible in federal large-scale, general population surveys.