particularly gay and bisexual men, as well as transgender women. The epidemic serves as an example of some of this report’s key themes: not only resilience, but also stigma, racial and ethnic disparities, and the importance of research funding.
A number of challenges are associated with conducting health research on LGBT populations. These include the following:
Sexual orientation and gender nonconformity are multifaceted concepts, and defining them operationally can be challenging.
Individuals may be reluctant to answer research questions about their same-sex sexual behavior or gender nonconformity.
Because LGBT populations represent a relatively small proportion of the U.S. population, it is labor-intensive and costly to recruit a large enough sample in general population surveys for meaningful analysis of these populations and their subgroups.
Despite these challenges, many researchers currently conduct health research on LGBT populations. In so doing, as with research on any populations, researchers must choose a sampling strategy and a data collection method. While probability samples allow findings to be generalized to the study’s target population, they are expensive and difficult to implement with LGBT populations given the research challenges listed above. However, the use of established statistical techniques makes it possible to improve the precision of estimates for small populations by combining two or more data sets.
Research on the health status of LGBT populations more commonly uses nonprobability samples. Even though the extent to which findings based on such samples accurately characterize these populations is unknown, these samples have yielded valuable information for expanding the field of LGBT research and identifying possible gaps in health services. In addition to providing general descriptive data for LGBT populations and their subgroups, nonprobability samples have served to reveal the existence of certain phenomena, to suggest relationships among variables, to identify possible differences among groups, and to generate hypotheses and formulate ideas that can be advanced for systematic study in the future. A variety of methods are used to generate nonprobability samples, including purposive, quota, and snowball sampling. While much of what is currently known about the health of LGBT populations comes from studies with nonprobability samples, the field of LGBT health would benefit if more data came from probability samples.