Interesting! The survey report PDF with data is here:
KFF/The Washington Post Trans Survey
METHODOLOGY
This KFF/Washington Post Trans Survey was designed and analyzed by public opinion researchers at the Kaiser Family Foundation (KFF) and The Washington Post. The survey was designed to reach a representative sample of adults in the U.S. who identify as transgender or as a trans adult along with a comparison representative of the general population of U.S. adults. The survey was conducted November 10 – December 1, 2022, online and by telephone among a nationally representative sample of 515 U.S. adults who identify as trans and another 823 cisgender U.S. adults who do not identify as trans and their gender is the same as their sex assigned at birth.1 All survey respondents received a financial incentive for participating in the survey.2
The comparison sample of the U.S. adult population (n=823) was conducted using the SSRS Opinion Panel either online (n=784) or panel members who do not use the internet were reached by phone (n=39). Twenty seven of these individuals completed the survey in Spanish.
The sample of trans adults was recruited using three probability-based online panels, the Gallup Panel (n=252), NORC’s AmeriSpeak panel (n=146), and the SSRS Opinion Panel (n=85). All three of these nationally representative panels are recruited using probability-based methodologies (SSRS and Gallup recruit using both RDD and ABS, while NORC relies on ABS). In order to interview under-surveyed populations, the project also includes telephone interviews from calling back respondents from previous KFF surveys (n=29) or from previous SSRS Omnibus surveys (n=3) who identified as lesbian, gay, bisexual, or trans. Overall, there were 515 interviews from trans adults, including 3 interviews conducted in Spanish.
All online completes were reviewed to ensure respondents were giving the survey adequate attention. Any respondent with over 30% item non-response or with a length less than one quarter of the mean length by mode were flagged and reviewed. Four cases were removed from the data that failed both of these quality checks. Additionally, the trans sample was subject to additional data checks including reviewing responses to open -ended questions and gender identity questions. Fourteen cases were removed due to either the respondent not actually identifying as a trans adult or that the respondent was not responding truthfully.
Trans adults from the combined phone and panel samples were weighted separately to match the sample’s demographics to the national U.S. adult trans population using data from the CDC’s Behavioral Risk Factor Surveillance System (BRFSS) as well as the Williams Institute’s analysis of BRFSS, which used small area estimation to model demographics for states that had not recently asked the sexual orientation and gender identity module. The weighting parameters included age, education, race/ethnicity, and region. The weights take into account differences in the probability of selection for each sample type (callback phone sample and panel). This includes adjustment for the sample design, within household probability of selection, and the design of the panel-recruitment procedure.
The full sample of U.S. adults was weighted to match the sample’s demographics to the national U.S. adult population using data from the Census Bureau’s 2021 Current Population Survey (CPS). Weighting parameters included sex, age, education, race/ethnicity, region, and education. The sample was also weighted to match patterns of civic engagement from the September 2019 Volunteering and Civic Life Supplement data from the CPS. The sample was also weighted to match frequency of internet use from the National Public Opinion Reference Survey (NPORS) for Pew Research Center. The weights take into account differences in the probability of selection for each sample type (callback phone sample and panel). This includes adjustment for the sample design, within household probability of selection, and the design of the panel -recruitment procedure.
The margin of sampling error including the design effect for the trans adult sample is plus or minus 7 percentage points and plus or minus 4 percentage points for the cisgender adult sample. Numbers of respondents and margins of sampling error for key subgroups are shown in the table below. For results based on other subgroups, the margin of sampling error may be higher. Sample sizes and margins of sampling error for other subgroups are available by request. Sampling error is only one of many potential sources of error and there may be other unmeasured error in this or any other public opinion poll. KFF and The Washington Post are a charter members of the Transparency Initiative of the American Association for Public Opinion Research.
Full report:
https://www.washingtonpost.com/documents/dfa015fb-e64f-4eb2-9cfd-048d9e9dc108.pdf?itid=lk_inline_manual_12