Repository logo

Queering Survey Research: Measuring LGBTQ+ Political Identity in the U.K.



Change log


Roundy, Nathan 


Despite the multitude of ways LGBTQ+ people and issues appear throughout contemporary politics, there is little research within political science that engages with LGBTQ+ topics. An even smaller amount of research engages with LGBTQ+ people as research subjects. This dissertation sheds light on this blind spot by asking what is the political identity of LGBTQ+ people in Britain? I use large-n data to answer this question through quantitative modeling. Despite the advantages of large-n quantitative data, is unclear whether current methodologies are capable of providing an answer to this question. Current measures of sexual orientation and gender identity (SOGI) may not be accurately counting and coding LGBTQ+ people. To assess the extent to which SOGI measures can or cannot measure LGBTQ+ people, the secondary question of this dissertation is how should sexual orientation and gender identity be measured on surveys? To answer the latter question, I conduct focus group interviews, using queer method as a frame, with LGBTQ+ people to evaluate current SOGI measures. I synthesize these interviews to construct new SOGI measures. I test the efficacy of the new SOGI measures against current measures in a series of survey experiments fielded with YouGov in collaboration with the Cambridge-YouGov Centre for Public Opinion Research. To answer the former question, I deploy effective SOGI measures on the 2022 Globalism Project’s annual survey, again in collaboration with the Cambridge-YouGov Centre for Public Opinion Research, to identify political hetero- and homogeneity among LGBTQ+ people and between LGBTQ+ people and the general population. This dissertation reveals several key findings. First, LGBTQ+ people are significantly excluded from political science research. Second, LGBTQ+ people want to be included in research, but are unable to do so due to the absence of or inadequate use of SOGI measures. Third, response bias from LGBTQ+ people in research can be mitigated by clearly communicating data protection measures and demonstrating the researcher’s awareness of LGBTQ+ identity. Fourth, constructing SOGI measures with the input of LGBTQ+ people reduce LGBTQ+ response bias. Fifth, LGBTQ+ people, as an in-group, are politically heterogenous. Sixth, LGBTQ+ people demonstrate a sexuality and gender identity gap in comparison to the general population. Seventh, sexuality and gender identity are important variables to include in quantitative models of behavior and opinion. And eighth, 10-12% of the British population will identify with an LGBTQ+ identity.





Foa, Roberto


LGBTQ+ politics, political behavior, Political identity, public opinion, research methods, Survey design


Doctor of Philosophy (PhD)

Awarding Institution

University of Cambridge
PhD funded by the King's College and Cambridge Trust International Scholarship