A study published in the July issue of PS: Political Science & Politics has found that how surveys ask about gender identity can significantly affect both the accuracy of data and how transgender and nonbinary people are represented. The findings underscore the importance of thoughtful survey design in a political climate where data is often used to shape debates over the rights of LGBTQ+ individuals.
The research team analyzed responses from the Canadian Election Study (CES) in both 2019 and 2021. In 2019, the CES asked participants to identify as “Man,” “Woman,” or “Other (e.g., trans, nonbinary, Two-Spirit).” That structure forced many transgender men and women into the “Other” category, even though they identified as male or female. The results blurred distinctions between trans and nonbinary populations, making it difficult to understand the true diversity within gender minorities.
By 2021, the CES updated its approach to a “two-step” model. The survey first asked participants to identify their gender, with options for man, woman, nonbinary, or another gender. It then followed with a second question on whether the respondent identified as transgender. This change offered a clearer and more inclusive picture of gender identity, allowing transgender men and women to identify their gender accurately while still acknowledging their trans identity.
The difference was significant. In 2019, about 0.8 percent of respondents chose “Other,” but researchers concluded that a large share were actually transgender men or women who had been mislabeled. By contrast, the 2021 model provided more precise results, with nonbinary respondents making up around 0.5 percent of the sample and transgender identity more accurately distributed across categories. The findings also revealed important demographic and political distinctions. Nonbinary respondents were generally younger, more likely to identify as queer, and more supportive of progressive parties compared to the broader transgender population.
The study’s authors caution against oversimplified survey options, noting that poor design can misrepresent marginalized groups and lead to flawed conclusions in political science, public policy, and social research. They recommend that surveys avoid collapsing “transgender” into a third category alongside “man” and “woman” and instead adopt the two-step model. This approach has been endorsed by LGBTQ+ advocacy groups for years and is considered best practice for capturing gender diversity.
Accurate data is not just a matter of statistics. It has real-world implications. Policymakers, educators, and health professionals often rely on survey results to shape decisions about funding, programming, and protections. When transgender and nonbinary identities are inaccurately measured, it can reinforce misconceptions, minimize visibility, and even fuel harmful narratives.
As debates over transgender rights continue to dominate headlines, this research serves as a reminder that representation begins at the most basic level, in how people are counted. By improving survey design, researchers can ensure that transgender and nonbinary individuals are not pushed into “Other” boxes but instead recognized for who they truly are.
For communities fighting every day to be seen and respected, that recognition matters.
