Even as many institutions in U.S. higher education work to create inclusive environments for trans and/or nonbinary people—especially students—current practices for collecting and using gender data impede these efforts. Recent reports by the National Academies of Sciences, Engineering and Medicine and the Center for American Progress on sexual orientation and gender identity data call for change in clinical and research data collection. In higher education, things are beginning to change, but some parts of the system still limit options to male and female, while others fail to consider the function of gender data, hindering institutions’ abilities to craft trans-inclusive policies—or even track the extent to which trans and/or nonbinary people are failed by current practices.
If colleges, universities and other higher ed institutions want their policies and practices to align with their rhetoric, we need to align our collection of gender data with a principle that starts with the why. What are these data for? And whose needs are we prioritizing when we collect, store and use them? Better gender data will not solve the escalating climate of antitrans panic, but it can make a difference to individual experiences and broader policy.
The three of us have complementary perspectives on this issue—one of us works on a college campus, one at an organization supporting the humanities and one at a professional association—but we all work with demographic data, and we have taken time to critically examine how gender is treated. Here are a few questions we consider before asking anyone to tick a box on a form. These are questions that many have been asking about other demographic data, such as data on race and ethnicity.
- Do you need to know?
- Why do you need to know?
- Who needs to see it?
- What is a respectful way to ask?
- What are you doing with the data?
Do you need to know? For many respondents, choosing a gender category is relatively straightforward, so those creating a data collection form might include it as standard procedure. Yet basic research ethics tells us not to ask for information we have no need or clearly envisioned use for. This is especially important when collecting gender data from trans and nonbinary people. Histories of violence and transphobia mean that trans students and staff may justifiably be wary of any gender data collection, irrespective of the goals and best intentions of the person asking, and especially if it prioritizes their assigned sex over their identity or requests that they reveal information without explaining why and how it is going to be used. Given the threats from legislators that trans people across the U.S. are currently facing, the best defense against gender data being used to persecute them may be simply not to collect or store those data at all.
Why do you need to know? In practice, “gender” (or “sex”) is sometimes used as a proxy for more specific questions, or it is collected simply because it always has been. Adding a nonbinary option, then, might not always be the most effective change. Ask yourself: Do I need to know a student’s lived gender so they can join the appropriate intramural sports team? Do I need to know their assigned sex so that health insurance will cover the correct cancer screenings? Do I need to know students’ gender identities as well as their trans status in the aggregate, so I can better capture the incidence of hostile environments for a campus climate survey? Am I trying to track how gender affects allocation of grant funding within a particular academic discipline? Or do I simply need to know how to address people socially—did I ask for gender when I meant pronouns?
Once the actual need is clear, we should then ask that question and, as much as possible, communicate with those from whom gender data is being requested, why we’re asking that question and what we’re going to do with their response. We should remember that data collection is also messaging; it communicates to those filling out a form what the institution or researcher thinks is important. By asking for personal information in a respectful way, we help to establish or re-establish trust, particularly with vulnerable populations.
Who needs to see it? At the height of COVID-19 testing regimes on campus, for example, one of us found themself faced with an unanticipated piece of gender data: an M or F, corresponding with individuals’ official records, was printed on sample vials. A cohort of staff members knew that making a small change—removing the visibility of the gender markers—would make the testing environment safer for trans and gender-nonconforming individuals. Removing these gender markers required coordination among the college, the testing institute and the third-party software vendor. In the end, the testing information infrastructure made it impossible to fully remove gender data from each person’s profile, but the visible marker was removed from the test vials at all institutions serviced by the testing institute. Addressing the issue at a root level through institutional coordination reduced the negative impact of the visible gender markers on trans and gender-nonconforming individuals. And through campus staff advocacy with receptive third-party vendors, the concrete improvement in trans inclusivity happened at a larger scale than had one institution simply tried for a local workaround.
What is a respectful way to ask? Rather than offering a single item to be used for all purposes, our argument is that questions related to gender should look different depending on what the data are for, who is being asked for their data and the legal compliance issues that may be at issue (e.g., Title VII, Title IX, FERPA, HIPAA).
For instance, if an institution seeks to craft trans-inclusive housing policies, it might revise the questions incoming students are asked on a student housing survey. These surveys prioritize student self-identification, while not requiring exhaustive disclosure, in aid of welcoming students to the university and informing their housing matches. Data collected in a different context—e.g., for health insurance (limited to male and female in many states)—would be insufficient and sometimes misleading for this purpose. In this context, questions about gender would be asked alongside questions about tidiness, sleep schedules and noise concerns. All these questions are shaped by understanding of nondiscrimination required by Title VII and Title IX. Conveying to incoming students where the data will be stored and who has access to them—and ensuring that they are secure—would be essential.
This case might lead to a two-question sequence, in which the first question asks about gender identity and the second asks about transgender status. (These particular options were informed by an article by Kronk et al., as well as conversations with Ash Bell and Avery Everhart).
- Are you (check all that you feel comfortable sharing here)
- A woman; female; a girl
- A man; male; a boy
- Another, nonbinary gender or no gender
- Questioning/exploring gender
- Prefer to self-describe (please specify):
- Do you describe yourself as transgender or having a history of gender transition?
- I prefer not to answer
You might follow this sequence with a question asking if students have a preference for living with someone who answered the same way they did. This not only allows trans students to indicate that they prefer to room with trans classmates of any gender; it also enables, for instance, women to ask to live with women regardless of trans status. Underneath each question, you could ensure you are offering a way for students to connect with housing staff if they have questions or concerns. Note that this item does not ask directly about assigned sex, although it does allow respondents to provide that information if they choose.
The above sequence is not exhaustive, nor is it perfect. It doesn’t include intersex or two-spirit as listed options. The ability to write free text offers maximum autonomy for survey takers, but it is a challenge for those working with the data. The decision to add these features will depend, as we said, on the purpose for which data are being collected.
This approach does not remove all possibilities for negative experiences, but it offers actionable data that staff can use to place students into housing that is most likely to work for them. A question sequence like this would have to come alongside a process that allows a more individualized accommodations process to ensure, for instance, compliance with the Americans With Disabilities Act and Title IX’s prohibition on gender-based discrimination, as well as the institution’s own nondiscrimination policies. Finally, crafting a survey like this and using it effectively would likely require bringing together LGBTQ student services, housing, legal counsel, information technology, accessibility/disability services, the registrar and Title IX offices. Without collaboration, any effort to improve the use of gender data in housing assignments is likely to fail or cause significant difficulty for individuals or the institution as a whole. It would be best to pursue this work in iterations over a few years and then to revisit as circumstances change.
What are you doing with the data? Even once a question has been drafted and accepted, this is no guarantee of trans inclusivity. We need to be vigilant to ensure that data collected for one purpose are not repurposed in unwarranted ways—not to assume, for instance, that a student’s housing preferences are a valid indication of their pronouns—and to provide simple, accessible procedures for records to be changed or corrected should circumstances warrant. The point is to begin with the fact that trans and nonbinary students, staff and faculty exist, so when we gather gender data, we must do so specifically with the aim of using it to either affect or inspire greater trans inclusion and gender equity. This is where many institutions in higher education have begun, and this is where we can continue to build from.