Better Sexual Orientation and Gender Identity Questions Get Better Results

Collecting data on sexual orientation and gender identity (SOGI) is increasingly common in the homeless sector: during program intakes, on employment equity surveys, and as part of point-in-time (PiT) homeless counts, to name a few examples. This data is critical, because 2SLGBTQ+ people (Two-Spirit, lesbian, gay, bisexual, queer, and other sexual and gender identities) are both overrepresented among those experiencing homelessness and underserved by the sector meant to support them. However, the collection of SOGI data is not always done in a respectful or inclusive manner, which is harmful and reduces accuracy, as 2SLGBTQ+ people are less likely to self-identify if they do not feel respected or see themselves represented in the questions and response options.

Why Collect SOGI Data?

Among the population of homeless youths, at least 25% identify as 2SLGBTQ+, despite representing only 5% to 10% of all youths. The pathways into homelessness are different for gender and sexual minority populations than they are for cisgender and heterosexual/straight youths. For example, identity-based family rejection and discrimination on the housing market have been reported as primary pathways. 2SLGBTQ+ people are more than twice as likely to experience homelessness compared to cisgender and straight people. Trans people underutilize housing programs due to discriminatory practices that perpetuate transphobia. For example, housing programs may use the information on a person’s ID rather than asking about their gender identity or how they identify.

These figures are meant to show the level of need for services tailored to 2SLGBTQ+ people, which can only be done if we have access to reliable data on their service use. However, the lack of gender identity data collection in housing, healthcare, and research settings makes it almost impossible to identify trans people. In recent years, the national PiT count and Toronto Street Needs Assessment have started to identify 2SLGBTQ+ people by including SOGI questions on surveys. Unfortunately, the demographic questions included on the majority of national surveys are inconsistent, with gender often still being a binary variable and sexual orientation being left out.

How Is SOGI Data Being Collected?

SOGI questions have evolved over time. For example, the 2016 national PiT count survey included one question on gender identity, an additional question on sexual orientation was added in 2018, and the questions were again refined for the 2020–2022 count.

In 2021, Canada became the first country in the world to collect and publish national data on gender identity using the Statistics Canada census. The 2021 census included a two-step approach, collecting data on sex assigned at birth and gender identity. However, this approach falls short on several levels. Notably, many people who have transitioned no longer identify as trans (identifying simply as the gender they transitioned to) and will be incorrectly classified as such based on these questions. And importantly, this question can be triggering for people of trans experience, particularly those who no longer identify as trans.

We can do better than this, and there are emerging best practices on how to collect SOGI data.

A Proposal for Asking about SOGI

We propose a three-step process:

• Sexual orientation
  1. Which of the following categories best describes your sexual orientation? This question is accompanied by a list of options, including straight/heterosexual, gay/lesbian, bisexual, Two-Spirit, and asexual.
  2. Describe your sexual orientation in your own terms.
• Gender identity
  1. Which of the following categories best describes your gender identity? This question is also accompanied by a list of options, including man, woman, trans man, trans woman, non-binary, and Two-Spirit.
  2. Describe your gender identity in your own terms.
• Do you identify as 2SLGBTQ+?

The two-step approach to questions 1 and 2 allows us to collect data about specific categories, while also getting more fine-grained data about how people identify in their own terms. In some situations, it may be practical to use a single-step model for these questions by just making the lists of options as broad as possible and then collapsing the categories during analysis, if necessary (e.g., non-binary, gender-queer, and gender-fluid can be collapsed into a single category, as can trans man and transmasculine).

Another important point about these questions is that it is rarely necessary to collect data on sex assigned at birth. This information is not necessary for understanding a person’s gender identity in the present moment, which is what counts. Importantly, question 2 does not specify that “man” and “woman” are cisgender people—they can be selected by anyone who does not specifically identify as trans, including people who have trans experience.

The third question is important because many people with trans experience still identify as 2SLGBTQ+ even if they don’t identify as trans, and this would not be captured by the other questions. For example, a person may select “heterosexual/ straight” as their sexual orientation and “woman” as their gender identity but selecting “yes” to the third question would allow them to identify as part of the 2SLGBTQ+ population. The third question is meant to reduce the degree to which 2SLGBTQ+ people are undercounted.

Considerations When Asking SOGI Questions

When asking SOGI questions, it is important to understand and be prepared to explain why these questions matter. It is also important to be ready for the different types of reactions people might have. It is recommended that you practise asking the questions in advance, because if you are not comfortable asking, people will not be comfortable responding. As well, all questions should be treated in a similar manner, meaning that it is important not to single out the SOGI questions (e.g., do not administer them in writing but then read out all of the other questions). If there is anything you are unsure about, you can consult the guidebook on inclusive and affirming PiT counts.

Again, SOGI data is important, and it needs to be collected respectfully, not least because this allows it to be as accurate as possible. A respectful program intake makes it more likely that 2SLGBTQ+ people will access services, and respectful questions during homeless counts will reduce undercounting and allow for better service decisions. Although there is no fixed best practice, our hope is to contribute to a dialogue on inclusive and respectful SOGI data collection with the goal of sharing what has worked.

For more information about 2SLGBTQ+ people experiencing homelessness, check out the free, self-paced training on the Homelessness Learning Hub.

Dr. Alex Abramovich is a senior scientist at the Institute for Mental Health Policy Research at the Centre for Addiction and Mental Health (CAMH) and an associate professor at the Dalla Lana School of Public Health at the University of Toronto. He holds a Canada Research Chair in 2SLGBTQ+ youth homelessness and mental health.

C.L. Michel is the senior French content coordinator at the Canadian Observatory on Homelessness.