The Institut de la statistique du Québec (ISQ) began compiling information on the gender of its survey participants in 2019 in order to recognize gender diversity among the population. The concepts, definitions and approaches are based on those of Statistics Canada, which published its first standards on gender and sex at birth in 2018.
What we mean by gender
A person’s gender refers to their personal and social identity as a man, woman, or non-binary person (meaning a person whose gender is not exclusively male or female).
- Gender identity refers to the gender that a person internally feels.
- Gender expression refers to the gender that a person displays through outward signs (such as clothing or accessories) that may traditionally be associated with a specific gender. The words used by a person to describe themselves also contribute to their gender expression.
A person’s gender may be different from their sex at birth and may change over time. It may also differ from what is recorded in official documents.
Sex at birth refers to the sex assigned to a person at birth based on their reproductive system and other physical characteristics.
It is recommended that statistics be produced according to gender when they relate to life choices, lifestyles, or living conditions. Data collection and production according to sex at birth remains relevant for some demographic and health statistics.
A few definitions
Cisgender persons: Persons whose current gender is the same as their sex assigned at birth.
Transgender persons: Persons whose current gender is not the same as their sex assigned at birth. A transgender woman is female; a transgender man is male.
Non-binary persons: Persons whose current gender is not exclusively male or female. Some people use the expression “non-binary” to describe their gender; others use terms such as agender, pangender, genderqueer, genderfluid, gender-nonconforming, or two-spirit, a term specific to some Indigenous peoples of North America.
The question used by the ISQ to collect data on gender in its surveys comes from Statistics Canada. This question was selected to ensure comparability between the statistics produced by both agencies, among other reasons.
If necessary, infotips provide additional information in the electronic questionnaires. When the questionnaires are administered by telephone, this information is read as needed.
This measure does not explicitly refer to the person’s gender identity so that the information can be provided by a third party based on their knowledge or perception of the gender of the person about whom information is requested (for example, a parent may have to provide information about their child). In such cases, the question is phrased slightly differently: “What is this person’s gender? refers to current gender, which may be different from sex assigned at birth or may be different from what is indicated on legal documents.”
The third response option, which consists in an open answer, aims to obtain a valid answer and allows people whose gender is not exclusively male or female to describe their gender in their own words.
Sex at birth must be compiled to produce statistics on the population of transgender persons: it’s the cross-classification between the answers to both questions – the one on gender and the one on sex at birth – that makes it possible to know if a person is transgender or cisgender.
Production and dissemination of results
The gender variable includes three categories for which results are produced: men, women, and non-binary persons. The latter includes all persons whose reported gender is not exclusively male or female. Moreover, the “men” category conceptually includes cisgender and transgender men, while the “women” category conceptually includes cisgender and transgender women.
Issues related to the precision of estimates and confidentiality
Given the small number of non-binary persons, very large samples are needed to produce statistics on this population that are acceptably precise and that ensure the confidentiality of the information collected. For more information on this topic, please read the document Notions statistiques pour l’analyse de données d’enquêtes (in French only).
When the production of such statistics is not possible, the data on non-binary persons are distributed between the two other gender categories to protect confidentiality. This distribution is done randomly and allows data on non-binary persons to be kept in the analyses. For most ISQ surveys, the dissemination of results by gender must be done with a binary gender variable (divided into two categories: “men” and “women”), which includes the distribution of non-binary persons.
Note that the same issues affect transgender persons, since they are also present in small numbers. The production of statistics on this population also requires large samples. In addition, as previously mentioned, we use both the question on gender and the one on sex at birth to produce statistics on this group of people.
Comparability over time
The publication of results with the two-category “gender” variable (after distribution of non-binary persons) rather than with the previously used “sex” variable should not pose any comparability problem over time given the small size of the non-binary and transgender populations. It is also probable that the information on sex that we previously had for transgender persons corresponded to their gender for some of them.
Gender diversity among the population
For the first time in 2021 in Canada, the concept of gender was introduced in the Census of Population. The census collected information from the Canadian population on gender and sex at birth, and not only on sex. This change made it possible to count the number of cisgender, transgender, and non-binary persons. Census data on gender diversity are produced at the provincial and census metropolitan area level and provide a distribution of the population according to the following categories:
- Cisgender persons: total; male; female;
- Transgender persons: total; male; female;
- Non-binary persons.