children playing

Canada’s “family portrait”: Measuring families in the census

Printer-friendly versionSend by emailPDF version
Nora Galbraith from Statistics Canada discusses the upcoming Census 2021 release on families in Canada.
Battams, Nathan
Publication Date: 
6 Jul 2022


How does Statistics Canada count and measure families, and how does this shape our understanding of families?

If we first zoom out from the topic of families, the main goal of the census is to count the population as accurately as possible. To do that, procedures are put in place to ensure that every person is counted once – and only once – and that they are counted as residing at only one dwelling and only one household.1 One of the ways the census accomplishes this is by focusing on the concept of usual place of residence: each person is counted as living in whatever is considered to be their “main” residence.

Under the usual place of residence concept, however, there are guidelines in the census questionnaire in order to determine where a person’s main home is. For instance, family members who live elsewhere for part of the year for work-related reasons should be included at their family’s home regardless of the amount of time they spend at this second residence.

Children who split their time throughout the year between the homes of two parents or guardians should be included in the home where they live most of the time. If they spend an equal amount of time with each parent or guardian, they should be included where they were staying on Census Day. Students who periodically return to their parents’ home should be listed only at this home, even though they may spend much less than half of the year there.

Once everyone has been recorded at their usual place of residence, families are formed based on the information provided in the census for a given household that describes who lives there and what relationships they have with each other.

Two key concepts used in the census release on families are that of census families and economic families.

A census family consists of a married or common-law couple – with or without children – or a one-parent family (formerly referred to as a lone-parent family). Children in census families may be biological or adoptive regardless of their age or marital status, as long as they live in the same dwelling as their parent(s) and do not have their own married spouse, common-law partner, or child living with them. By definition, anyone who is part of a census family is also part of an economic family.

An economic family refers to a group of two or more persons who live in the same dwelling and are related to each other by blood, marriage, common-law union, adoption, or foster relationship. An individual living with their aunt, uncle, or cousin, for example, would form an economic family. This perspective can be useful when examining housing or household income indicators. The concept of the economic family is mostly used in research related to income and housing.

This usual place of residence lens is an important consideration for those who use census data to examine family topics. We know that more and more people and their families divide their time throughout the year living at more than one address, whether it’s for work, school, shared parenting time arrangements, or other reasons. Census data may not always reflect the full complexity and fluidity of families, households, and living arrangements, but it provides a rigorous and reliable basis to examine trends at detailed levels of geography and for subpopulations of interest.

How has the concept of family within the census changed over the years, and what’s new in the upcoming release?

Perhaps more than any other census topic, concepts related to families and households have evolved over time, along with societal changes. If we go way back to the early decades of the census, prior to 1941, there was really no “census family” concept, and little distinction was made between household and family units.

Beginning in 1981, it became possible to identify common-law couples. The 2001 Census was also a key “turning point” for the family topic: same-sex couples were identified for the first time, and the concept of census families was expanded to include, among other changes, skip-generation families of a grandchild and grandparent living together with no middle-generation present.

Concepts also evolve as a result of legislative changes. For instance, in 2006 the census began distinguishing same-sex married couples following the legalization of same-sex marriage in the year prior. More recently, in 2011, foster children and stepfamilies were added to the types of family and living arrangements that can be measured in the census.

This evolution continues with the upcoming 2021 Census release, which will include new information on the gender diversity status of couples.

With the new gender variable, Census 2021 will now be reporting on transgender and non-binary people. How might this strengthen our understanding of families, family life, and family wellbeing in Canada?

As a result of the new gender variable, the upcoming census release on families will include new information on the gender diversity status of couples for the first time. This represents a new step forward in our ability to capture and understand the diversity of families in Canada.

As with the previous release on gender diversity, we hope that this information will mean families are better able to see themselves reflected in the census data. The information released on July 13 will be a launch point in our understanding of couples that include transgender or non-binary persons, in particular, but also same-gender couples and different-gender couples.5

Practically speaking, how exactly does Statistics Canada go about changing how families are measured? Why does it change (or not), and what are the impacts of these changes?

Statistics Canada strives to provide information about families that is timely and relevant. To do so, it is important to measure emerging forms of families and households. To this end, the agency is involved in ongoing consultation and engagement with stakeholders, data users, and the Canadian public in order to meet information gaps and respond to legislative changes. With the Census of Population, any potential changes to the questionnaire are arrived at in coordination with these ongoing engagements and are tested extensively using both qualitative and quantitative test formats.

Since one of the key benefits of the census is its long-time series, there is always a balancing act at play. Historical comparability and continuity of concepts are important, but at the same time certain concepts may need to evolve, expand, or be revised to be more inclusive and relevant to today’s realities.

As I mentioned earlier, as society continues to evolve, the growing complexity of families and living arrangements means that many families or living situations cannot always be adequately captured by a single term. In acknowledgement of these changes to our society, and following feedback from users, Statistics Canada is currently investigating how persons with multiple residences might best be measured using various statistical vehicles, such as the census, surveys, and other data collection activities.

In the coming months, Statistics Canada will consult with Canadians, stakeholders, and partners on this issue, and will assess the feasibility of testing content on multiple residences as part of preparations for the 2026 Census.

Broadly speaking, what does the census tell us about families and family life compared with other data sources, such as the General Social Survey (GSS)?

The Family cycles of the General Social Survey and the Census of Population have very different objectives and capture different data, and, as a result, have different strengths. In the Census of Population, the content relating to families represents only one of the many different topics it needs to cover, including income, housing, linguistic diversity, Indigenous identity, citizenship and immigration, ethnocultural and religious composition, mobility, and migration. The census is often deemed a “snapshot” of the population at a point in time, and families are identified based on their usual place of residence, as I discussed earlier.

In contrast, the GSS Family cycle is an in-depth, focused examination of the family characteristics of adults, including their conjugal and fertility history. The GSS dives into much more detail and teases out the “hidden complexities” of people’s family situations, for instance, whether they are in a relationship with someone who lives elsewhere (a “living apart together,” or LAT, relationship) or what sorts of parenting arrangements and decision-making responsibilities are in place with their ex-spouse or ex-partner. In other words, it has a unique strength of being able to explore family relationships that go beyond the door of the family home.

Unlike the census, the GSS also permits examination of intergenerational links, capturing information about their family origins and the characteristics of their parents. Finally, the GSS allows us to get a pulse on people’s values and views via questions about fertility intentions and intentions to marry in the future.

In contrast to the GSS, which is a sample survey covering the 10 provinces, the census often permits the examination of more detailed population groups of interest (e.g., less-populated geographic areas, or racialized or Indigenous people) that may not always be feasible with survey data based on a sample of the population. In the case of families and living arrangements, the census can capture less prevalent situations, such as skip-generation families, foster children, or multiple families living within the same household.6

Another advantage of the census is that it’s been conducted for a long time and offers a lot of data. With many of the core concepts, such as marital status, persons living alone, private households, and so on, we can go back nearly a century to examine long-term trends. This proves to be a very helpful perspective when we learn, for instance, that situations like young adults living with their parent(s), or children living in one-parent families with their father, are really not new phenomena.

I see the census and GSS-Family as sisters, each bringing different things to the table and together permitting us to achieve a broader understanding of families and family life in Canada.