Enduring Diversity: Living Arrangements of Children in Canada over 100 Years of the Census
The first half of the 20th century
Information identified as archived is provided for reference, research or recordkeeping purposes. It is not subject to the Government of Canada Web Standards and has not been altered or updated since it was archived. Please "contact us" to request a format other than those available.
Over the first half of the 20th century, Canada’s social landscape changed dramatically. Increased urbanization and industrialization, medical and sanitary innovations, the Great Depression and two world wars were among the factors that influenced the everyday lives of children in Canadian families.
At the turn of the 20th century, Canada was in the midst of a demographic transition from an era of relatively high mortality and fertility to one of improved population health, life expectancy and increased fertility control and limitation. At the same time, Canada became a major receiver of international immigrants, affecting the regional and ethnic distribution of the population.Note 2 Glimpses from early censuses show that, at this time of great transition, many children in Canada had a very different experience of childhood compared to today.
The census has only recently begun to distinguish certain diverse family forms such as stepfamilies, skip-generation families and multigenerational households. However, children’s living arrangements have always varied to some degree. At the turn of the 20th century, census families (a couple, with or without children, or a lone parent with one or more children) were much more open to admitting people other than immediate family members into their homes, either for additional income or to give or receive care or financial support. Within this “culture of household sharing”,Note 3 it is estimated that about one in three (30.8%Note 4) census-family households in 1901 contained additional persons (non-census family persons and/or other census families) compared to 9.2% in 2011.Note 5 Most of these additional household members in 1901 were other relatives, lodgers, boarders or employees of the family head.Note 6 A wide variety of extended relatives, such as “stepdaughter-in-law”, “half sister”, “great nephew” and “goddaughter” were among the household relationships recorded by 1901 Census enumerators.Note 7 These diverse family living arrangements were in many cases a result of the death of one or more family members.
Death within the family—of siblings, of mothers during or following complications from childbirth, of fathers serving in war, for example—was a much more common experience for young children in the early 20th century than today. In 1921, about 1 in 11 (8.9%) children aged 15 and under had experienced the death of at least one parent, while 4.1% had experienced the death of both parents.Note 8, Note 9 In addition, children themselves experienced a much higher risk of death a century ago compared to today (Box 1).
In the early 20th century, the death of one or both parents often resulted in children living with relatives or non-relatives. Parents could also send their children to live elsewhere to attend school (Box 2), become an apprentice, earn wages (Box 3), or simply reduce the economic burden on families experiencing financial difficulties.Note 10 It is estimated that approximately 55,000 children aged 14 and under had a non-parental guardian in 1901,Note 11 representing 3.0% of the total population in that age group. In comparison, there were 29,600 foster children aged 14 and under reported in the 2011 Census, representing 0.5% of this population.Note 12
This living arrangement continued to exist during the difficult times of the Great Depression, as evidenced by a 1931 Census monograph that analyzed ‘guardianship children’, defined as ”persons other than own children” of the household head.Note 13 Guardianship of this nature was more widespread than institutional care at that time, with guardianship children living in private homes outnumbering children in institutions 4.3 to 1 in 1931.Note 14 Most guardianship children lived with a member of their immediate or extended family: approximately one-third (34.4%) lived with a grandparent, 28.5% lived with an uncle or aunt and 8.2% lived with a brother or sister.Note 15
Despite the prevalence of guardianship families during that era, many children remained with one parent in a lone-parent family. It is estimated that over 100,000 children aged 14 and under were living in a household with only one parent present in 1901, among which just over 60% lived with a lone mother.Note 16, Note 17 Indeed, the proportion of children who lived with a lone parent was nearly as high in 1931 (11.9%) as it was in 1981 (12.7%), as seen in Figure 1. Lone-parent families in the early 20th century were found across the socioeconomic strata of the country, and thus the experiences of children raised in such families were quite diverse. Remarriage also occurred, however, and many children experienced several different family structures over the course of their childhood. As with every census snapshot, an even higher share of children may have experienced transitions between two-parent families and lone-parent families than what has been captured by the census data.
Throughout the early 20th century, families continued to be relatively large owing to the influence of religion on family life, combined with less effective means of contraception, as well as the value of children in what was still a largely rural environment: it was not until 1931 that the proportion of the population living in urban areas surpassed the population living in rural areas in Canada.Note 18 From an estimated 6.56 children per woman in 1851,Note 19 the total fertility rate decreased to 3.48Note 20 children per woman in 1931; still well above the population replacement level at that time of 2.40 children per womanNote 21 but lower than for previous generations of women. Several changes may have played a role in this decrease, such as the difficult economic circumstances of this era, the uncertainties due to war, the transition to greater urbanization in conjunction with evolving modes of production and farming methods, the growing dependency of households on wages for their sustenance, changing attitudes of women and the growing cost of childrearing. A 1938 manuscript based on 1931 Census data addressed some of the reasons for the fertility decline as follows:
The early Canadian settlers were great individualists…In this society large families were common and children were generally regarded as an asset and a blessing…During the last seventy years, production has been centralized and activity of the individual producers has been narrowed to a specific job. Consequently, the family has become much less self-sufficient.Note 22
As life becomes more comfortable…an increasing emphasis is placed on the sacrifices which women must make to bear children. Regardless of other factors, an improvement in living conditions for the human race per se makes women more reluctant to undergo the travail and inconvenience of bearing child after child.Note 23
The same manuscript included a series of policy recommendations to reverse these growing tendencies and stimulate the birth rate. Among them were the implementations of unemployment insurance and family allowances—both relatively novel propositions at that time. The growing role of immigration in population growth was not anticipated, as it was remarked that “[i]f the present downward trend in natural increase of population continues, there is a real possibility that actual stability or retrogression [of the population] will be reached.”Note 24
- McInnis, M. 2000. “Canada’s Population in the 20th Century”, A Population History of North America, editors M.R. Haines and R.H. Steckel, Cambridge University Press, pages 371 to 432.
- Burke, S.D.A. 2007. “Transitions in household and family structure: Canada in 1901 and 1991”, Household Counts: Canadian Households and Families in 1901, editors E.W. Sager and P. Baskerville, University of Toronto Press, page 32.
- Calculated from Burke (2007). Tables 1.1 and 1.2.
- 2011 Census of Population.
- In the 1971 and the previous censuses of population, in a husband-wife family, the husband if present was automatically designated the family head. In a lone-parent family, the male or female parent was always the family head. In the 1976 Census, the term 'family head' was eliminated. Source: Wargon, S.T. 1979. Children in Canadian Families, Statistics Canada Catalogue no. 98-810.
- The Canadian Families Project. 2002. The National Sample of the 1901 Census of Canada: User’s Guide. The Canadian Families Project created a computerized national sample of individual-level information from the 1901 Census of Canada. The database contains a 5% random sample of dwellings, stratified by microfilm reel.
- 1921 Census of Canada volume III, Statistics Canada Catalogue no. 98-1921, Table 31.
- In comparison, in 2011, less than 1% of children aged 0 to 14 lived in a lone-parent family in which the parent was widowed.
- Darroch, G. 2001. “Home and away: Patterns of residence, schooling and work among children and never married young adults, Canada, 1871 and 1901”, Journal of Family History, volume 26, pages 220 to 250.
- Darroch, G. 2007. “Families, Fostering and Flying the Coop: Lessons in Liberal Cultural Formation, 1871-1901”, Household Counts: Canadian Households and Families in 1901, editors E.W. Sager and P. Baskerville, University of Toronto Press, pages 197 to 246.
- Milan, A. and N. Bohnert. 2012. Portrait of Families and Living Arrangements in Canada, Statistics Canada Catalogue no. 98-312-X-2011001.
- Pelletier, A.J., F.D. Thompson and A. Rochon. 1938. The Canadian Family, 1931 Census, monograph no. 7, Statistics Canada Catalogue no. 98-1931CM-7.
- Pelletier et al. 1938, page 94.
- The remaining 28.9% lived with an adoptive guardian or other person. Source: Pelletier et al. 1938, Table LXVII.
- Bradbury, B. 2007. “Canadian Children Who Lived with One Parent in 1901”, Household Counts: Canadian Households and Families in 1901, editors E.W. Sager and P. Baskerville, University of Toronto Press, pages 247 to 301.
- Bradbury notes that due to the difficulties in raising a child, many male parents, upon the loss of their spouse, would have given their children to relatives or for adoption. Thus, the children of surviving fathers were more likely to appear in the census data as living in households in which they had no parent present.
- Martel, L. and J. Chagnon. 2013. “Canada’s rural population since 1851”, Census in Brief, Statistics Canada Catalogue no. 98-310-X-2011003, Figure 1.
- Gee, E.M. 1987. “Historical Change in the Family Life Course of Canadian Men and Women”, Aging in Canada: Social Perspectives, editor V.W. Marshall, Fitzhenry and Whiteside, Table 4.
- Statistics Canada, Canadian Vital Statistics, Births Database, 1926 to 2011, Survey 3231 and Demography Division, Population Estimates Program.
- The cohort replacement rate varies over time as a function of the mean age of childbearing and female mortality rates. The 1931 replacement rate of 2.40 children per woman is calculated as 1 / (r * m) where r equals the female-to-male sex ratio at birth (0.488) and m equals the probability of survival of females from birth to the mean age of childbearing, which was approximately age 30 in 1931 (0.8536).
- Pelletier et al. 1938, page 193.
- Pelletier et al. 1938, page 195.
- Pelletier et al. 1938, page 195.