Supplement to Statistics Canada’s Generic Privacy Impact Assessment related to the Survey on Family Transitions

Date: March 2024

Program manager:

  • Director, Centre for Social Data Integration and Development
  • Director General, Social Data Insights, Integration and Innovation

Reference to Personal Information Bank (PIB):

Personal information collected through the Survey on Family Transitions is described in Statistics Canada’s “General Social Survey” Personal Information Bank. The Personal Information Bank refers to information collected through Statistics Canada’s ad hoc surveys, which are not part of the regular survey taking activities of the Agency. The General Social Survey covers rotating themes including: social support, time use; giving, volunteering and participating; victimization; families; and caregiving and care receiving. Personal information may include the first and last name and date of birth of each member of the household as reported by a member of the household; and the first and last name and date of birth of the selected respondent as reported by the selected respondent.

"General Social Survey" Class of Personal Information (Bank number: StatCan StatCan PPU 155) is published on the Statistics Canada website under the latest Information about Programs and Information Holdings chapter.

Description of statistical activity:

Under the authority of the Statistics ActFootnote 1, Statistics Canada has been conducting the voluntary General Social Survey (GSS) on Families approximately every 5 years since its inception in 1990. Following a program redesign, it has been renamed the Survey on Family Transitions (SFT) which is scheduled for collection from April to September 2024. Aiming at providing key insights into Canada’s demographic dynamics, the SFT focuses on the union and childbearing trajectories of the respondent.

The SFT is designed to prioritize comparability to the previous GSS family cycles (1990, 1995, 2001, 2006, 2011 and 2017), thereby allowing for the analysis of historical and emerging trends in family formation and dissolution. This is required for the analysis of sociodemographic trends in the country. The results of the new iteration are also key to the analysis of potential impacts of the COVID-19 pandemic on family and fertility behaviour.

The survey was developed by Statistics Canada, in close collaboration with government and academic stakeholders including:

  • Economic and Social Development Canada (ESDC)
  • Immigration, Refugees and Citizenship Canada (IRCC)
  • Canadian Heritage (PCH)
  • McGill University
  • Guelph University
  • Université de Montréal
  • University of Saskatchewan
  • Women and Gender Equality Canada (WaGE)
  • the Public Health Agency of Canada (PHAC)
  • Ministère de la famille et des aînés du Québec
  • Institut national de la recherche Scientifique (Montréal)
  • University of Western Ontario
  • University of Alberta
  • Institut de la famille du Québec

The SFT will be the only source of information in Canada that collects information on the timing of family trajectories and that will allow to elaborate statistics on union and fertility (for instance on life-long prevalence of common-law union, separation, repartnering or multiple-partner fertility).

The SFT will sample a total of 51,000 individuals that are representative of the Canadian population aged 20 to 79 years old living in private dwellings in the 10 Canadian provinces excluding those living in institutions and in other collective dwellings, such as the Canadian Armed Forces bases. The sample will be selected from the 2021 short-form Census survey frame as well as from administrative sourcesFootnote 2, which ensures that the immigrant population who landed after the 2021 Census is also represented in the sample. The SFT aims to provide reliable data on family life trajectories for persons in each region in Canada, and by immigration status.

The sample is selected from respondents that provided valid contact information for the census, which was then used to create a sample of targeted respondents with an oversamplingFootnote 3 of individuals who are immigrants to Canada to ensure enough data will be captured to be able to release disaggregated statistics on relevant sub-groups that meet Statistics Canada’s Quality Guidelines. Given this oversample, in its 2024 iteration, the SFT will receive funding under the Disaggregated Data Action Plan (DDAP) which was part of the Government of Canada’s federal Budget 2021. It will collect information about family life experiences to provide information for specific immigrant populations in Canada.

If a respondent is confirmed to meet the survey selection criteria to continue with the survey (based on age and province of residence), they will be asked questions related to the following topics:

  • Conjugal unions: length of current and previous common-law unions and marriages, number of children born in each union.
  • Spouses/Partners: gender and age (or month and year of birth) of current and previous spouses or partners. Socio-demographic characteristics of current partner (labour market activity, population group, Indigenous identity, place of birth and educational attainment).
  • Children:number, gender and age (or month and year of birth) of children born/adopted or step from any previous or the current union, residential status of each child (if younger than 19).
  • Grandchildren: number, and age (or month and year when became a grandparent for the first time).
  • Family background: whether the respondent grew up with both parents, one or none, socio-demographic characteristics of parents (age or month and year of birth, place of birth and educational attainment).
  • Leaving the parental home: whether they left or returned and when (age/month or year).
  • Basic sociodemographic information on the respondent (harmonized content): date of birth or age, gender/sex at birth, population group, place of birth, immigration, educational attainment, labour market participation, main activity, school attendance, Indigenous identity, population group, language, religion, sexual orientation, general health, and life satisfaction.
  • Family related information: marriage and fertility intentions (situation and age depending), division of household tasks, and other people living in the household (age, gender, and relation between household members).

While the respondent is the sole sampling unit and unit of analysis, some information about their family members is collected in order to properly characterize the respondent’s trajectory and socio-economic status. The personal information is used to produce statistical data on social trends in order to monitor changes in the family trajectories and well-being of Canadians over time, and to provide immediate information on specific social policy issues of current or emerging interest: population ageing, fertility decline, childcare needs, family instability and well-being. Statistical analyses are to only be made about the respondent, rather than about any other individual whose information may also be collected from the respondent.

For recent cycles of the GSS, the information has been linked to income tax data files (PIB: StatCan PPU 111) in order to obtain personal and household income and reduce burden on respondents. In addition to the information collected from respondents, microdata linkages will also be performed for the SFT in order to reduce respondent burden and enhance the analytical value of the data. This includes linkages to tax data files and the Longitudinal Immigrant Database (IMDB). The IMDB includes information on immigrants that is not collected within the survey (e.g., admission category, intended destination, knowledge of official languages, etc.). The additional information obtained through such linkage will allow conducting more in-depth comparative analysis of family trajectories, especially between non-immigrant families and different types of immigrant families. This type of analysis, made possible by the linkage, is particularly relevant given that the survey includes an oversample of immigrants.

These linkages will support and improve insight and decision-making by encouraging and facilitating more in-depth policy analysis with regards to the well-being of persons with different types of family trajectories, which are of special interest to policymakers and researchers. This will improve insight with regards to, for example, the socioeconomic well-being of Canadians who experience trajectories of union instability, allowing research on the intersectionality between economic well-being, lived family experiences, gender and immigration status.

Statistics Canada’s microdata linkage and related statistical activities were assessed in Statistics Canada’s Generic Privacy Impact Assessment.Footnote 4 All data linkage activities are subject to established governanceFootnote 5, and are assessed against Statistics Canada’s principles of necessity and proportionalityFootnote 6. All approved linkages will be published on Statistics Canada’s websiteFootnote 7.

Availability and access to the microdata from the SFT will focus on using existing Statistics Canada access mechanisms such as the Federal Research Data Centre (FDRC), and research data centre network (RDC)Footnote 8. The data will also be available in non-confidential aggregated formats via Statistics Canada’s website.

Reason for supplement:

While the Generic Privacy Impact Assessment (PIA) addresses most of the privacy and security risks related to statistical activities conducted by Statistics Canada, this supplement describes potential new risks associated with the collection, processing, and use of data related to some of the Survey on Family transitions (SFT) content, and the possible concerns of Canadians about the intrusiveness of the collection.

Though respondents must be aged 20-79, some questions may be particularly sensitive. The SFT will collect information about the respondents’ family life and trajectories, including some personal information regarding dates of family events, gender identity, sexual orientation, ex-spouses/ex-partners, children, and other family/household members. As is the case with all PIAs, Statistics Canada's privacy framework ensures that elements of privacy protection and privacy controls are documented and applied.

Necessity and Proportionality

The collection of personal information for the Survey on Family Transitions (SFT) can be justified against Statistics Canada’s Necessity and Proportionality Framework:

  1. Necessity:

    The SFT collects information on the timing of family events: it focuses on when a series of life events happened, allowing researchers and policy makers alike to understand how several aspects of the family life intersect. Leaving the parental home, getting married, and having children, have all been increasingly delayed in recent decadesFootnote 9, while the number of children Canadians have has also significantly gone downFootnote 10. Understanding the associations between all these processes has relevant policy implications for adequately targeting interventions to the appropriate stage of Canadians’ life cycles. For instance, if fertility decline is linked to later union formation and childbearing delays - which have, in turn been linked to increasing difficulties in establishing residential and financial independence from the family of origin, policy-makers concerned with fertility and population aging might want to focus their efforts in policies targeted at facilitating home leaving for the youth, rather than on policies focusing on childbearing itself, which might be ineffective in this contextFootnote 11.

    The survey collects information on past events, so that this biographical information can be used to establish different types of life trajectories and their links to current socio-economic well-being. These questions allow analyzing how individuals with different types of family life trajectories fare in the present day, what type of family structures they currently have, and what specific needs they might have depending on their family trajectories. Some of the family events whose timing the SFT collects are known to have a long-lasting impact on the individual’s socio-economic well-being (for instance a separation or divorceFootnote 12). The collected information thus allows identifying crucial turning points in the individual life course that may lead to increasing vulnerability and exclusion, information that is needed by both researchers and lawmakers.

    In order to identify the long-term impacts of family events (reported in the survey) on socio-economic well-being, linkages between the survey data and administrative datasets are required to obtain information on income, if the respondent consents to it. This allows assessing whether certain types of family trajectories (past events reported in the survey) are linked to current socio-economic outcomes. These correlations are not only of interest to researchers, but they also allow policy-makers to identify sources of inequality and inform policy. For example, targeted policies might prove more effective to prevent financial strain during the post-separation period for specific types of families that tend to experience increased financial strain, for instance when there is childbearing across householdsFootnote 13. Events and when these were experienced are collected to measure the diversity of family situations and the plurality of family paths. The basic socio-economic characteristics (age and gender and previous parity) collected in the SFT about former spouses or partners are used to demonstrate what types of families might be affected by family transitions such as union dissolution.

    Some of the aspects covered by the survey are of particular importance in a post pandemic period as it will provide insight as to the changes to family trajectories brought by the pandemic. The socio-demographic changes observed in the post-pandemic periodFootnote 14 generate a context in which family scholars have expressed their interest in comparing the newly gathered data with previous cyclesFootnote 15, in order to assess emerging trends in family behaviourFootnote 16 and identify needs among Canadians.

    Given the importance of understanding socio-demographic processes as those mentioned above, Federal and Provincial Departments (e.g., Employment and Social Development Canada, Justice Canada, Women and Gender Equality, Ministère de la Famille du Québec) make use of the data to better understand how the needs of Canadian families have been changing over the past decades and assess the relevance of programs and policies related to families. Such programs and policies focus on reducing family and child poverty (child tax benefits or transfer payments to families), work-life balance (enhancing childcare services and leave practices by parents, promoting women’s participation in/return to the labour force), and gender-based analysis (pay equity between men and women, gender equality in the involvement of mothers and fathers in family life). The SFT will be a key data source for these programs and policies. In addition, these data are used to assess not only the progress Canada is making towards such goals, but also how it fares in the international scene compared with other countries.

  2. Effectiveness - Working assumptions:

    The 2024 SFT was carefully designed to produce relevant, high priority, and statistically meaningful information in order to meet the outlined goals. The personal information collected will help meet the objective of gathering timely data on sociodemographic processes. The SFT is tailored to provide specific information for measuring the types of family life courses experienced by Canadians and has taken steps to ensure it captures experiences from respondents of diverse backgrounds, including gender and immigration status. Information on the timing of family events, particularly on childbearing, provides for an accurate way to calculate demographic estimates and a complementary source to census data.

    The SFT includes retrospective biographical questions that ask the month and year, or the age, at which several events occurred. This allows to establish not only what type of family events happened in the life course of the respondent (home leaving, marriage, common-law, separation, divorce, childbearing), but also when these events happened, and in which sequence. Collecting this retrospective biographical data has been proven as an effective way to accurately assess the family trajectories of individuals not only by the three decades of the General Social Survey on Families, but also by its international counterpartFootnote 17. The SFT collects thus not only effective but also internationally comparable data on the timing and types of family trajectories, which has been used in countless academic articles and government reports.

    The oversample of immigrants is an effective method to fill data gaps on issues related to family formation and dissolution among an underrepresented demographic group. Due to the lower prevalence of immigrants in the general population, contrasted with the need to produce more detailed disaggregated data on this population, immigrants will be oversampled in the SFT. The differences in sampling rates between the various target groups makes it possible to carry out more disaggregated and intersectional analysis with this survey than a survey with a general population sample design.This allows for the identification of family patterns that might be specific to certain groups, and to assess whether specific policies need to be put in place to increase their socio-economic wellbeing.

    Thus, the survey methodology combined with the rigorous testing of the questionnaire content enables the effective production of robust statistics that meet Statistics Canada’s quality guidelines and provide insights into differences and similarities in experiences among Canadians of different socio-economic backgrounds, and especially among immigrants, compared to the rest of the Canadian population.

  3. Proportionality:

    Some of the questions included in the questionnaire of the SFT might be considered of a sensitive nature. These questions were assessed as required to determine the extent of the effect of family related experiences on respondents’ lives. Statistics Canada undertook a thorough assessment of the content of SFT’s predecessor, the GSS on Families, in order to remove content that is not essential to meeting the objectives of the SFT and to reduce respondent burden. This exercise resulted in the elimination of approximately one quarter of the previous content.

    The questionnaire content was developed in collaboration with subject matter experts from the academic sector as well as from within Statistics Canada. Statistics Canada's Questionnaire Design Resource Centre (QDRC) provided input and feedback for the survey questionnaire after conducting a series of qualitative testing interviews with individuals who have similar characteristics of the sample. Overall, QDRC participants reacted positively, and the topics covered in the survey were of general interest, and no issues or major concerns were observed about the proposed topics.

    Given its retrospective nature, the SFT collects detailed information on family events that took place outside the country (births, adoptions, union formation and dissolution). These include key family events experienced by immigrants before their arrival in Canada, or by Canadians while living abroad. This information provides the full family life course of individuals, thus allowing to make comparisons between immigrants and non-immigrants. This, in turn, allows for the identification of family patterns that might be specific to immigrantsFootnote 18, and to assess whether specific policies need to be put in place to increase their socio-economic wellbeing as it relates to their family trajectories. The content of the SFT is comparable to what is being done elsewhere. Therefore, it is possible to conduct comparative studies with survey data from other countries, for example, the Generations and Gender Survey (GGS), the German Family Panel Data (PAIRFAM), and the National Survey of Family Growth in the United States (NSFG).

    The survey collects information about certain individuals related to the respondent (e.g., spouses or partners, former spouses or partners, children, etc.). The personal information collected is purposely limited in scope and it can only be analyzed in ways that relate to the respondents themselves. The information collected about the family members of the respondent can thus only be used in ways that: a) define an attribute or characteristic about the respondents themselves (e.g., respondents with a spouse; respondents with younger/older children) and b) is at a level of disaggregation large enough to guarantee the protection of the respondents’ family members privacy.

    For particularly sensitive variables, the following considerations were taken into account:

    • Gender identity and sexual orientation: The collection of gender identity and sexual orientation allows for a better understanding of the diversity of family configurations pertaining to the 2SLGBTQIA population. Improving data collection on the latter was an important priority identified by the Privy Council Office (PCO)Footnote 19. The collection of these variables will allow analysing this segment of the population previously not covered, which aligns with the intentions of the Disaggregated Data Action Plan and also the Government of Canada’s 2SLGBTQI+ Action PlanFootnote 20. The latter highlights the need for intersectionality in policy design in order to serve “underserved and disenfranchised populations, particularly 2SLGBTQI+”. Whereas other studies have separately pointed out poorer socio-economic outcomes among the 2SLGBTQI+ populationFootnote 21, and their more instability-prone family trajectoriesFootnote 22, little is known about how these factors intersect and how they play out over the life course of the 2SLGBTQI+ population. The inclusion of both gender and sexual orientation allows to measure the potentially distinct family life trajectories and transitions that might make individuals more prone to several well documented outcomes and the “disproportionate health, social and economic inequities experienced by some 2SLGBTQI+ communities”Footnote 23. In order to formulate adequate policies targeting potentially vulnerable life course transitions (home leaving, separation) for this population, adequate data needs to be collected that will allow identifying such turning points in family trajectories.
    • Religion: Religious affiliation is one of the dimensions that allows understanding the background of the respondent: it is indicative of the social norms the respondent ascribes to, and those constitute a key aspect influencing family behaviours. Given the increasing share of immigrants in the Canadian populationFootnote 24 and the greater diversity in religious affiliation among themFootnote 25, the different makeup of religious affiliation of the recent immigrants could impact their family behaviour, affecting for instance fertility rates in Canada. Religion has indeed been thoroughly documented as a factor in explaining fertility behaviour: more religious individuals tend to have higher fertilityFootnote 26. Religious affiliation is also related to divorce: in Canada, more religious individuals are less prone to divorceFootnote 27.  The SFT predecessor, the GSS Families, has been used to disentangle the effects of religious attendance and religious affiliation from other factors on the risk of union dissolution in Quebec. Individuals with no religion, and those who attend religious services less frequently are more prone to experience dissolution, even after controlling for numerous other variablesFootnote 28. Given the well-known link between religion and family behaviour, including these variables is necessary to carry out accurate analysis of family behaviour. Isolating the potential role of religion from other factors that affect family transitions is also key in order to formulate appropriate family policy: in order to properly understand which characteristics can drive certain behaviours and outcomes, it is necessary to control for normative variables such as religion. This is especially applicable in the context of an immigrant oversample in the survey, where a higher diversity of religious backgrounds is expected.
    • Names or pseudonyms of related individuals: The SFT asks about partnerships, former partnerships and children, as well as characteristics of the respondent’s parents. The first name or pseudonym of those individuals is asked in order to automatically populate in pursuant questions to assist respondents in keeping track of questions about each of these persons. Qualitative testing demonstrated that when a first name or pseudonym is incorporated directly into the associated questions, it helped respondents better understand and answer these questions.

      Further relationships with related individuals are assessed in the following ways:

      • Current spouse/partners: Questions are asked about the characteristics of couples in Canada, whether they are married, living common-law or living apart together (i.e. in a relationship and living in separate dwellings). These questions allow examining the diversity of couple forms in Canada, partnership duration (stability and instability), and the fertility of couples in Canada.
      • Former spouse/partners: When couples separate or divorce, it impacts the lives of both former spouses/partners, as well as their children’s. The SFT collects data to shed light on the different impacts of separation or divorce: economic and financial consequences, the ongoing responsibilities for care of children and child custody, the role of each ex-spouse/ex-partner, and work-family arrangements. The length of time the spouses/partner have been together and their ages at the time of separation are important variables to study the prevalence and impacts of separation, repartnering and multiple-partner fertilityFootnote 29.
      • Children: Determining the number of children a person has had or adopted is an important measure of fertility, especially when Canada has hit a record low total fertility rate in 2022. Asking about children (young or old) allows to learn more about family compositions in Canada (two-parent intact families, blended families, single-parent families), arrangements separated parents may have, and to identify needs in terms of parental leave and childcare services.
      • Parent(s): Family environment while growing up, and changes to this environment are very important variables when studying the family life course of Canadians. Experiencing parental separation or divorce during childhood has been demonstrated to be associated with partnership stability during adulthoodFootnote 30. In addition to finding out whether the respondent lived with both parents at birth and up to age 15, new questions ask about the main reason for not living with both parents. Moreover, instead of asking questions about the characteristics of “the mother” and “the father” of the respondent, as was done in the past, the questionnaire first asks respondents who they consider as their parents (if any). This allows to better capture family diversity while ensuring better inclusiveness of all family types.
      • Other household members: The inclusion of details on other household members allows to better understand, complement, and supplement the collected information on life course trajectories. During consultations, stakeholders and partners deemed this information as critical to a survey on families. Furthermore, it allows identifying relationships in multiple family households and to identify otherwise concealed families (especially important in multi-generational households).

    Family identity and diversity: As society becomes increasingly diverse, family identity and diversity play a large role in shaping individuals’ life experiences and outcomes, including family trajectories. These can be influenced by many factors: family traditions, country of origin, geographic regions, ethnic identities, cultural groups, community norms, religion, etc. Since individuals belonging to vulnerable populations have specific family trajectories, the analysis needs to take into account information on family identity and diversity as these are factors that might make individuals more or less vulnerable to specific transitions, events and outcomes. For instance, lesbian, gay and bisexual individuals are much less likely to be married and to live with children, and more likely to live in less stable couple arrangements like common-law unionsFootnote 31. Moreover, the Indigenous population is younger than the non-Indigenous population. and a higher share of Indigenous children live with grandparents or with only one parentFootnote 32. A younger age structure and higher rates of lone parenthood are also more frequent among the black populationFootnote 33. By collecting information on population group, gender, sexual orientation and Indigenous identity, the SFT would provide the latest information on family composition and family trajectories for specific population groups, and that would contribute to the development of program and policy for these respective populations.

    Statistics Canada is committed to providing information about family identity and diversity to inform decision making that supports family well-being (childcare services, parental leave, financial agreements and child custody after a family breakup, workplace policies and practices, etc.). Stakeholders require up-to-date information on the above sensitive topics to respond to changing social conditions where family life reflects evolving notions of what constitutes a family.

  4. Alternatives:

    While most Statistics Canada’s surveys, as well as the census, are focused on providing an accurate snapshot of society at the time of the interview, the SFT is unique in its retrospective nature. In other words, the survey is not focused on the measurement of the respondent’s current situation, but on the family events the respondent has experienced in the past, and in how and when they got to their current living and family situation. More precisely, this survey collects information on the timing of family events, which is not accessible through any other means.

    The SFT covers information that is not otherwise collected by Statistics Canada. No alternative data sources currently exist which include sufficiently precise, complete, or robust data to replace the information collected by the survey. In particular, the SFT questions are the sole data source for information on:

    • Biographical information on the family life course of Canadians
    • General indicators on couple relationships in Canada
    • Family histories for events that took place outside of Canada
    • Internationally comparable data

    While some administrative or fiscal data sources can complement the type of information collected on the SFT, administrative data are not collected with the same goals in mind and are thus not an accurate replacement for biographical data on family events and transitions.

    The SFT will also be linked to other administrative data in order to reduce the burden placed on respondents to answer additional questions where possible. This linkage is the only way to derive the additional insights from the personal information collected as there are no alternative methods to analyze this information.

Mitigation factors:

While some questions contained in the survey can be considered sensitive, the overall risk of harm to the survey respondents has been deemed manageable with existing Statistics Canada safeguards that are described in Statistics Canada’s Generic Privacy Impact Assessment, as well as with some additional measures. These include the following that are of particular importance in the context of this survey:

Sensitivity

  • A full review of the content was conducted with key stakeholders to keep only content that is directly relevant to the objective. As a result, about a quarter of questions were cut from the previous GSS on Families.
  • Respondents are provided with the possibility to skip questions or provide partial responses.
  • Short explanatory texts have been inserted where deemed necessary. These address the usefulness and purpose of the questions and help survey participants be made aware of the upcoming content before reading questions that are potentially sensitive to them.

Transparency and consent

  • Prior to the survey, respondents will be informed that the survey is voluntary, and of the survey purpose, allowing them to decide whether they wish to participate. The SFT is collected using both respondent self-completed electronic questionnaire (rEQ) and interviewer assisted telephone interview (iEQ) collection methods. Interviewers will be thoroughly trained and have information readily available (interviewer manual, reference cards, PowerPoint presentation) to reply to respondent questions about the survey should the need arise.
  • Individuals selected for the SFT are informed of survey content and privacy, using all means available to the agency. An invitation letter with an information brochure (see Appendix A) will be mailed out to selected respondents and will be available in the six most spoken non-official languages in Canada. The information regarding the survey content and its objective is also available on the Information for Survey Participants (ISP) Statistics Canada webpage, on the survey specific webpage, and in the questionnaire (which can also be found on the survey specific web page).

Conclusion:

While the survey raises potential privacy concerns, this assessment concludes that, with the existing Statistics Canada safeguards including those listed above, any remaining risks are such that Statistics Canada is prepared to accept and manage the risk.

Date modified: