Individuals at greater risk of poverty in Canada have been identified in a number of studies. Hatfield (2004), in “Vulnerability to persistent low income”, identified five population groups that were particularly vulnerable: lone parents, unattached persons aged 45-64, recent immigrants, persons with work-limiting disabilities, and Aboriginal populations. Human Resources and Skills Development Canada (2009), in its call for proposals entitled “Resilience among at-risk groups in Canada”, adds young adults as a sixth group. Beaujot et al. (2010), in “Low income status by population groups, 1961-2001,” document the extensive change in the groups at higher risk of low income since it was first measured in the 1961 census, noting the increased vulnerability of young families with children, especially in families with only one earner.
The research program will focus on family and life course factors that increase the vulnerability of individuals and the factors that make them resilient in spite of their vulnerabilities. Thus, among the population groups listed above, we will concentrate on lone parents, young adults and unattached men and women at ages 45-64. As “the long arm of demography” links family transitions to vulnerability (Kiernan, 2002), our research will also bring in children; that is, early transitions can mean low human capital investments from parents and the broader society, making for vulnerability to lone-parenthood and thus, of “fragile families” in the next generation.
In spite of their vulnerabilities, many individuals do manage to deal with the risks and overcome their disadvantages. Our research will endeavor to examine resilience, defined as the ability to adapt positively to adversity (Garmezy, 1991; Masten, 1994; Werner and Smith, 1982; Werner, 1994; Luthar et al., 2000; Patterson, 2002; Lamond et al., 2008), and the factors and conditions under which some individuals are resilient in comparison to others in similar circumstances. We intend to develop theories specific to the influences of individual agency, familial support, and social networks, and to analyze empirical data to help us understand the management of risks and development of resilience over the life course.
We will use a life course framework, guided by the assumptions that life is longitudinal and multifaceted, lives are linked, and lives are lived in local, regional, societal and historical contexts. Stages of the life course will be taken into account with focus on various transitions, including the transition to adulthood in early life, transition to lone parenthood or to detachment from family.
We will analyze longitudinal data from the Survey of Labour and Income Dynamics and from three General Social Surveys (GSS) on Time Use, on Family Histories, and on Social Support, which have been conducted a number of times over the past 20 years. These data will allow us to examine trends in risks and resilience over time, paying particular attention to the more recent surveys including the 2007 GSS on Family, Social Support, and Retirement, the 2008 GSS on Social Networks, the 2010 GSS on Time-stress and Well-being, and the 2011 GSS on Families expected to be released for analysis by Statistics Canada in the next two to three years.
Social Sciences and Humanities Research Council (SSHRC) 2010-2014 grant of $74,000 for the project: "Vulnerable families and individuals: Risk and resilience over the adult life course." (Investigators: Roderic Beaujot, Zenaida Ravanera and Jianye Liu)
Changes in Canadian families from the 1960s, often collectively referred to as the second demographic transition, have increased the risk to poverty of certain groups of families and individuals. For many young adults, the transition to adulthood has been made more difficult by the need to pursue longer years of schooling in order to successfully integrate into a more competitive labour market. Increases in the levels of divorce and in common-law unions that include children have made lone parents, especially women, vulnerable to low income. Family dissolutions have left more men and women unattached, exposing them to greater economic, social and health risks.
This program of research aims at understanding the trends in risks brought about by family changes and proposes to examine the factors that allow individuals to overcome their disadvantages. While adaptation to adversities greatly depends on individual characteristics (Patterson, 2002), resilience is often facilitated by support also from families and from social networks, the main factors that will be considered in our analysis, along with identified characteristics of individuals (Luthar et al., 2000:552).
Our research will be empirical and based on data sets gathered nationally by Statistics Canada, specifically, the Survey of Labour and Income Dynamics, a longitudinal survey, and the General Social Surveys on families, on time use, on social support and on social networks. Using more recent data collected by these surveys and the 2006 Census, the first component of our research program builds on our previous studies and updates the trends over time on levels of vulnerabilities of the various groups of interest (namely, children, youth, lone parents, and unattached individuals).
The second component focuses on factors contributing to the resilience in specific vulnerable groups at adult stages of the life course: in early adulthood – young adults in low social class as they make their transition to adulthood through leaving the parental home, completion of schooling , entry into the labour force, and entry into relationship and parenthood; in mid adulthood – lone parents as they raise their children and balance their family and work life; and in later adult life – men and women as they cope with detachment from family.
In terms of policy concerns, the research will address issues with regard to alternative priorities of social support to individuals and families, the institutional arrangements associated with work-life balance, and the family and work contexts of social differentiation, including the situation of children.
1. Second Demographic Transition: Changes in families and increased vulnerabilities
The current research program, like our previous studies, places our work in the theoretical context of the Second Demographic Transition (Beaujot and Ravanera, 2008; Ravanera, Beaujot and Liu, 2009; Beaujot, 2000; Beaujot, 2011). We have paid particular attention to the changes in earning and caring as central to this family change, along with the associated diversity across families and the associated new forms of risk and inequality (Beaujot, 2010, Ravanera and Rajulton, 2007). As demographers, we have also focused our attention to the childbearing side of family change (Beaujot and Wang, 2010; Ravanera and Beaujot, 2010; Beaujot and Muhammad, 2006) and the impact on children (Beaujot and Ravanera, 2009, Beaujot, Ravanera and Du, 2010).
Demographers have largely theorized family change in terms of two demographic transitions: a long-term change (from about 1870 to 1950), which brought smaller families; and another change (from about 1960 to the present), which especially involved increased flexibility in marital relationships (Lesthaeghe, 1995; Lesthaeghe, 2010; Beaujot, 2000: 85-96; Beaujot and Ravanera, 2008). The broader explanation of the transitions involves both structural and economic questions (macro-level structural changes and micro-level economic calculus) and cultural questions (attitudes and value orientations).
The first transition involved a change in the economic costs and benefits of children, along with a cultural environment that made it more appropriate to control family size. The second demographic transition has been linked to secularization and the growing importance of individual autonomy. This includes a weakening of the norms against divorce, pre-marital sex, cohabitation and voluntary childlessness. Value change has promoted individual rights along with less regulation of the private lives of individuals by the larger community. There is a heightened sense that both women and men should make their own choices in terms of relationships and childbearing. Diversity is valued, in living arrangements and in family forms.
As indicators of this family change, the annual divorces per 100,000 married couples increased from 600 in 1971, to over 1100 in the period 1981-2006, common-law couples amounted to less than 1% of all couples in 1976, compared to 18.6% in 2006, births to non-married women increased from 9.0% in 1971 to 37.7% of births in 2006, and lone parent families as a proportion of all families with children increased from 11.4% in 1961 to 25.8% in 2006 (Beaujot, 2010). The median age at first marriage increased from 21 years for brides and 23 for grooms in 1961-71, to 27 and 29 years respectively in 2004. Similarly, the age at women’s first birth increased from a median of 22.8 years in 1961, to 28.1 in 2006.
(a) On young adults: Assortative mating has increased selection into marriage based on education, with highly educated women tending to marry highly educated men. This pattern of selectivity by socio-economic status implies that persons who make transitions early can be relatively disadvantaged. Focusing on women born between 1922 and 1980 in the 2001 Canadian General Social Survey, Ravanera and Rajulton (2006) find women with high social status are more likely to have delayed their entry into motherhood, having first completed post-secondary education. In contrast, women with low social status are more likely to become mothers at a younger age, often without first completing post-secondary education or having a period of regular full-time work. These authors also find that the 10% who do get married at a young age are more likely to have fathers with less education (Ravanera and Rajulton, 2007).
Later home leaving can bring more transfer of resources from parents to children. Later entry into relationships, and especially later childbearing, enables young people to better handle the trade-offs between investing in themselves and investing in reproduction. Drolet (2002) finds that the wages of women who had their children later did not differ from those who had no children, but women who had their children earlier than the average for their level of education had lower average wages.
(b) On lone parents and their children: The contrasting patterns of family formation have led to polarizing patterns over the life course. On the one hand, there are persons who marry and have children at a young age, without having the training necessary for proper establishment in the labour market, and who are also more likely to experience family disruption and lone parenthood (Ravanera and Rajulton, 2007: 62). In contrast are those who complete their education, marry and have children later in the context of dual earner families that are more stable. Other research indicates that more educated mothers and fathers spend more time in child care, giving their children further advantages (Gauthier et al., 2004; Sayer et al., 2004). Women who have more education and other resources are more likely to be in “shared roles” relationships, and thus their children are more likely to benefit from father’s involvement (Ravanera et al., 2009).
In terms of family instability, Kennedy and Thomson (2010) find that, in Sweden, educational differentials in family instability were small in the 1970s, but have since increased due to the rising union disruption among less-educated parents. Consequently, children in more advantaged families experience less lone parenthood and family instability. Similarly, in Canada, Bohnert (2010) finds that employment difficulties are associated with increased relative risks of union dissolution, while home ownership has the opposite effect.
(c) Long term effect of family changes: Extending the analysis over generations, Kiernan (2002) has proposed the concept of “the long arm of demography.” That is, early transitions can mean low human capital investments from parents and society, making for vulnerability to lone-parenthood and “fragile families” in the next generation. Le Bourdais and Marcil-Gratton (1998) found that, in Canada, young people who had experienced their parent’s separation were more likely to enter cohabiting relationships early, less likely to have a direct marriage, more likely to give birth before age 20, and more likely to experience union dissolution (see also Hofferth and Goldscheider, 2010; Lappegard et al., 2009). Likewise, Bignami-Van Assche and Adjiwanou (2009) find that girls who experienced their parent’s separation as children were more likely to also experience earlier sexual activity in comparison to children from intact families.
In summarizing how children are faring in the second demographic transition, McLanahan (2004) uses the concept of “diverging destinies”. For instance, Tash and her colleagues (2010) find that, for parents who are not married, the father’s involvement declines sharply after the end of the relationship, or the mother’s transition to a new romantic relationship. Even in Quebec, where the majority of children are now born in cohabiting relationships, these relationships are less stable, and there is less father involvement after the relationship ends, in comparison to children born to married parents (Le Bourdais and Lapierre-Adamcyk, 2004). Father involvement is important to children, as documented by the fact that men’s interactions with children are more likely to promote physical activity, risk taking and independence (Doucet, 2009).
(d) On unattached adults at later life: At ages 45 and over, there has been an increase in the proportion of the population who are considered unattached since they are not living in a family situation. In many instances, this is brought about by family dissolution through separation and divorce or not having entered into relationships in early adulthood. At ages before public pensions are available, a significant proportion of these unattached are experiencing poverty. Between 1976 and 2007, the low income rate for unattached seniors has declined to 13.0% for men and 14.3% for women, but for the non-elderly unattached it remains 29.7% for men and 35.1% for women (Beaujot, 2010). This population has not been extensively analyzed, but non-attachment to families probably overlaps for men who have lost their employment and women who have had poor labour market connection when they were in breadwinner families.
2. Resilience: Families and individuals adjusting to adversities
Many studies on changes on family life brought about by the second demographic transition, including our own studies, have focused on aggregate differences by categories, including social class and family structures (Liu et al., 2010; Kerr and Beaujot, 2003; Ravanera, 1995; Ravanera and Rajulton, 2006). However, there are variations within categories, examination of which could be instructive and could point to finding ways by which individuals and families overcome adversities. As Furstenburg and colleagues (1999:5) note, studies focusing on “success under unfavourable conditions can provide useful clues about which public policies could be adopted to assist children and families to cope with these conditions, thereby reducing the handicaps imposed on those economically disadvantaged at birth.”
In a review of the literature, Patterson (2002:350) notes that the concept of resilience “emerged primarily from studies of children who functioned competently despite exposure to adversity when psychopathology was expected.” From psychology, studies using the concept have since expanded to other disciplines, including the disciplines of public health, sociology, and family science. Common to the various frameworks that guide the study of resilience is the emphasis of multiple levels of influence, namely the level of the community, the family, and the individual (Luthar, et al., 2000; Saint-Jacques, 2009). An ecological perspective points to interaction between the individual (developing child) and the contexts (such as culture, neighborhood, and family) wherein he/she is embedded (Bronfenbrenner, 1977; Cicchetti and Lynch, 1993).
We intend to examine resilience in adulthood using concepts that have been applied mostly in studies focused on children. As Luthar, Cicchetti, and Becker (2000: 555) state, “resilience can be achieved at any point in the life cycle, and there is a need for additional work on at-risk individuals’ achievement of positive outcomes in later life”. We intend to examine hypotheses specific to the influences of individual agency, familial support, and social networks, and to analyze empirical data to help us understand the management of risks and development of resilience over the adult life course. As in the study of Furstenberg and colleagues (1999), we will attempt to examine the immediate environment (including their family and social networks) that distinguishes individuals who succeed in spite of adversities.
3. Implications for policies
In the past, family policy followed the breadwinner model with a focus on men’s family wage and associated pension and health benefits, along with widowhood and orphanhood provisions in the case of the premature death of breadwinners. That is, the focus of family policy was the loss of a breadwinner and supporting the elderly who were beyond working ages. The challenge of current policy is to accommodate children who receive less parental investments, young lone parents who have difficulty coping with both the earning and caring functions, the disadvantages faced by couples where neither has secure employment, and the difficulties of unattached persons at older labour force ages who have limited employment potential. While there has clearly been a decline in the proportion of the population who have low income status, this has especially benefitted the elderly, and there are new forms of inequality across individuals and families (Zyblock, 1996; Beaujot et al., 2010, Beaujot, 2010; Myles et al., 2007; see also Richards, 2010). If one thinks of policy as promoting individual self-sufficiency and supporting dependents through their families, how is social policy to operate when families are more diverse and less stable? What policies would promote resilience among those with greater vulnerabilities brought about by changes in families?
C. Previous Studies
The proposed program of research builds mainly upon two previous SSHRC standard research grants dealing with families:
1. Family and work: Models of earning and caring, 2007-2010 (Beaujot and Ravanera).
This research explored the key changes affecting the family and work domains and the interplay between them, building on the work done for Earning and Caring in Canadian Families (Beaujot, 2000) and on Toward an SDC (Social Development Canada) Family Research Framework (Beaujot, Ravanera, and Burch, 2005), the lead paper in a SDC Expert Roundtable on Challenges for Canadian Families. Studies from this grant were based on a series of Canadian General Social Surveys on time-use and on family, including the following:
Beaujot, Roderic. 2011. Families. Pp. 228-256 in J. White, T. Hewitt and J. Teevan (editors), Introduction to Sociology: A Canadian focus Tenth Edition. Toronto: Pearson.
Beaujot, Roderic. 2010. Change in earning and caring: Implications for family diversity and inequality. Paper to be presented at conference on Family as a value with respect to religion, tradition and modernity, Antalya, Turkey, 26-27 November 2010.
Beaujot, Roderic and Zenaida Ravanera. 2008. Family change and implications for family solidarity and social cohesion. Canadian Studies in Population 35(1): 73-101.
Beaujot, Roderic and Zenaida Ravanera. 2009. Family models for earning and caring: Implications for child care and for family policy. Canadian Studies in Population 36(1-2): 145-166.
Beaujot, Roderic, Zenaida Ravanera and Ching Du. 2010. Child care: Preferences and opportunity costs. Paper presented at the Statistics Canada Socio-Economic Conference, Ottawa, 26-27 April 2010.
Beaujot, Roderic and Juyan Wang. 2010. Low fertility in Canada: The Nordic model in Quebec and the U.S. model in Alberta. Canadian Studies in Population 37(3-4):411-443.
Ravanera, Zenaida R. and Roderic Beaujot. 2010. Childlessness and socio-economic characteristics: What does the Canadian 2006 General Social Survey tell us? in Lorne Tepperman (ed) Reading Sociology. 2012 Oxford University Press.
Ravanera, Zenaida R. and Roderic Beaujot. 2010. Childlessness and Economic Self-sufficiency. Paper presented at the meetings of the Canadian Sociological Association, Montreal, June 2010.
Ravanera, Zenaida R. and Roderic Beaujot. 2010. Childlessness of Men in Canada’s Second Demographic Transition. Paper presented at the Annual Meeting of the Canadian Population Society, Montreal, June 1-June 3. (Submitted for publication in Canadian Studies in Population)
Ravanera, Zenaida R., Roderic Beaujot and Jianye Liu. 2009. Models of earning and caring: Determinants of the division of work. Canadian Review of Sociology 46(4): 319-337.
Ravanera, Zenaida and Roderic Beaujot. 2009. Life course and structural factors in childlessness: The waiting game and constrained choices in the Second Demographic Transition. Paper presented at the 26th Conference of the International Union for the Scientific Study of Population, Marrakesh, September 2009, and at the 2010 Statistics Canada Socio-Economic Conference, Gatineau, April 26-27.
2. Gender, Interpersonal Risk and Childbearing, 2002-2005 (Beaujot, David Hall and Ravanera).
This program of research analyzed child-bearing in the context of the family lives and life goals of women and men. Childbearing is viewed in the context of a life course where adults have a variety of objectives with associated trade-offs, and in the context of risks that are envisioned for the various desired and undesired outcomes. This research produced a number of studies, one of which is a collaboration of two of the research team members: Beaujot, Roderic and Jianye Liu. 2005. Models of Time Use in Paid and Unpaid Work. Journal of Family Issues 26(7).
Other grants of research team that produced studies relevant to the proposed research program:
Human Resources and Skills Development Canada on Earning and Caring, 2006-2008 (coordinated by Rod Beaujot)
Beaujot, Roderic, Zenaida R. Ravanera, and Jianye Liu. 2008. Models of earning and caring: Trends, determinants and implications. Report submitted to Human Resources and Skills Development Canada.
Beaujot, Roderic, Zenaida R. Ravanera, and Jianye Liu. 2009. Models of Earning and Caring: Trends, Determinants, and Implications. Population Change and Lifecourse Strategic Knowledge Cluster Research Brief No. 2. (Note: This research brief appeared in several popular media, including the 125th Anniversary edition of Good Housekeeping, May 2010. See www.sociology.uwo.ca/cluster)
Rajulton, Fernando, Thomas K. Burch, and Zenaida Ravanera. 2008. Influence of Opportunity Structures on Family Formation. Report submitted to Human Resources and Skills Development Canada.
Ravanera, Zenaida, Rajulton Fernando, Roderic Beaujot.2008. Family Structures and Social Capital. Report submitted to Human Resources and Skills Development Canada.
Social Development Canada, 2006:
Beaujot, Roderic, Zenaida R. Ravanera, and Thomas K. Burch. 2006. Toward an SDC (Social Development Canada) Family Research Framework. Paper prepared for Social Development Canada Expert Roundtable on Challenges for Canadian Families, Ottawa, 1-2 December 2005.
SSHRC-CURA Project on Changing Fatherhood: Supporting Involvement, 2003-2008 (Kerry Daly, Guelph University as Principal Investigator):
Ravanera, Zenaida R. and John Hoffman. 2010. Canadian Fathers: Demographic and Socio-Economic Profiles from Census and National Surveys. Chapter 2 in Kerry Daly and Jessica Ball (eds) Engaging Fathers in Social Change: Lessons from Canada (tentative title). Under review for publication by UBC Press.
Ravanera, Zenaida. 2008. Profiles of Fathers in Canada. Report prepared for Changing Fatherhood: Supporting Involvement, a SSHRC-CURA project of the Father Involvement Research Alliance.
Department of Canadian Heritage, 2009:
Ravanera, Zenaida R. and Roderic Beaujot. 2009. Synthesis Report on Canadian Youth: A Focus on Minority Youth. Report prepared for Multiculturalism and Human Rights Program, Department of Canadian Heritage. April 2009.
SSHRC Strategic Research Grant on Family Transformation and Social Cohesion, under the Program Social Cohesion in Globalizing Era, 2000-2004 (final product):
McQuillan, Kevin and Zenaida R. Ravanera (eds) 2006. Canada’s Changing Families: Implications for Individuals and Society. Toronto: University of Toronto Press.
Data and Methods
The research program will first examine the trend in the vulnerable groups subject to life course risk. Since resilience of children has been extensively studied, we will focus on adults, namely, young adults, lone parents, and unattached persons at older labour force ages. This will be done especially on the basis of data from censuses 1981-2006 and the General Social Survey on Families (1990, 1995, and 2001, 2006 and 2011) to determine the extent of vulnerability, the trend in the depth of disadvantage and the relative importance of each group in the total disadvantaged population.
Secondly, focusing on the groups at risk, statistical analysis (including appropriate longitudinal, multi-level and multivariate techniques of analysis) will be used to determine the factors responsible for higher or lower amount of actual vulnerability of given individuals within the group. This analysis will use the longitudinal Survey of Labour and Income Dynamics, focusing on vulnerable groups at various stages of the life course, roughly categorized by age groups: 18-29, 30-44, and 45-64.
The Survey of Labour and Income Dynamics (SLID) is a longitudinal panel survey conducted by Statistics Canada since 1993. It aims at understanding the economic wellbeing of Canadians and collects cross-sectional as well as retrospective and prospective data on a variety of transitions, durations, and occurrences relating to finances, work and family. SLID collects information from a panel of respondents over a six-year period, with a new panel selected every three years. Analysis of these longitudinal survey data will be done at the Research Data Centre at the University of Western Ontario.
In addition to the SLID and the GSS on families, we will supplement our analysis with data from recent General Social Surveys: the 2007 GSS on Family, Social Support, and Retirement – that would be most useful for understanding unattached individuals aged 45-64, the 2008 GSS on Social Networks – that would be most useful for understanding the roles of relatives and friends in dealing with adversities for all three vulnerable groups, and the 2010 GSS on Time-stress and Well-being – that would allow analysis of time spent with families, relatives, and friends and its relation to well-being.
An important consideration in our research is the measurement of outcomes and of resilience. Levels of income would be a major measure of outcome. We will, however, consider other outcomes appropriate for particular life course stage. Thus, at early life, where level of income may not be a good outcome measure for young adults who have not settled in regular work or careers, benchmarks of transition to adulthood will be used, including completion of post-secondary education, entry into regular work, and family formation through cohabitation, marriage, and parenthood. Similarly, at later ages, we will consider health status as a possible outcome measure, in addition to levels of income. For lone parents, in addition to level of income, we will explore variables related to their children’s outcomes, and to time spent with children.
Indicators of resilience are seen here as tied with the measures of outcome. We will explore ways of assessing resilience, based on the experience of previous studies. One possibility would be to measure “better than expected outcomes”, with expected outcomes derived from regression analysis that includes the variable of interest, for example, categories of social status (low, middle, high) in the case of analysis of young adults. Individuals with low social status whose actual outcomes are better than the derived expected outcomes would be considered resilient.