Academic Achievement and Ethnicity in Secondary Schools: A Longitudinal Study.
Mme Hanh Hoang Tran , Education, Université de Montréal

Migration is part of the contribution to the growth of the population of Que´bec. This phenomenon could be best observed in the demographic profile of the student enrollment.

In Que´bec, the Charter of the French language (bill 101 in 1977) requested the immigrants to enroll their children in French schools, unless one of the parents or one of the children in the family, is educated in English. These last decades, many studies from Canada and from several countries of the G7, pointed to the young immigrants as the sources of influence of the academic achievement in schools. How much this thesis could be confirmed or rejected?

The present longitudinal study uses the file of the Montreal school board. All students of secondary I are followed during 5 years, the normal time to finish high school in Que´bec. Five cohorts are taken into account to get a significant results, in order to confirm or reject the thesis. The sample has more than 5000 students for the first year, of whom 3500 will graduate in the fifth year.

Three dependent variables are observed: **teaching language (English or French), mathematics, second language (English or French) and 7 independent variables are considered: ** age, sex, mother tongue, place of birth, family status, father's schooling, mother's schooling. S.A.S is used for the multiple regression of these variables. Part 1 of this paper covers the theories of academic achievement and the effects of some variables; part 2 discusses the sample and the methodology; part 3 analyzes the results of the research. Tables and graphs are attached and finally the conclusion.

Note: The subject is part of the author's Ph.D. thesis to be submitted at the end of 1999.

Awakening to the Intergenerational Crisis: A Content Analysis of Media Pension Coverage in Canada 1993-1998
David Foot, University of Toronto and Rosemary Venne, University of Saskatchewan

With declining fertility rates and increasing life expectancy, Canada, like many industrialized countries, is facing an aging population. Specifically, the growth in elderly dependents is a significant concern for countries with pay-as-you-go public pension systems (where income from today's workers is recycled to current seniors). This is especially a concern in North America which experienced a prolonged post-WWII fertility boom followed by below-replacement level fertility rates. When this bulge of boomers attain retirement age (in the 2010s and 2020s) public pensions plans are predicted to run dry in Canada as the pension system attempts to accommodate the demographic bulge of the boomers.

Conflict over the issue of intergenerational equity (referring to equality in treatment and opportunities for different generations) was noticed in the US media over the 1980s and early 1990s but in Canada attention to this issue was virtually nil. This research analyzes public discussion of the intergenerational equity issue from 1993 to 1998 through a media content analysis with special attention to the looming pension crisis in Canada. The issue of intergenerational equity seems to have surfaced in Canada during the mid 1990s with the realization that the Canada Pension Plan might "run dry" sometime in the 2010s under current contribution rates. Though the realization resulted in some policy responses (e.g., effective January 1998 Finance Minister Martin increased contributions and slightly decreased payouts to assure that the CPP system does not "run dry"), nonetheless intergenerational equity issues continued to prevail in the popular media. Documentation of the emergence of this issue will focus on public pensions, demography and business realities.

Childbearing After Age 30
Lindy MacNeill, & Zheng Wu, Sociology, University of Victoria

The growing trend to delay motherhood until after age 30 is well documented in Canada. Women have long continued to have children late into their reproductive years, but many women are now beginning their reproductive careers when most women were finishing them in the past. Fertility theories suggest that women are now more able, and willing, to postpone motherhood to focus on education, employment and personal development and to ensure the quality of life of their children. But little is known about the socioeconomic and cultural factors that influence women to animate this trend. Using data from the 1995 General Social Survey and survival model techniques, we examine childbearing behaviour among Canadian women who remain childless at age 30. We estimate the probability that women who were still childless at age 30 would ever have a child. We also examine the effects of several individual-level characteristics, such as employment activity, schooling, and marital history, on the hazard rate of first births to women who were childless at age 30.

Consequences of Childhood Residential Mobility: Findings from the 1986 Canadian General Social Survey
Darcy Hango

Childhood events and circumstances can have a long-term impact on socioeconomic status. Factors that influence the status attainment of the next generation, or the intergenerational transfer of resources, is an area of considerable concern. An important indicator of socioeconomic status is the completion of a postsecondary degree or diploma, or at the very least, the completion of a high school diploma. Therefore, the factors that reduce one's chances of completing high school or postsecondary education are particularly salient.

The socioeconomic status of parents is a significant factor predicting status attainment in the next generation. Generally, the more financial capital available to parents, the better the prospects for children. Intergenerational transfers, however, are not confined to monetary rewards; the human capital possessed by parents can also benefit the child in the long run. Children of parents with high human capital have a greater chance of obtaining a significant amount of education. However, the positive effects of human and financial capital may not be as significant if an important relationship does not exist between parents and children. In other words, the 'social capital' expended on children by their parents is necessary for financial and human capital to be successful determinants for human capital accumulation in the next generation. Therefore, the factors that enhance or impede the relationship between parents and children, and between the family and others in the community, are an important area of study.

A significant life event that involves many families and their children is residential mobility. A family move can be seen as a major life course event in the lives of children with long-term consequences exhibited on educational attainment. The loss of friendships and the stress that may accompany family moves could potentially undermine the positive economic effects of the family move. This paper examines the effect of childhood residential mobility on educational attainment on the basis of the 1986 Canadian General Social Survey. The social capital perspective suggests that children's educational attainment is negatively affected in families that move because of the loss of important peer and community relationships. Conversely, the human capital perspective suggests that families move to increase their resources, thus attenuating the negative impact of the loss of important relationships.

The vast majority of studies support the social capital perspective, suggesting that children's educational attainment is negatively influenced by residential mobility. However, the present results do not support the social capital perspective, since the children who experienced residential mobility had an increased probability of high school and postsecondary completion. In addition, a cohort analysis further suggests that the effects of residential mobility on the formation of social capital may have declined over time. These results are an anomaly in the literature and the limitations of the data as a possible reason is discussed.

Delays and Rapid changes in Quebec fertility: 1860-1996.
Rod Beaujot, Sociology, University of Western Ontario

The delay in the first demographic transition can be linked to delays in socio-economic and socio-cultural transformations in Quebec. In comparison to Ontario, this delay is particularly marked around 1921 when the fertility is 5.3 in Quebec and 3.2 in Ontario. The subsequent changes are stronger in Quebec, where the baby boom was less important, such that by 1960 the two provinces arrive at the same figure of 3.8 births per woman. The changes of the 1960s are particularly noteworthy because they represent the culmination of the first demographic transition as well as the beginning of the second, bringing Quebec fertility from 3.76 in 1960 to 1.97 in 1970. This second transition relates especially to marital relations which have become looser, as seen in the frequency of divorce and cohabitation. While cohabitation is especially prevalent in Quebec, a higher proportion of Ontario births occur after age 30. The fertility levels of Quebec and Ontario are very similar in 1996, at 1.56 births per woman.

Demographic Imperative and Informal Care to Seniors
Norah Keating, University of Alberta

Keating: Population aging has been viewed with alarm by policy makers who believe that we are ill-equipped to deal with the needs of Canadian seniors. The focus of this paper is to place seniors receipt of assistance and care within the context of other sectors of the population. Using data from the 1996 Statistics Canada General Social Survey on Social Support, we illustrate that seniors are no more likely to receive assistance than are those in other age groups. Further, we show that among seniors, a small proportion is in need of care because of chronic health problems. We argue that policies for an aging population should be based on consideration of the resource needs of younger people, the broad set of tasks that enhance independence, the transitions from assistance to care, and the support that might reduce transitions from community to institutional care for seniors.

Demography, Legislation and Ethnic Mobility: Considerations and Implications for Projections of Aboriginal Populations
Mary Jane Norris, Dan Beavon and Stewart Clatworthy, Indian and Northern Affairs Canada

Well-defined and understood components of growth are critical in assessing not only the current, but also the future sizes and composition of any population. Today, the size and growth of Canada's various Aboriginal populations are not simply functions of the usual demographic components of growth (fertility, mortality and migration), but have become an increasingly complex interplay of these demographics with components of legislation (e.g. Revisions to the Indian Act) and ethnic mobility.

Ethnic mobility, which can be described as a change in the reporting of individual affiliations with an Aboriginal group over census periods, has been shown to be a significant component in past intercensal growth (Guimond, 1999), and could likely continue in the future. In developing projections of four major Aboriginal groups (registered Indian, non-status Indians, Me´tis and Inuit) (either Identity or Ancestry based) assumptions are not only required for the usual demographic components but must also be considered in relation to legislative impacts on population sizes, especially for registered and non-status Indian groups. Legislative components related to the 1985 amendments to the Indian Act affect population growth largely through reinstatements (largely related to intermarriage of registered Indians prior to 1985) and status inheritance (a function of intermarriage). Ethnic mobility (which can also be affected by intermarriage) can affect the growth of Aboriginal groups either positively or negatively- but is clearly a less well-defined factor for future population size and is not explicitly projected. Nevertheless as a component of growth ethnic mobility assumptions are implicit in that respondents are assumed to remain consistent in reporting their group affiliations from one census to another and from one generation to the next.

The impact of each component of growth, be it demographic, legislative or related to ethnic mobility, varies by Aboriginal group, such that some groups may be more or less affected by a particular component, or even not at all. For each Aboriginal group this paper considers each component and its implications for projections along four dimensions: measurement (direct or indirect) of the component; impact on growth (positive or negative; predictability or certainty of component; and, interplay with other components and Aboriginal groups. Recent developments affecting measurement, refinements (such as estimating fertility of registered Indians by reinstated and non-reinstated) and questions are discussed along these various dimensions. For example, what will be the future identity of those offspring of registered Indians, who are not eligible for registered status through intermarriage? Intermarriage and its effect on status inheritance have the potential to impact not just the registered Indian population, but also the size and composition of the non-status and perhaps even the Me´tis populations. In addition, the paper also looks at the aspect of projecting band membership of registered Indians within the complex interplay of demographics, legislation, band membership rules and intermarriage.

Note:Reinstatement refers to the restoration of Indian status to those persons who had lost status as a result of provisions in the earlier act, especially as the result of out-marriage of registered women to non-status men. Status inheritance refers to a set of descent rules that establish entitlement to Indian status at birth based on the extent of out-marriage between status and non-status (whether Aboriginal or non-Aboriginal) persons.

Effects of Maternal Age on Child Outcomes
B. Ram, N. Montsion, and M. Moore

Over the past two decades, Canada like most industrialized countries has witnessed two important and interrelated demographic shifts. First, fertility has fallen much below the level required for population to replace itself. Second, increasing proportions of women have been delaying having a child. Effects of these shifts on the lives of mothers have been studied extensively. However, their effects on the lives of children have not been fully examined, especially in Canada. In an earlier study, we examined the effects of the number of siblings and birth order on the intellectual, emotional, and behavioral traits of younger children.

This paper complements our earlier research, by examining the effects of maternal age on three outcome measures for children below age 12. These outcome measures are: MATH skills for children between 6 and 11 years, as measured by the Mathematics Computation Test; cognitive ability for children 4 to 5 years, as measured by the Peabody Picture Vocabulary Test (PPVT); and motor and social skills development (MSD) for children 3 years and younger. Maternal age which is our major independent variable, is measured by age of the mother when the child under study was born. The analysis is based on the first cycle of the National Longitudinal Surveys of Children and Youth (NLSCY) done by Statistics Canada in 1994-95.

Enhancing Analytical Credibility in Fertility Research
Susan McDaniel, Sociology, University of Alberta

Demographic analysis has been situated in, and socially constructed by, the societal and the economic rather than in and by human agency, either collective or individual. This may pose particularly perplexing challenges to explaining fertility as a social process. Indeed, the quest for explanation of fertility patterns and trends had expanded in multiple directions including, for example, the socio-cultural, the political and economic, the familial in global perspective, and the micro-economic social psychological, suggesting a sense of frustration with the explanatory capacities of existing demographic approaches.

This paper proposes the development and application of new conceptual and methodological approaches to the apprehension and analysis of fertility by reference to the concept of 'analytical credibility', a concept used judiciously by Anatole Romaniuc, to open and extend active engagement of people with a future that they make. By invigorating demographic analysis with agency, particularly the agency of women as childbearers, a reconciliation of agency (human and societal) with structural constraint and opportunity is enabled. Fertility research is enhanced not only in explanatory power, but in its capacity to understand what is happening and why, and how we as agents might act to make a future or our choosing.

Estimating the Fertility Level of Registered Indians in Canada: A Challenging Endeavor
Shirley Loh and M.V. George, Demography Division, Statistics Canada

The Indian Register maintained by the Department of Indian Affairs and Northern Development (DIAND) is the most comprehensive source of fertility data pertaining to Registered Indians. A reasonable time series for the births and population under question can be derived. However, it is well known that births registration in the Register suffers from two major limitations: late reporting and underreporting. It is of prime importance to properly adjust these data in order to estimate the fertility level of Registered Indians.

The purpose of this paper is to discuss the methodologies used to adjust time series data on births and procedures used to derive accurate estimates of fertility, namely total fertility rates. Estimating the fertility level of Registered Indians have become more complicated since the passing of the 1985 amendments to the Indian Act which includes a set of descent rules that establish entitlement to Indian status. The impact of the amendments on deriving accurate total fertility rates for Registered Indians will be addressed in the paper.

Ethnic Mobility and Demographic Growth of Aboriginal Populations in Canada, 1986-1996
Eric Guimond, Norbert Robitaille, Andrew J. Siggner, and Gustave Goldmann

The demographic growth of Aboriginal populations of Canada during the 1986-1996 period is exceptionally high. Observed growth rates exceed by far the theoretical maximum of 5,5% per annum for a closed population, which is basically the case of Aboriginal populations at the Canada level. Traditional components of demographic growth - fertility, mortality and migration - and population coverage issues can not account for all of the observed growth. The present component analysis shows that ethnic mobility, defined as changes in self-reporting of ethnicity, is an important contributor to the growth of aboriginal populations and that this phenomenon has manifested itself outside an Indian reserves, mostly in urban areas, because of legal considerations. Changes in self-reporting of ethnicity has also been observed in Aboriginal population of the United States, Australia and New Zealand.

Ethnic Youth
John Valentine

Much of Canada's character as a society in the next century will be determined by the social, economic and ethnic attributes of its younger age cohorts. Using the 1996 population Census as its data source, the purpose of this research is to provide selected graphic profiles of age cohorts comprising children and youth of different ethnic and racial backgrounds as well as places of residence in Canada.

In 1996, there were 9.7 million individuals under the age of 25 (34% of the total population of Canada). This population was made-up by four age-groups: young children under 6 years of age (2.3 Million), elementary school children aged 6-11 (2.4 Million), high school and college youth aged 12-18 (2.8 Million) and older youth aged 19-24 (2.3 Million). A slight majority ( 51% ) of them were females while 49% were males. According to the census, the majority of young age cohorts lived in Ontario (37%), Quebec (24%), British Columbia (12%) and Alberta (10%). Major concentrations of this population were found in Metropolitan Toronto (14%), Montreal (11%) and Vancouver (6%).

The Family and Immigrant Adjustment
Derrick Thomas, Housing, Family and Social Statistics, Statistics Canada

Background/Issues: The behavior of immigrants frequently makes sense only when viewed in the context of a family strategy. Migrating individuals face uncertainty, they often make sacrifices and many experience diminished social status in what they perceive are the long-term interests of their children and other family members. From the point of view of individuals, the entire objective of migration is not infrequently to be with or rejoin family members.

Family reunification has, in fact, been described as the "backbone of Canada's immigration program". Aside from the benefit conferred on Canadian residents who wish to be joined by their loved ones from abroad, family reunification is thought to represent good policy from the perspective of the wider public. Established family in Canada should help to ease the burden of adjustment for new immigrants and minimize social costs for all concerned. Families ostensibly represent a source of support as immigrants get settled, learn an official language, and upgrade their qualifications. The Federal government's family reunification program is even designed to ensure that a significant proportion of the potential costs of resettling Family Class immigrants is born by their relatives in Canada. Canadian residents who wish to sponsor the immigration of relatives from abroad are required to make a formal undertaking to support that relative free of public assistance for a period of 10 years.

The extent and duration to which new immigrants indeed live in economic families with existing residents is unknown. An even more critical question, from a policy point of view, is the degree to which such living arrangements mitigate against social costs in the form of economic dependency and low incomes. Important ancillary questions concern the nature of vanguard and retinue elements and the impact of different characteristics on outcomes.

Methodology: Beyond its advantages with respect to sample size, the Census of the Population is superior to other sources in several respects. While Census data does not contain information on the legal category of immigration, it does contain unparalleled information about actual living arrangements and the family composition. It also allows for an examination of outcomes over a relatively long time frame.

Broadly speaking, three types of economic families can be delineated in Census data.

  1. Families consisting of new immigrants and Canadian born relatives whom they have joined,
  2. Families consisting of new immigrants and immigrants who arrived at an earlier period,
  3. Families consisting of new immigrants who arrived together as a unit.

The approach will be to compare these three types of families for a cohort of new immigrants arrived at the same period. It will also be necessary to control for the, the age (average), sex (ratio), and level of education (highest & lowest among those out of school) of family members. The incidence of low income and dependence on transfers as well as overall family incomes will serve as outcome measures. The incidence of involvement in business and/or self-employment could also be an interesting outcome measure.

Among families consisting of immigrants who have arrived at different times further analysis will focus on the characteristics of vanguard elements. These characteristics could include family structure, duration in Canada (average if necessary), age, sex and education levels. The questions addressed might include: Is female lead migration more or less successful? Are the education levels of vanguard members important to the outcome? Is it better that families are long established before being joined by relatives? Interactions between the characteristics of the vanguard and those of the retinue will also be examined.

A number of statistical techniques might be employed in the analysis. A set of multivariate logistic regression models might be used to predict the probability of low income or dependence on transfers based on the family characteristics. Ordinary multivariate regression could be used to predict overall family income. The objective would be to isolate and quantify the independent effect of family living arraignments (e.g. migrating vanguard and retinue living together or family consisting only of migrants who arrived at one time).

Family Diversity and Children's Health and Well-Being
Teresa Abada, Sociology, University of Alberta

This paper examines the implications of different family structures on several aspects of children's health and well-being - including dimensions of cognitive functioning, social and emotional development, physical growth and health and academic achievement. It examines how family dynamics and parent-child relationships differ across different family environments and how they account for differences in children's development and well-being.

Are children in single parent families more likely to suffer from physical and mental health problems than children in two-parent families? Do children whose single parent has never married fare better or worse than children whose parents are divorced? To what extent does family income account for differences in parenting in children's outcomes? This paper also addresses the degree to which differences between children in lone-parent families and those in two-parent families are a function of income. Building on the first wave of the NLSCY, this paper examines the separate effects that family structure, family relationships and other factors such as poverty and social support have on children's health and well-being.

Family Structure and Remarriage in Alsace, 1750-187
Kevin McQuillan, Sociology, University of Western Ontario

This paper examines the influence of the number, age and gender of surviving children on the likelihood of remarriage for widows and widowers in Alsace in the period from 1750 to 1870. Using data from a family reconstitution study of five villages in the region, the paper employs a hazards regression model to assess the effects of family structure variables on remarriage while controlling for the influence of age at widowhood, the period in which the dissolution occurred, occupation, and religious affiliation. The findings suggest that the age-sex distribution of surviving children had different effects on the remarriage chances of men and women. For men, the number of dependent children significantly increased the likelihood of remarriage, while the presence of an older daughter in the household decreased the probability of remarriage. For women, teenaged sons do not appear to have played the role of substitute spouse, and having an older son in the household did not significantly decrease the probability of remarriage. However, the presence of a teenaged daughter did have a marked negative effect on the likelihood of widows remarrying.

First Nation/Band Affiliation of Canada's Aboriginal Population Off- Reserve: An Analysis and Interpretation of 1996 Census Data for Selected Census Metropolitan Areas
Stewart Clathworthy, Mary Jane Norris, Dan Beavon, and Paula Saunders

There are roughly 600 different Indian bands or First Nations in Canada. Although the majority of the populations associated with these First Nations reside on reserve, a sizable minority of the populations of many First Nations resides off reserve, most commonly in large metropolitan areas. Presently, little is known about the First Nation or band affiliation of the off-reserve populations residing in specific localities.

The 1996 Census contained a question on Band/First Nation membership and provides an opportunity to explore this aspect within the context of individual off-reserve communities. Using the 1996 Census, this paper presents data and analyses on the First Nation affiliations of Canada's Aboriginal populations living off reserve in six Census Metropolitan Areas (CMAs) or Census Agglomerations (CAs), including: Edmonton, Calgary, Regina, Saskatoon, Winnipeg and Thunder Bay. The composition and heterogeneity of a CMA/CA's First Nation population is examined in relation to a number of factors, including: the size of the particular First Nations' source populations (on & off reserve); the location and distance of First Nation reserves in relation to the CMA/CA; and recent (1991-1996) migration patterns between reserves and the CMA/CA.

In interpreting the CMA's First Nation composition, other considerations include the proximity of other (intervening) urban areas and their impact on First Nation settlement patterns, both recent and long-term, as well as accessibility and transportation (constraints and opportunities) between the reserve and CMA/CA.

A preliminary analysis of data for Winnipeg shows that the CMA has a highly heterogeneous First Nation population of some 18,000 (mainly registered Indian) persons, representing some 65 individual First Nations. The population affiliated with 10 First Nations, however, is particularly large and represents just over half of Winnipeg's total First Nation population. These "10" First Nations are characterized as having large source populations of at least 2000 persons and have their reserves located within 500 km of Winnipeg. Recent migration from these First Nations, however, does not appear to be a major contributing factor to size of the First Nations population in Winnipeg. Also, this preliminary look at the data suggests that the greater the distance a First Nation's reserve is from a given CMA/CA, the smaller its share of population in that CMA. However, further insight into the roles of accessibility, the proximity of (and access to) other urban areas is required in order to better understand the factors contributing to the size and composition of Winnipeg's First Nation population.

By looking at the First Nation composition of a number of CMA/CAs, this paper will provide a comparative perspective and hopefully, contribute to a better understanding of the dynamics in the size and growth of First Nation populations in some of Canada's metropolitan areas.

Housing Alberta's Seniors
Alison Yacyshyn, Sociology, The University of Western Ontario

Housing of Alberta's seniors aged 55+ is addressed on the basis of population projections and data from the 1991 and 1996 censuses, including collective dwelling data. The fields of demography, ageing, and housing are relevant to understanding elderly housing. The availability of the data relevant to this area of focus is also assessed. Using applied demographic techniques, the results of the research contributes to an understanding of Alberta's aging population and to a projection of future housing usage. The research also has social policy implications. The projection results suggest that the majority of the elderly Albertans in 2016 will reside in single detached houses. Grants for home repair and the increasing demand for home care are relevant policy issues. Housing projections by type can address the expected needs of specified age groups across the later stages of the life course.

Immigration and Gender Segregation in Occupations
T.R. Balakrishnan and Stephen Obeng Gyimah, Sociology, University of Western Ontario

That immigrant women, especially women of color are disadvantaged in the labour market has been fairly well established by many studies. However, whether they do worse than men when education and occupational skills are controlled is less known. An earlier study by the authors have shown that occupational segregation among the immigrant groups, without regard to gender, has been decreasing over time. Occupational assimilation is found to be a function of duration of stay in Canada as well as official language facility. This paper examines sex differentials in occupational assimilation taking account of these factors.

Using public use sample tapes from the Canadian censuses the study will construct measures of occupational assimilation by gender for the period 1971-1996. Indices of dissimilarity will be used extensively as well as other ways of looking at integration. Comparisons will be made not only with native born women of European origins but with men, both immigrant and native born. It is hypothesized that because of differences in the occupational distribution of men and women in general, the modes of assimilation for women will be different from that of men. We will also examine whether the place of origin such as from Europe, Asia and Africa makes a difference in the occupational assimilation of immigrant men and women. It is hopes that this study will yield results that will be of interest for public policy.

Immigration, Ethnicity and Labour Force Outcomes in Canada: 1981-1996
Stephen Obeng Gyimah, Sociology, University of Western Ontario


Immigration continues to play a significant role in sustaining the population of Canada, contributing to about half the annual population growth in recent years. There has been a significant shift in the ethnic origin of immigrants to Canada over time. Whereas most immigrants before the 1970s came predominantly from Europe, the past three decades have witnessed a remarkable shift to developing country origins. Thus, a historical overview of the ethnic composition of the population of Canada shows an increasing diversification, particularly since the 1970s. This trend has drawn considerable interest in socio-demographic research, most notably on socio-economic differentials and inequalities among the ethnic groups. Many studies have focused on Canadian-born and foreign-born ethnic populations using indicators such as education, occupation and income (for example, Pendakur and Pendakur, 1998; Hou and Balakrishnan, 1996; Maxim, 1992, 1994; Shamai, 1992; Herberg, 1990; Li, 1988; Balakrishnan, 1988; Beaujot and Rappak, 1988; Beaujot et al. 1988; Kalbach, 1970; Kalbach and Richard, 1988; Richmond and Kalbach, 1980; Tandon,1978).

Almost invariably, these studies demonstrate the complex nature of the relationship between ethnicity and inequality. In occupations, Hou and Balakrishnan's study (1996) shows the probability of working in managerial and professional occupations is least among visible minorities , although the Chinese seem to be doing better than the other minority groups. With regard to wage income, multi-variate models show the disadvantaged position of minorities, especially Chinese and Blacks (Li, 1988). Our contention in this paper is that the causal connection between education and income operates through occupational achievement. The level of education determines the kind of occupation one has which in turn determines one's level of income. All things being equal, higher levels of educational attainment are accompanied by higher probabilities of being employed in higher status occupations (managerial and professional) leading to higher earnings. While this causal connection might be theoretically sound, we believe its empirical validity might not be tenable for all groups in Canada, especially immigrants. The relationship is more likely to hold for the Canadian born than the foreign born population although the latter group has a higher educational attainment. This derives from the fact that most foreign educational degrees are not recognized at similar levels as Canadian degrees (Pendakur and Pendakur,1998).

Recognizing the ethnic diversity of the Canadian population, we further hypothesize that visible minorities are more likely to be employed in occupations which do not commensurate with their higher levels of education, and thus receive less income for their higher educational achievements. To what extent does 'visibility' determine occupational and income status irrespective of the level of education? Our principal objective in this study is to answer this question by examining the causal connection in a historical context. While previous research provides considerable evidence that Southern Europeans (Portuguese, Greek, Italians) were greatly disadvantaged with respect to earnings and high status occupations compared to the charter groups (English and French) and Northern/Western Europeans, recent studies reveal that such disadvantages are declining and the differences are fast diminishing. Could a similar situation be happening to visible minorities? What is the relative magnitude of such decline among southern Europeans compared to visible minorities, if any? Are there differences by gender?

Data and Methods
The data for this study will come largely from the Public Use Micro Files (PUMF) of individuals from the Canadian censuses from 1981 to 1996. The PUMF is a 3 percent sample from the population enumerated in the censuses (about 700,000 cases in each data set) and provides detailed information at the individual level on a number of variables. On methodological considerations, the study will be restricted to individuals aged 20-60 years with non zero earnings and not enrolled as full time students at the time of the censuses. Some of the variables we will be looking at are ethnicity, immigration status and age at immigration, age, sex, wage earnings, level of education and occupational status, among others. In our multi-variate analyses, two models will be presented. A multi-nomial logistic model of occupational status (dependent variable) with three categories ( managerial, professional, and other occupations) will be used to examine the relationship among ethnicity, education, and occupational status after controlling for other factors. Also, a hierarchical MCA model of wage (log) earnings will be used to examine wage income differentials among ethnic groups in a multi-variate context.

Since all the variables we are interested in are available in each of the PUMF from 1981 to1996, we would be able to examine the causal connections among ethnicity, education and labor force outcomes overtime. While we expect the labor market disadvantages to be diminishing over time given the official rhetoric about equity issues in recent times, we further hypothesize that such disadvantages with the charter groups and western Europeans are fast diminishing among Southern Europeans than among visible minorities.

The Impact of Ethnic Mobility on Educational Levels of Aboriginal Peoples
Andrew Siggner, Statistics Canada

For the first time data from three censuses and the 1991 Aboriginal Peoples Survey have been organized in an historical manner with a consistent set of definitions and geographies. Having established that the Aboriginal population, both in terms of origin and identity, has grown rapidly and that an important component of that growth has come from personal changes in the ethnic affiliation, this finding begs the question as to what are the characteristics of the people changing their affiliation? If we do not know more about the group or groups who ethnically have moved into the Aboriginal identity population, how can we rely on the aggregate Aboriginal statistics? How can policy-makers and planners really know what impact their policies and programs are having on the target populations?

This paper will focus on the highest level of schooling among those reporting Aboriginal origins and Aboriginal identity and how this characteristic is changing over time, namely from 1986 to 1996. The ability to monitor these changes will allow policy makers and planners, particularly those within various types of Aboriginal governments, to know more about the composition and dynamics of the population's skill capacity, where it is lacking and what improvements may be needed. The study will demonstrate that ethnic mobility has an important impact on the schooling levels among Aboriginal peoples using an age-cohort analysis. This is followed by presenting a more descriptive analysis of the highest level of schooling data by regional distributions, place of residence, the population concentration of Aboriginal peoples in a certain types of geographic areas (a variable we call, "enclavity"), and by age and gender.

Influences of Community on the Sexual and Reproductive Behavior of Adolescents and Their Expression in Demographics Variables
Grisell Rodriguez Gomez

The community, as a level of individuals' social re-grouping, filters the influence of others levels or entities in which men group themselves also, in such way that human behavior and its expressions at social level are marked by the imprint of community dynamics. This report seeks to define those community's factors, of a psico?social nature, intrinsically linked to sexual and reproductive behavior and ruling the actions of the adolescent population of a peri?urban community of the city of Havana.

The author has carried out an inquiry of this subject using quantitative and qualitative techniques to measure some indicators of this phenomenon in a representative sample of this sub?population in the community mentioned above, to define some of these determinant community factors and with the study of them, to assess the possible nexus with fertility, mortality and migration, rescuing them as an expression of the impact of the community level on the population dynamics.

Interethnic Connections in Canada
Madeline Kalbach

Previous research demonstrates a trend toward increasing ethnic intermarriage both in the population as a whole and for most ethnic groups in Canada. In addition, husbands and wives tend to select spouses from the Charter groups or from those groups similar to themselves such as geography, culture or religion (Richard, 1991; Kalbach, 1999). This paper examines the interethnic connections of English, French, German, Italian, Indo-Chinese, Chinese and Black and Caribbean husbands and wives. It will compare their patterns of intermarriage and marital choices in the 5 major Prairie CMAs (Edmonton, Calgary, Regina, Saskatoon, and Winnipeg) and in Canadas three largest Census Metropolitan Areas (Toronto, Montreal and Vancouver). This research utilizes custom tabulations from the 1991 Census of Canada.

Mortality and survival in Cuba in the nineties
Juan Carlos Albizu-Campos Espiñeira

Although Cuba made a rapid process of demographic transition, in general, and of mortality, in particular, the seventies and the eighties were the testimony of the stagnation of the improvements of the capacity of survival of the population while the infant mortality continued to decline. That was the genesic agent of the current reductions of the life expectancy at birth in both sexes while the deep economic and social crisis that hit the Cuban society in present days constituted the factor of acceleration. What happened to that indicator and which were the determinant factors of the changes are the questions to be responded.

Multiple Levels of Analysis: Prospects and Challenges
Fernando Rajulton and Zenaida Ravanera, Sociology, University of Western Ontario

This relates to methodological challenges (and how to meet them) if and when we do the various studies planned for the 'Family Transformation and Social Cohesion' project. To meet the policy and information needs that the partners of the proposed project have articulated, we foresee that different levels of analysis will need to be done: individual level, group level, and combined or multi-level analysis. We intend to discuss theories and hypotheses, and the appropriate levels of analysis, data, and techniques.

Neighbourhood Ethnic Transition and Socioeconomic Status in Montreal, Toronto, and Vancouver
Feng Hou, Department, Institutition

The present study examined the relationship between changes in neighbourhood ethnic composition and socioeconomic status in Canada's three largest Census Metropolitan Areas. Analyses were based on 1986 and 1996 census tract profile data, and focussed on blacks, Chinese, and South Asians. The results indicated that the three CMAs had very few neighbourhoods that were comprised predominantly of visible minorities, but that the number of tracts with a strong presence of visible minority populations had increased rapidly. Proportions of blacks tended to increase rapidly in neighbourhoods with low SES, but rapid increases in black proportions did not necessarily lead to declines in neighbourhood SES. Chinese proportions tended to increase rapidly in neighbourhoods with high SES, but rapid increases in Chinese proportions did not necessarily result in improvement of neighbourhood SES. Among tracts that were over-represented by South Asians in 1986, those tracts that experienced rapid increases in South Asian proportions were associated with lower initial SES and declines in SES.

New Demographic Approach to Population Adjustment
R. Lachapelle and D. Kerr

The 1996 Canadian Census is adjusted for coverage error as estimated primarily through the Reverse Record Check (RRC). In this paper, we will show how there is a wealth of additional information from the 1996 Reverse Record Check of direct value of population estimation. Beyond its ability to estimate coverage error, it is possible to extend the RRC classification results to obtain an alternate estimate of demographic growth - potentially decomposed by component (including emigration and temporarily abroad). This added feature of the Reverse Record Check provides promise in the evaluation of estimated census coverage error as well as insight as to possible problems in the estimation of selected components in the population estimates program.

On Their Own Again: Sources of Income Inequality Among Unpartnered Divorced or Separated Women, 1985-1995
Charles L. Jones

This research examines the sources of inequalities in life style and total personal income among unpartnered women who are currently divorced or separated. Changes in divorce legislation increase the probability that women will pass through such a life stage and changes in related aspects of family law and social welfare policy assume that they can be self-supporting. Analysis of data from ten national surveys carried out in Canada between 1985 and 1995 clearly demonstrates that while the presence of dependent children is an undoubted handicap, formal educational qualifications, particularly the possession of a post-secondary diploma play a key role in the economic stratification of such women.

One Hundred Years of Aboriginal Demography: An Analysis within the Canadian Context
Mary Jane Norris, Eric Guimond and Paula Saunders, Statistics Canada

Canada's Aboriginal population is demographically distinct from Canada's mainstream population, being notably younger, with significantly higher fertility and mortality. This paper begins with the issue of defining Canada's Aboriginal populations, using both census concepts (based on ethnic origin (Aboriginal ancestry) and Aboriginal identity) and legal definitions of Registered Indians. In tracing the growth and size of Aboriginal populations data are used from Statistics Canada's censuses from 1871 to 1996, and from the Indian Register of the Department of Indian and Northern Affairs (DIAND). In providing historical trends in Registered Indian and Inuit fertility and mortality the paper makes reference to works in the literature by Romaniuc, George and Robitaille, and are also used to demonstrate the process of the demographic transition that these populations are undergoing.

This study also assesses the changing composition of Canada's overall Aboriginal population in terms of different Aboriginal groups and classifications such as Registered Indians, non-status Indians, North American Indians, Me´tis and Inuit, in relation to the concepts of Aboriginal origins/ancestry and Identity. The impact of legislative changes in shaping the composition of the Aboriginal population, such as revisions to the Indian Act and the effect of reinstatements and status inheritance are examined based on reports and studies by Clatworthy, and Statistics Canada's various Aboriginal and Registered Indian projections. As well, the impacts of both internal migration and reinstatements are also considered, in relation to on/off reserve and rural/urban distributions of different Aboriginal populations, based on studies by Clatworthy and Norris.

The Pure Relationship and Below-replacement fertility in Canada
David Hall, Department of Sociology, Nipissing University, North Bay, Ontario

Canadian society has been characterized by below-replacement fertility for over two decades. During this same time frame, rates of divorce and cohabitation have increased dramatically. Recent theories developed from the rational-choice perspective, and in the sociological theory of Anthony Giddens have both explored the possible link between changes in fertility and changes in intimate relationships. Specifically, while the rational-choice "uncertainty reduction hypothesis" holds that couples will use childbearing as a way of reducing uncertainty in their marriages; Giddens' writings on the "pure relationship" indicate that childbearing will be minimized or avoided altogether in the face of relationship uncertainty. Using data from the 1995 Canadian General Social Survey, this paper tests the contrasting hypotheses that flow from the above theories and discusses the broader implications for family demography in Canada.

Some Demographic Consequences of Revising the Definition of Old Age to Reflect Future Changes in Life Table Proabilities
Frank T. Denton and Byron G. Spencer, McMaster University

Denton/Spencer Sixty-five has long been used to define the beginning of 'old age'. Yet it is clear that that definition is arbitrary, and with continuing reductions in mortality and morbidity rates it will become increasingly inappropriate as time passes. In this paper we consider how the definition might be modified to reflect changes in life table probabilities, and how the future numbers and proportions in 'old age' would be affect

A Stochastic Population Forecast For Canada
Nan Li and Shripad Tuljapurkar, Mountain View Research

The uncertainty in the future change of population is significant. Traditional forecasts use the range of scenarios called high-medium-low to cover this uncertainty. How much uncertainty has been taken by this range, however, is yet a question. The method provided by Lee and Tuljapurkar (LT), which models the uncertainty in history and hence in forecasting systematically, answers this question. This paper uses the LT method to assess how much uncertainty has been taken by the official forecast of Canada. It concludes that about 25% of uncertainty is covered by the fertility trajectory range, more than 95% of uncertainty is in the mortality trajectory range, and about 80% of uncertainty has been taken by the population scenario range.

Structural Adjustment Programs and Their Impact on Fertility Determinants
Reem Attieh, Sociology, University of Western Ontario

It is widely argued in the development literature that a high fertility rate is one of the major obstacles to development. While there may be a great deal of debate about the best route to induce fertility reduction, the improvement of women's status, which includes both economic and social, is believed to be a positive way in which fertility can be reduced. However, structural adjustment progams (SAP) that are imposed on "developing" countries by the International Monetary Fund (IMF), which are intended to improve the economic health of the recipient countries, have had a detrimental effect on women's status in the "developing countries". An examination of the literature reveals that these austerity measures, which include demands for reduced government social spending and increased taxes, have had a negative impact on some of the determinants of fertility, such as women's employment, and education. It seems that the SAPs are undermining the progression towards fertility decline, which in turn undermines the advance in development.

Theoretical and Analytical Advances in Longitudinal Research
Fernando Rajulton & Zenaida Ravanera, Sociology, University of Western Ontario

This will take off and use materials from the recent workshop that we organized here at the Population Studies Centre on longitudinal analysis for social science. We plan on discussing theoretical considerations, technical approaches, and software for longitudinal analysis.

Trends in Occupational Attainment and Mobility in Canada, 1920-1994
Richard A. Wanner, Sociology, University of Calgary

During the course of the twentieth century Canada has experienced a dramatic increase in both the average level of educational attainment and in the proportion of young people going on to postsecondary education. Has this increase in educational qualifications served to loosen the ascriptive links between origin status and destination occupational status among Canadian men and women? Using data from the 1973 Canadian Mobility Study and the 1986 and 1994 Statistics Canada General Social Surveys and a cohort design, I address this question using two modeling strategies. In the first, continuous measures of occupational status are combined with conventional OLS regression methods, which provide a single-parameter estimate of the linear trend the effect of father's occupational status on offspring's status net of education, mother tongue, and labour force experience. In the second, more recently developed multinomial conditional logit (MCL) models are applied. These combine the ability of loglinear methods to model flows across occupational categories with the ability of regression methods to include covariates. The OLS regression models show that the effect of socioeconomic origins, as indicated by father's occupational status, on son's occupational status declined considerably in Canada during the twentieth century, as did the effect of language for both men and women. However, the usual indicator of achievement, education, does not appear to have had an increasing effect on occupational status for either men or women. Results from the MCL models in most respects confirm the OLS regression results. In addition, they show that the declining effect of father's occupational status on son's occupational status is due to declining levels of immobility rather than an increased flow between occupational classes.

Union Transition and Health
Zhen Wu, & Randy Hart, Sociology, University of Victoria

Research has been consistent that married people enjoy better mental and physical health than those who are not married. There is also evidence that once a marriage has ended and partners have separated, many of the health benefits derived from marriage also dissolve. As nonmarital cohabitation has rapidly become a new form of Canadian family living, it raises the question as to whether the negative health effects associated with marital dissolution are also felt by people who experienced nonmarital union dissolution. Using longitudinal data from the 1994-1997 National Population Health Survey (NPHS), we investigate the impact of union transition on a range of health indicators (including physical, mental, and health risk factors). Our analysis focuses on the transition in and out of cohabitation. We anticipate that the entry into cohabitation and transition from cohabitation to marriage may enhance physical and mental health and inhibit health-risk behaviours, whereas the exit out of cohabitation via union separation may be detrimental to health and induce health-risk behaviours. Separate analyses were conducted for women and for men. The discussion of our findings is conceptualized by the changing marriage and cohabitation patterns in Canada.

The Use of POHEM in Assessing the Health Impacts of Earlier Diagnosis of Colorectal Cancer in Canada: An Evaluation
Edward Ng, C. Le Petit, W. Flanagan, J.M. Berthelot, Social and Economic Studies (SES), Statistics Canada

While colorectal cancer (CRC) is one of the leading causes of death due to malignancy in North America, the burden of CRC care can be reduced by early detection. Statistics Canada has developed a microsimulation model, the Population Health Model (POHEM), that recently incorporated the incidence, disease progression and outcome of CRC. This presentation discusses the CRC disease module within POHEM and provides estimates of the health impacts of an earlier diagnosis of CRC, as a proxy for a nationally coordinated screening effort. POHEM was used to estimate the health effects of a simple cancer stage-shift starting at age 50. The impact of such a stage-shift for a simulated cohort was a CRC-related mortality reduction of approximately 9% and 12% for females and males respectively. We evaluate the impact of age-sex structure on our results and find that when the age-sex specific rates were applied to the 1995 population, the mortality reductions were higher, especially among the seniors.

York Region Private Schools Prepare for the Double Cohort
Jim Jackson

Abstract not available.