Abstracts

Age-Uniformity in Early Life Transitions: What Does the 1995 GSS Tell Us?
Zenaida R. Ravanera, Fernando Rajulton, and Thomas K. Burch, Sociology, University of Western Ontario

Are the ages at experiencing early life course events more homogeneous now than in the past? For Canada, Marshall and McPherson (1994) seem to think so: "The life course is becoming more structured over time in that there is much less variability than there once was around the average years at which these events occur" (p.9). This was observed for the United States too. In an oft-cited study on social change and transitions to adulthood, Modell, Furstenberg, and Hershberg (1976) found that compared to their forebears, modern youth experienced greater uniformity and more routinization in the transition to adulthood. The ages at experiencing school completion, home-leaving, or marriage were concentrated at narrower age ranges.

Perceptions regarding age uniformity (particularly of early life events) may need revision in the light of more recent data. In this paper, we argue that this trend has been reversed in recent years. We examine the age ranges at which Canadian men and women experience early life course events. While there are certain differences by gender and by type of events, the general trend over cohorts (born from 1916-20 to 1961-65) is toward uniformity until around the 1941-45 cohort, but later cohorts experience a reversal of the trend. The analysis uses data from the 1995 General Social Survey of the Family. Six early life course events are examined: school completion, start of regular work, home-leaving, first union, first marriage, and first birth. We make use of life tables from which inter-quartile ranges are derived and used as indicator of age uniformity. Differences by gender for each event and for the processes for 'taking off' and 'settling down' are also explored.

Childbearing in Co-habitating Unions and Union Stability
Wendy Manning, Sociology, Bowling Green State University

Cohabitation provides a two-parent family context outside of marriage to have and raise children. Yet we know little about the conditions under which cohabitors decide to have children and the implications of childbearing on the stability of cohabiting unions. Drawing on data from the 1995 National Survey of Family Growth it appears that almost one third of women had a completed pregnancy and one-fifth of women gave birth during their cohabiting union. These levels of childbearing are notable given the relatively short duration of cohabiting unions. Childbearing within cohabitation is much more common for Hispanic women than White or African American women. Furthermore, children born during cohabitation to Hispanic women are less likely to be unplanned than those to women of other race and ethnic groups. The effect of these children on the stability of cohabiting unions is not the same as their effect on the stability of marriages. Almost three-fifths of cohabiting unions are still intact five years after the birth of a child. Net of other factors children born during cohabiting unions do not speed up or slow down the timing of the end of the relationship. Instead, conceiving a child during cohabitation tends to decrease the hazard of dissolution via an increased hazard of marriage. Generally, conceiving a child during cohabitation has similar effects on the stability of cohabitation for race and ethnic subgroups. The longer-term effects of children on the stability of subsequent marriages are also considered.

Contraceptive Choice in Canada
Zheng Wu and Kelly Martin Sociology, University of Victoria

Contraceptive choice, as an important aspect of reproductive health, has two primary goals: to meet individuals' needs for fertility regulation, and to provide protection from sexually transmitted diseases (STDs). While the 1984 Canadian Fertility Survey (CFS) provides an extensive account of women's contraceptive use in Canada in the mid 1980s, we know virtually nothing about contraceptive use among Canadians in the 1990s. Using data from the 1995 General Social Survey (GSS-95) and a reproductive health perspective, this study updates our knowledge of contraceptive use in Canada in the 1990s. We examine contraceptive choice among women and men in all marital statuses, focusing particularly on condom use in unmarried and non-cohabiting populations. Our results suggest that there has been an increase in condom use but a reduction in pill use among adult Canadians. Contraceptive use, particularly condom use, is associated with individuals' age, marital status, fertility and fertility intentions, education, religion, and church attendance. The implications of these results are discussed in the context of the recent fertility decline and the spread of STDs including HIV/AIDS.

The Current Health Status of Ghanaians
Kwame A. Boadu, Sociology, University of Alberta

From the mid-1970s to the early 1980s, a combination of economic mismanagement and internal and external shocks contributed to a deepening crisis in the Ghanaian economy. As a result, Ghana embarked on an economic recovery program in 1984, commonly referred to as structural adjustment policy. This policy, according to its major sponsors, the International Monetary Fund (IMF) and the World Bank is aimed at promoting accelerated or sustainable development. However, several studies have provided differing dimensions of the impact of structural adjustment policy on the Ghanaian population in general, and on certain segments. Overall, it has been established that economic reforms have caused significant structural changes in the population, which factor invariably has resulted in health status differentials. This study therefore proposes to investigate the current health status of Ghanaians based on present socioeconomic conditions.

Determinants of Divorce Across the Duration of Marriage
Teresa Abada, Sociology, University of Alberta

It is well-established that the probability of divorce varies over the marital life course. The probability of divorce is known to be lowest in the first two years of marriage, increases up to the 5th-7th year of marriage and decreases thereafter. In the later years of marriage, marital dissolutions are more often attributed to mortality rather than divorce. In effect, marriages may serve different functions over time; hence the determinants of divorce should also vary over the duration of marriage. This study focuses on the social and economic factors, including female labour force participation, male unemployment, geographic mobility and the presence of children and their risks to divorce across the duration of marriage for Canada, 1991.

Determinants of Early Birth Intervals in Bangladesh
Haider Rashid Mannan, Sociology, University of Western Ontario

This paper provides a strong empirical evidence for the association between some selected socio-economic factors and subsequent birth intervals in Bangladesh using the most recently conducted national survey data (Demographic and Health Survey 96-97). The methodological strength of this study is that it tackles the problems of both censoring (by using life table technique) and selectivity. The study also utilizes Cox's proportional hazards model to find the independent effect of each selected factor on the timing of subsequent birth. The results of hazards analysis show that rural women have shorter birth intervals than urban women with the differential increasing with parity. Women's education has significant independent effect on the timing of subsequent births. Women's Current work status, current age and age at the start of interval also have significant effects on the birth intervals studied. The survival status of preceding child is a very strong determinant of the timing of second and higher births. The study finally draws a number of very important policy implications.

Determinants of Health Care Utilization Patterns in Ontario
Margaret L. DeWit, Southwest Region Health Information Partnership, London, Ontario

Research in the area of health care utilization suggests a variety of factors which may impact on the level and nature of assistance received from service providers such as general practitioners, specialists and hospitals. These indicators may correspond to either the individual patient, health care provider, or to features of the health care delivery system itself. All, in turn, may have an important influence on the health status of a population. A previous study by Burch et al. (1991) ("Variations in Healthcare Service Use in Ontario: An Analysis of Associated Factors") employed data from the 1985 General Social Survey in order to examine factors associated with variations in health care service use in Ontario. Multiple regression and path analytic techniques were used to study predictors of contact with a number of separate health service providers, including visits to general practitioners; visits to specialists; visits to dentists; consultations with nurses; and nights spent in hospital, nursing or convalescence homes. The results of these analyses indicated that, at least in a quantitative sense, the system of access to health care seems to be working reasonably well since the existence of health problems is found to be a key predictor of service use as are contacts with general practitioners and medical specialists. Using data from the 1996-97 Ontario Health Survey, the present study will update and expand on this earlier work by investigating the dynamics of health care provision and receipt with path analytic models corresponding to specific sources of health care, including general practitioners, specialists, dentists, nurses, hospital stays or nights spent in convalescent or nursing homes. Special attention will be paid to regional variations in patterns of utilization as well as to differentials according to age and sex. Analyses of these data will help to establish not only current levels of service utilization but also determinants and predictors of the intensity of use of specific types of health care. For purposes of establishing a baseline for change in utilization patterns of the Ontario population, these results will also be compared against findings from the earlier Ontario Health Survey (1990-91).

The Determinants of Immigrant Acculturation
Gustave Goldman, Housing, Family and Social Statistics Division, Statistics Canada

Given current demographic trends of population growth in Canada, immigration will play an increasingly important role in the continuing development of Canadian society. Sociological and psychological studies have shown that patterns of adaptation and acculturation vary depending on the specific socio-cultural characteristics of the migrants. The research presented in this paper builds on existing demographic, sociological and psychological work by analysing the determinants of immigrant adaptation and acculturation, taking into account the ethnic diversity of the migrants. Data from the census of population and from administrative sources such as the longitudinal immigration database maintained by Statistics Canada are used to derive an index of acculturation and to develop analytical models that show the relative contribution of selected socio-cultural and economic characteristics to the outcome of the acculturation process. The results of this research will serve to inform the public policy and decision making process with respect to immigrant adaptation and acculturation.

Determinants of Infant Mortality in Nepal
Juhee Suwal, Sociology, University of Alberta

Infant mortality is a sensitive human development indicator of a nation. Infant mortality has reached a stable low rate in developed countries while it is still high and on a slow decline in developing countries. There are many factors which contribute to the incidence of a high or low level of infant mortality. Although credit for contributing to the lowering of infant mortality has been given to health programs by public health personnel and to the improvement in socio-economic status by social scientists, in a traditional and agricultural country like Nepal, both these factors are found to influence infant mortality. Data on infant mortality obtained by the 1991 Demographic Health Survey of Nepal are analysed in this study. A logistic regression model is used for analysing the data. Several hypotheses are set up to explain the incidence of infant mortality in Nepal. All the hypotheses are supported. The various reasons for the persistence of high infant mortality and the difficulties in lowering it are discussed. The findings suggest that among all the variables analysed in the study, parity, residence, immunization, and ethnicity influence infant mortality the most.

Direction - North: Employment Related Inflows of Americans for Temporary Residence in Canada
Margaret Michalowski, Demography Division, Statistics Canada

Temporary migration and its causes are part of a broader mobility context involving movements and exchanges of capital, goods and information. Although temporary migration is not a new form of geographical mobility, the labour component of temporary migration is constantly evolving in response to increased intensity of international trade and movements of capital accompanied by more restrictive immigration policies of the major receiving countries.

Canada and the United States are prime exemplars of the changing nature of temporary migration flows. Within the context of bilateral exchanges between Canada and the United States, the major contributing factors which have influenced the development of temporary flows are: the Free Trade Agreement (FTA) provision for facilitating the migration of technical and managerial personnel in connection with cross-border direct investment introduced in 1989, and a long tradition of post-secondary education exchanges. Although migration was deliberately excluded from the negotiations of this agreement, the ratification of the North American Free Trade Agreement (NAFTA) in 1994 is expected to further intensify labour migration between North American countries.

A motivating reason for examining temporary flows derives from recent studies that demonstrate that the increased use of temporary visas for international personnel movement and the labour force dynamics of knowledge-based economic activities reflect the realities of the contemporary global marketplace (Keely, 1998; Fellegi, 1997). Moreover, the changes to this movement are not of temporarily character and will continue being a part of overall mobility. It is important, therefore, to assess not only the size of the movement but the human capital, which it represents as well.

The objective of this paper is to assess inflows of Americans to Canada for temporary residence related to employment, and to study the importance of its highly skilled component. The main sources of data for this study are the 1996 and 1991 Censuses of Canada and administrative data sources (the Citizenship and Immigration Canada files used to create a comprehensive database on the temporary authorisations' holders). Based on the latter, estimates of occupational structure of American workers in Canada were produced for the 1985 to 1996 period. The analysis of these estimates points to the significant changes in the occupational structure in general, and specifically to the important shifts among professionals. An added dimension of the analysis is the specific meaning of the notion of "temporarily". The census data demonstrate that although the majority of Americans residing temporarily in Canada in 1991 stated that their place of residence five years ago was the United States (55%), the large proportion of them was already living in Canada (40%). Five years later, in 1996, the proportion of those Americans who have lived in Canada for at least 5 years decreased but was still high, as one out of 3 of them was found in this situation.

Ethnic Variations in Voluntary Association Memberships in Canada
Fernando Mata, Multiculturalism, Department of Canadian Heritage

This paper will focus on the memberships in voluntary associations reported by Canadians of different immigrant status and ethnic origins in Canada. The database used for the analysis comprises about 17,000 respondents to the 1997 National Survey of Giving, Volunteering and Participating conducted by Statistics Canada. Patterns of membership in six types of voluntary associations will be examined: religious, cultural, fraternal, sports related, community and political type of organizations. Using correspondence analysis as an exploratory technique the analysis will move further into logistic regression to investigate the net effects of ethnicity and other covariates such as age, gender, SES, immigrant status and length of residence.

Factors Influencing the Migration of Registered Indians Between On- and Off-Reserve Location sin Canada, 1966-1991
Stewart Clatworthy with Martin J. Cooke

Using data collected by the 1991 Aboriginal Peoples Survey and the 1991 Census of Canada, this study explores several aspects of the migration patterns of Registered Indians during the 1986-1991 time period. In addition to descriptive analyses, which document the characteristics of migrants and in-, out- and net-migration flows for various geographical areas (including provinces, north-south regions, reserves, and rural and urban off-reserve locations), statistical models have been developed to explore the contribution of selected factors to Registered Indian migration flows between on- and off-reserve locations. Factors considered in the models include various personal attributes, as well as characteristics of migrant origin and destination communities.

Fertility Projections for Canada, Provinces and Territories, 1998-2031
Shirley Loh, Ravi Verma and M. V. George, Demography Division, Statistics Canada

This paper discusses the development of the fertility assumptions for the 1998-based population projections for Canada, provinces and territories, 1998 to 2031, produced by Statistics Canada. After a slight increase in the fertility rates during the late 1980s, the total fertility rate fluctuated around 1.70 children per woman between 1990 and 1995. Since then, the fertility rate has plunged from 1.67 in 1995 to 1.62 children per woman in 1996. This is a drop of 3% in just one year, after the fertility rate had remained fairly stable around 1.70 for the preceding five years. A similar drop has occurred at the provincial level as well. Does this rapid decline in fertility suggest a further downward trend? From the point of view of fertility projections, how low can fertility go? Will it stabilize, at the present level, or move upward? To address these questions, three fertility assumptions together with their rationales have been developed, taking into consideration the experience of other industrialized countries, previous Canadian fertility levels, changing socioeconomic factors affecting Canadian fertility behaviour, and studies on fertility intentions. Assumptions for the total fertility rates were first developed at the national level, then the projected values of the TFRs for the provinces and territories were derived, using an index method based on the ratios of observed TFRs between the provincial and national levels. An analysis of the trends of these provincial/national ratios over the past two decades, and their use in projecting total fertility rates for the provinces and territories, is also presented.

Health Insurance Coverage and Utiliztion Demographics in the United States
Rongli Chen & K.V. Rao

Abstract unavailable at this time.

High Fertility Trends And Maternal Health In India, 1981-1994
Ravi R. Sinha, Population Research Center, Arun K. Sinha, Department of Statistics, Patna University, & Ravi B. P. Verma, Statistics Canada

The objectives of this paper are: (1) to analyze the changes in the total fertility rates over the period 1981-1994, (2) to correlate the 1993 total fertility rates with the maternal health care indicators, i.e. life expectancy in the reproductive ages (15-49) by gender, antenatal care, place of delivery, assistance during delivery, etc., and (3) to suggest ways of improving the situation in five Indian states, Assam, Bihar, Uttar Pradesh, Madhya Pradesh and Rajasthan. The data on fertility indicators are derived from the Sample Registration Bulletin ,Vol. 32, No. 1, Registrar General, India, January 1998. The maternal health indicators are extracted from the National Family Health Survey conducted in 1992-1993, and the Indian Life Tables, 1988-92; We also use the evidence that has emerged from surveys carried out in Bihar about the promoting family planning and MCH (Maternal and Child Health) Care through Dairy Co-operatives (sponsored by the Population Council, New York), Operations research on Spacing Methods (sponsored by MOHFW, Govt. Of India, New Delhi), Integrated Nutrition and Health Programme (Sponsored by CARE-India, New Delhi). The Population Research Center, Department of Statistics, Patna University during the 1992-1996 period conducted these surveys. In this paper, the fertility and maternal health indicators by state are compared with the national figures. Rates and proportions of events are computed. The relative differences between the province and national figures are analyzed to support the negative consequences of high fertility on maternal health indicators.

Fertility in India has declined substantially, but it is still higher than the replacement level of 2.1 children per woman. Though the total fertility rate (TFR) of 4.5 children per woman in 1981 declined to 3.5 in 1994, its states vary widely with respect to decline in fertility. While the reduction is much higher than the overall national average in the southern and western states, the states like Assam, Bihar, Uttar Pradesh, Madhya Pradesh and Rajasthan present entirely a different scenario of very slow decline. In these five states, the relative differences in TFRs over the national figure have also increased between 1981 and 1994. Several studies also report that the prevailing high fertility causes poorer maternal health and higher mortality.

How to Count: A Comprehensive Methodology for Counting Unaccompanied Youth
Laurie Goldman, Sociology, Carleton University

Historically the homeless population was believed to be homogeneous in composition. Today this group is comprised of people from all walks of life - male and female, young and old, sick and healthy, single individuals and families who find themselves on the streets and/or in shelters. Of these subgroups, unaccompanied youth are becoming an increasingly visible faction of the homeless population (e.g. squeegee kids). Services are required to help these youth before they become homeless adults. But how can services be designed without an accurate count of the target population and some information about their characteristics? Current data on the extent of homelessness suffers from underestimation to overestimation, depending on the purpose of the research. Using a combination of enumeration techniques, such as a street count and a shelter count, and the capture-recapture method of estimation which has been proven to work for hard to measure animal populations, an accurate count of the unaccompanied youth population of the Ottawa region will be attainable. As well, it will be possible to collect some basic demographic characteristics about this segment of the homeless population through this process.

Impact of Family and Socio-demographic Factors on Patterns of Leaving the Parental Home
Pascale Beaupré and Céline Le Bourdais, Centre interuniversitaire d'études démographiques Institut national de la recherche scientifique/Université de Montréal

As with other developed countries, rates of leaving home in Canada rose until the 1980s before starting to progressively decline. Using the retrospective data collected in the General Social Survey carried out by Statistics Canada in 1995, our study first aims to document the extent to which children's age at departure from the parental home has varied across generations. We then use proportional hazards models to examine the influence that family and socio-demographic factors exercise on the propensity of children to leave home early. Particular attention is given to the effect of family disruption experienced in childhood and to the effect of the changes observed in the patterns of entry into the labour market and into conjugal life.

Life Course Trajectories Before and After Retirement
Fernando Rajulton and Zenaida R. Ravanera, Sociology, University of Western Ontario

This study analyzes the timing of retirement from work and the trajectories leading to and following retirement. Trajectories before and after retirement will mainly consider events in late life, namely, home-leaving of last child (or empty nesting), activity limitation in old age, widowhood, and return to work after retirement. We use data provided through the Canadian 1995 General Social Survey of the Family, and apply various event history techniques of analysis.

Initial findings from life tables reveal that men's average age at retirement has decreased over birth cohorts. This is expected and in line with several studies showing that Canadians desire retirement before age 65 and that the trend is towards early retirement. Analysis of trajectories indicates some interesting changes that have taken place over cohorts. In addition to analysis of changes over cohorts, we examine whether or not the timing of retirement and sequences of transitions vary across social classes, regions, and different cultures.

Lone-Father Families in Canada
Kevin McQuillan, Sociology, University of Western Ontario with Marilyn Belle

Lone-father families were, until recently, a demographic oddity, and were virtually ignored in analyses of changing family structures. With recent changes in patterns of marriage, cohabitation, and divorce, however, lone-father families have begun to attract attention from sociologists and demographers. This paper reviews the changing dimensions of lone-father families in Canada, focussing largely on the period from 1971-1996. The paper charts the changing nature of these families and briefly discusses some of the policy issues that are raised by the growth of this form of family living.

Population Estimates from Remotely Sensed Data
Jerry Wicks, David Swanson, Robert Vicent & José Luis Pereira de Almeida

Abstract unavailable at this time.

Regional Analysis of Health Status in Canada
Srimanta Mohanty, Sociology, University of Western Ontario

Health is an important aspect of the quality of human life. Regional imbalances in development occur in most societies. This study is an example of comparing health status across the provinces of Canada. In this study five variables are investigated, which reflect the various dimensions of health status. These variables are aggregate in nature. Using the taxonomic method, the variations among the provinces are examined. The analysis reveals that Ontario has the highest scores and Quebec has the lowest scores on the health development index. The results also show the aspects of health status that are not satisfactory in Canada relate especially to problems in the health care system. The research methodology involves a novel statistical application. The method facilitates the ranking, classification and comparison of the provinces by levels of health status. This study is useful in identifying indicators of spatial imbalances in health status with a view to setting up targets in allocating scarce resources. It is hoped that this study will provide new direction for health status in Canada.

Registered Indian Migration: An Analysis of Five-Year and One-year Mobility and Migration Data from the 1996 Census
Mary Jane Norris, Department of Indian and Northern Affairs

This paper explores the mobility and migration patterns of Canada's Registered Indians using data from the 1996 Census. Mobility status and internal migration of registered Indians are examined by age and gender, with a focus on migration to and from reserves. The 1996 Census contains two questions on mobility/migration: one is based on place of residence five years ago, the second is on place of residence one year ago. While the five-year question has been a standard question for many censuses, the one-year question was first introduced in the 1991 Census but was limited to place of residence at the provincial level. For the first time in 1996, place of residence one year ago includes subprovincial geography, i.e. at the Census Subdivision level. This is particularly relevant to Registered Indian migration since it permits an analysis of migration to and from reserves based on a one-year period, in addition to the standard five-year period. Flows of migrants over each migration period (1991-1996 and 1995-1996) are examined by origin-destination using the following origin-destination categories: on-reserve; off reserve- rural; off-reserve urban Census Metropolitan Areas (CMAs); and off-reserve urban non-CMAs.

Self-Destruction in the new Land: Immigrant Suicide in Canada During 1990-92
Frank Trovato, Sociology, University of Alberta

In this paper, differentials in suicide among immigrants in Canada during the three-year period surrounding the 1991 census are investigated. Deaths are classified by type of method employed to commit self-destruction (solids/liquids, gases/vapours, hanging/strangulation, firearm/explosives, other means) to better assess the possible underlying dynamics of self-imposed death. A conceptual model of immigrant mortality is developed which considers the risk of suicide as a function of four sets of causal agents: culture of origin of the immigrants; aspects related to immigrants' experience in the host society; group differences in demographic composition; and migration selectivity effects. From this framework, five predictions are specified for empirical valuation: (1) immigrant suicide risk in Canada is directly linked to the suicide rate of the home nations; (2) the likelihood of self-death among foreigners is a function of their duration of residence in Canada; (3) variations in ethnic community integration account for differences in suicide risk (4) the chance of self-inflicted death is inversely related to socioeconomic status; and (5) new wave immigrant groups (i.e., mostly immigrants from third world regions) will share low suicide propensities independent of all other factors in the multivariate model.

A Study of Poverty of Immigrants in Canada
S. S. Halli and A. Kazemipur, Sociology, University of Manitoba

The most recent UN report on human development revealed that Canada, while still the most desirable country to live in, has not had an admirable record in dealing with poverty. What composes the basis of UN report is the overall rate of poverty in different countries. The domestic distribution of poverty, however, is another issue of concern. The poverty trends in Canada during 1990s showed that poverty tends to concentrate socially among certain groups, e.g., women, single parents, disabled, young adults, etc. Another potentially vulnerable, yet largely ignored, group in this regard is immigrants. The present paper will show such trends in different Canadian cities. A statistical model is also developed to explain the dynamics of poverty-generation trends among immigrants.

Wage Income of Immigrants in Canada: Spatial Differentials Within and between Ethnic Groups
Stephen Obeng-M. Gyimah, Sociology, University of Western Ontario

The changing ethnic diversification of the immigrant population in Canada has tended to draw considerable interest in socio-demographic research, most notably, on socio-economic differentials and inequalities among ethnic groups. What seems notably lacking, however, are the spatial variations within and between ethnic groups with respect to the Census Metropolitan Area (CMA) of residence. While most studies suggest that visible minorities earn relatively low wages than non-visible minorities after controlling for other factors, the literature is silent on the relative spatial differentials that might exist within and between these broad categories in the major CMAs.

Using the 1991 Public Use Sample Tapes, the primary objective of this paper is to examine the relative economic position of selected immigrant groups (visible and non-visible) in a multivariate context and compare such performances within and across major CMAs.