Volume 51, Part 1, 1997

Author Alaka Malwade Basu
The 'politicisation' of fertility to achieve non-demographic objectives

It is argued in this paper that 'perceptions' about the determinants of fertility and of fertility decline can be 'politicised' by various special interest groups: that is, these perceptions can be used to push for policies and interventions which often have an ambiguous relationship to actual fertility, but are important because they already appear on the political agenda of these groups.

Such politicisation is facilitated by the near-universal consensus that fertility decline is a legitimate goal in the developing world, by the increasing evidence that there can be no grand theory of fertility decline, and by the willingness of scholars to attach a policy significance to all their findings. Two examples of such politicisation in India are presented, one of which has a socially beneficial impact, whilst the other is potentially disruptive, to illustrate that such politicisation is not without its dangers.

Author Jan M Hoem
Educational gradients in divorce risks in Sweden in recent decades

Many investigators have found that divorce risks decrease as you move from groups with little educational or social capital to groups with more. This negative educational gradient fits with the notion that people with more education are better at selecting spouses and better at making a marriage work. Other investigators have found a positive gradient, often in populations where the situation is dominated by the individual's ability to handle the divorce process and to cope with the economic and other problems that follow in the wake of a divorce. The sign of the educational gradient in divorce risks seems to depend on the balance between countervailing influences.

Information about the gradient over a few educational levels is about as much as you can expect to get from the interview data of a normal-sized general survey. With access to the data from a full-coverage system of the population and educational registers of a sizeable population like that of Sweden, educational effects can be studied in much greater detail. We begin to tap this source in the present paper. When we do, the educational gradient in divorce risks turns out to be too slippery a basis for the general theories that have been developed around it so far, at least in a population where it is reasonably easy to get a divorce and where the hurtful consequences to the divorcees are more limited than elsewhere.

There has been no uniform relation between education level and divorce risk of Swedish women at the various educational levels during the 1970s and 1980s; developments in recent decades in Swedish first-marriage divorce risks have been much more favourable to the more highly educated than to women with less education, and the result is that the educational gradient has become negative as we leave the 1980s. The educational gradient changed correspondingly between cohorts born in the mid-1940s and cohorts from the mid-1960s.

In a society such as Sweden, it may be more important to explain the trends in divorce risks by educational level than to explain the gradient of educational effects. Most normal-sized data sets are too small to permit the inclusion of secular changes in the effect of education on divorce rates, so analysts risk working with a seriously mis-specified model if real educational impacts change over time.

Author Omondi-Odhiambo
Men's participation in family planning decisions in Kenya

From its inception in 1967 to the mid 1980s, the Kenyan national family planning programme suffered from a lack of popular support and confidence within the general population, absence of active local participation at all levels and, above all, the absence of men's involvement in this patriarchal nation. This study measures the effects of men's participation in family planning decisions, and identifies the conditions which would stimulate greater participation by men in family planning decisions. The principal conclusions are that Kenyan men do participate in these decisions, take an interest in planning their families, support family planning and use contraception to achieve their goals.

The recent transition to lower fertility is probably due, at least in part, to changes in men's attitudes. In particular, the study shows that lack of communication between husband and wife may be a more important obstacle to the adoption of contraception than men's opposition. Couple communication has the strongest positive influence on current contraceptive use followed by: residence in regions of the country in which conformity to traditional reproductive practices is weaker (Nairobi, Central, and Eastern regions), employment in higher-status occupations, the number of living children, higher levels of education, and the wife's current age.

Giving higher priority to the inclusion of men's needs and concerns in the design of family planning programmes should improve their success.

Authors Kathleen E Kiernan and John Hobcraft
Parental divorce during childhood: age at first intercourse, partnership and parenthood

It is well established that young people whose parents divorced or experienced marital breakdown during their childhood are likely to enter into first partnerships and into parenthood earlier than those whose parents remained married. In this paper, using data from the British National Sexual Attitudes and Lifestyles Survey, we examine how far the timing of first coitus plays a role in the genesis of this demographic behaviour for children of divorced parents. Other factors, including the timing of menarche, attitudes to sexual activity, degree of parental strictness and religiosity, were also examined. In general, these factors had little explanatory power. The analysis showed that earlier sexual activity for men and women from disrupted families is an important proximate determinant of their earlier entry into partnerships and parenthood, compared with those brought up with both natural parents.

Authors Abbas Bhuiya and Mushtaque Chowdury
Title The effect of divorce on child survival in a rural area of Bangladesh

The data for this study come from Matlab, a rural area of Bangladesh, where a continuous registration of demographic events has been maintained by the Internal Centre for Diarrhoeal Disease Research, Bangladesh since 1966. A total of 11,951 first marriages of Muslims that took place in the area between 1975 and 1987 were followed until the end of 1989, to examine the relationship between parental marriage breakdown and survival of first live-born children.

The impact of divorce on survival of children during infancy and childhood was examined, using hazard analysis. Other independent variables included age of mother at birth, and mother's education, year of birth, sex of children, and residence at the time of childbirth. It is shown that the net odds of death among children of divorced mothers in infancy and childhood were respectively 3.2 and 1.4 times higher than those of mothers whose marriages continued.

The paper also discusses the possible mechanisms which link divorce and child survival.

Author Warren C Robinson
The economic theory of fertility over three decades

After a promising start some three decades ago, the application of micro-economic analysis to fertility studies has proved disappointing. It has not led to an increased understanding of fertility decisions nor to the policy insights which had been expected. This paper considers the reasons for this disappointment. It reviews briefly the development of the now dominant version of the economic approach to fertility analysis, the so-called 'Chicago Model'. It concludes that several basic conceptual and theoretical weaknesses of this approach have led it up a blind alley. The paper concludes with suggestions for new assumptions and approaches which may make the theory more relevant for policy programmes.

Authors Alice Goldstein, Guo Zhigang, and Sidney Goldstein
The relation of migration to changing household headship patterns in China 1982-1987

Radical changes in fertility, economic structure, and level of development occurred in China between 1982 and 1987. Nonetheless, during this period family size remained relatively stable because the decline in household size due to lower fertility was offset by an increase in the number of adults. A major explanatory factor has been the government's changing migration policies which led first to family fission and then to fusion. Migration and household composition data from the 1982 census of China and the 1987 sample survey show that during spousal separation women often assumed the headship of their household, and in many instances retained it after the return of spouse.

Since this pattern is most pronounced in cities, we suggest that women's headship is related to changing norms that engender greater acceptance of equality between the sexes. It also reflects pragmatic recognition that these women have developed their own important networks for the efficient operation of their household. With the dramatic rise in migration in China resulting from the economic reforms, household size and headship patterns are likely to continue to be affected by spousal separation.

Authors M Khlat, M Deeb and Y Courbage
Fertility levels and differentials in Beirut during wartime: an indirect estimation based on maternity registers

In this paper, total fertility estimates for Greater Beirut in the mid-eighties and early nineties are presented, and changes in socio-religious differentials of fertility across time are explored.

The baseline information was recorded from registration details for all maternities in Beirut and its inner suburbs in 1984 and 1991: age of mother, number of children ever-born, hospital class, and religion of newborn. An indirect method was used to estimate total fertility from the joint distribution of mothers by age and parity, and using hospital class as a proxy for social class, differentials in fertility were investigated by Poisson regression.

The estimates of total fertility for Beirut shifted 2.60 in 1984 to 2.52 in 1991, and were higher for Muslims than for Christians in the two periods. The regression analysis showed that: (1) the difference between the two religious groups persisted after control for social class, and in fact applied to the lower social class; (2) fertility dropped between the two dates in the lower social class, and more so for Muslims than for Christians. In comparison with other countries of the region, the decline in Beirut was found to be relatively modest.

If the trends assessed in this study were to continue, the religious-based fertility differentials would taper off progressively in the capital city of Lebanon.

Authors François Légaré, and Robert Bourbeau
Mortality in Quebec during the nineteenth century: from the state to the cities

The aim of this paper is to explore mortality in Quebec during the nineteenth century from a demographic perspective. During the nineteenth century, there was excess urban mortality in various countries; in order to identify such mortality differentials, we compared mortality indicators for the province of Quebec and then for the urban areas of Montreal and Quebec City. Using data from various studies, we produced life tables and compared life expectancies.

We show that at different times during the nineteenth century, spatial variations in mortality levels across the province of Quebec and its urban areas were significant. According to the data we analysed, mortality is undoubtedly higher in urban areas even though a convergence in trends took place towards the end of the century, resulting in an overall reduction in mortality. Also, exploring life expectancies within a cohort approach at times of fast-changing mortality patterns has proved to be instructive.

Life expectancy estimates based on a cross-sectional approach were systematically lower than those resulting from a cohort-specific one. Trends diverged to a greater extent beginning with the 1870 cohort, reflecting improvements made from that point on to World War II. Since current mortality levels are substantially determined by the cumulative effects of past behaviour specific to each generation, it is quite obvious that mortality analysis will reveal its true meaning only with the help of cohort life tables.