Continuing fertility transitions in a plural society: ethnic trends and differentials in Peninsular Malaysia

Fertility in Peninsular Malaysia has declined continuously from the late 1950s, reaching a total fertility rate of 3735 in 1983. All ethnic groups in Malaysia have contributed to this modern demographic transition but the rate of change has been most rapid for Chinese and Indians, Malay fertility having reached a plateau in the early 1980s. The effect of age structure, marital patterns and marital fertility (by parity) on the fertility declines for each ethnic community are analyzed. There has been a tendency, in each ethnic group, for the age distribution within the group of reproductive-age women to grow younger, reflecting the entry into the younger reproductive ages of the large birth cohorts of the 1950s and early 1960s. The effect of this on crude birth rates is hard to determine, because rising age at marriage and increasing use of contraception meant that fertility was increasingly concentrated in the more central reproductive ages. By the 1990s, the earlier declines in fertility will bring about a decline in the proportion of the total population made up of females in the main reproductive ages. After that point, further declines in fertility will be reflected in a sharper decline in the crude birth rate and hence the rate of population increase. Between 1947 and 1980, the age at marriage changed dramatically for females of all ethnic groups. The transition to higher age at marriage for Chinese was completed earlier, and since 1970 has risen by only a year. For Malays and Indians, the rise began later, proceeded faster and continued right up to 1980 when the medium ages at 1st marriage were Malays 22, Indians 23, Chinese 24 years. In 1980, Malay women on average were marrying 5 years later, and Indian women 6 years later than had their mothers' generation in 1947. The proportion never-married among Malay and Indian women aged 20-24 rose from 1/10 to 1/2 over this period; relatively greater changes are evident at ages 25-29. Other factors are the almost complete shift from parent-arranged to self-arranged marriages. Family size desired has decreased for all groups and the decline in breastfeeding has been offset by the sharp increase in the practice of contraception. Continuation of these trends would lead to replacement-level fertility for Malaysian Chinese and Indians by the year 2000. Malay fertility is likely to continue to decline but at a more moderate pace.     

Resources and reproductive success in women with an example from the Kipsigis of Kenya

Contradictory results regarding the relationship between resources and reproductive success of women have led some social scientists to conclude that evolutionary biological models are inappropriate to the study of human social behavior. This paper suggests instead that the variability across societies in this relationship reflects an inadequate specification of the nature and availability of the resources critical to reproduction as well as a failure to understand the mechanisms whereby resources confer reproductive success in traditional, developing, preindustrial, and modern societies. These methodological and conceptual issues are illustrated through use of data on the association between wealth and reproductive success from the Kipsigis, a polygynous agropastoralist population in southwestern Kenya. In this society, land is owned by men, and women gain access to land through marriage. In 3 of the 5 marriage cohorts studied, women with access to larger land plots had higher lifelong reproductive success than their poorer counterparts both in terms of enhanced fertility and survivorship of offspring. This association was independent of confounding factors such as education, age at menarche, husband's age, or occupation. Moreover, wealthy women were found not to make greater use of modern medical child health services when their children were sick than poor women. The Kipsigis data indicate that wealthy women had more nutritional resources than poor women and were able to introduce more suitable weaning foods, leading to a lower incidence of episodes of illness in offspring. Overall, the findings suggest that wealth-related differences in the nutrition and health of mothers and children are important factors in reproductive differentials in Kipsigis society.     

Canadian fertility trends in perspective

This paper examines the broad movements of Canadian period and cohort fertility over the past 100 years, and compares them with corresponding trends in the US and other industrialized countries. The main movement in Canada was a decline in fertility extending from the 19th century to the present, interrupted in the 1940s and 1950s by a baby boom. Between 1871 and 1937 the total fertility rate (TFR) fell at about an average of 1.4% annually. The rate of fall in the US was similar, with the result that in the late 1930s the Canadian TFR was about 20% that of the US. The fertility boom that followed was steeper in the US than in Canada, and in the downswing that later followed, the rate of decline was similar in the 2 countries (3.4-4% annual average). But, the decline continued longer-- indeed still continues--in Canada, whereas the TFR in the US reached its lowest point in 1976. Moreover, the recent decline in fertility has been more severe in Canada than in almost any other industrialized country. The TFR relates to fertility in a single year and is highly sensitive to short-term changes in the timing of births. For the purposes of understanding and explaining long-term trends in fertility, the completed fertility rate (CFR) is a better index because it measures the ultimate family size of cohorts. Cohort fertility can be seen to fluctuate much less than does period fertility. In both the US and Canada, the peak cohorts, born in the early 1930s, had roughly the same completed fertility, and later cohorts continued to remain closely in step as the CFR fell sharply. In Canada the fall continues, though at a reduced rate, in the latest cohorts for which there is information. Apart from differences in amplitude, the dates of turning points and the shapes of the TFR and CFR curves of Canadian fertility are fairly similar. The long decline in cohort fertility is largely explained by the decrease in the proportions of families of 6 or more children. During the baby boom, for Catholics and non-Catholics alike, the proportion of ever-married women remaining childless fell by about 40%, the proportion having 2 children changed little, and the proportions having 3,4 and 5 children tended to increase substantially. The crucial difference between the 2 groups was in the proportions of women having 6 or more children. For non-Catholics, the proportion fell by over 4% from a high level. For non-Catholics, on the other hand, the proportion rose somewhat, though even after the rise, it was barely more than 1/2 the level to which the Catholic proportion had declined. Among Catholics, the effect of the massive decline in the proportion of women having 6 or more children was to swamp the effects of the increase in the proportions of women having 3, 4 and 5 children. The net effect was that fertility declined. Among non-Catholics, however, the comparable increases in the proportions of families of 3, 4 and 5 children, were not offset by any fall in the proportion of larger families, with the result that a baby boom occurred.     

Marital fertility in polygynous unions in Nigeria

The effect of polygyny on fertility, by age cohort, was examined in 4 regions of Nigeria through use of data from the 1981-82 Nigerian Fertility Survey. Simple tabulation of numbers of live births by maternal age and type of marriage union indicated that, with the exception of the 30-39-year age group, fertility in polygynous unions tends to be higher than that in monogamous marriages. Overall, such tabulations reveal an average of 3.90 births among polygynous women and 3.47 births among monogamous women. However, when age-specific fertility rates were compared, except for women under 20 and over 40 years of age, rates were higher in monogamous unions (7.286 overall) than for women in polygynous households (7.200 overall). Mean completed fertility, taking into account marriage duration as well, shows a tendency for women in polygynous unions to be more fertile (with the exception of women aged 25-44 years who had been married 10-19 years). The absolute difference in fertility between the 2 types of marital unions ranged from 0.32 children in the northwest to 0.64 in the southeast. With adjustment for independent variables such as education, the absolute mean differences were reduced, from between 0.28 children in the southeast to 0.42 in the northeast, but the direction of high fertility was still in favor of women in polygynous unions. But when the covariates (e.g., age at marriage) were taken into account as well as the independent variables, there was a dramatic reduction in the mean difference between the fertility of these 2 groups of women. These findings suggest that some changes in reproductive behavior are taking place in Nigeria that are restricting the fertility of women in polygynous unions. These changes are hypothesized to reflect the spread of formal education in Nigeria, with the expectation that women will contribute to the costs involved in educating their children.     

The life histories of American stepfathers in evolutionary perspective

This paper presents an analysis of the characteristics of men who become stepfathers, and their subsequent fertility patterns and lifetime reproductive success. Because women who already have children are ranked lower in the marriage market than women without children, men who marry women with children (e.g., stepfathers) are likely to have lower rankings in the marriage market as well. Using retrospective fertility and marital histories from the Panel Study of Income Dynamics (PSID), I show that men who become stepfathers have lower levels of education, less income, and are more likely to have been divorced before and to already have children, all characteristics that lower their rankings in the marriage market. Men with one or two stepchildren are just as likely to have children within a marriage as non-stepfathers, although men with three stepchildren show decreased fertility. Among men age 45 and older, stepfathers have lower lifetime fertility than non-stepfathers, although the difference disappears when men’s age at first marriage is controlled for. Additionally, stepfathers have significantly higher fertility than men who never marry. The results suggest that some men become stepfathers to procure mates and fertility benefits that they would otherwise have been unlikely to obtain; for these men, raising other men’s children serves as a form of mating effort. Preliminary versions of this paper were presented at the Evolution and Human Adaptation Program at the University of Michigan, and the Human Behavior and Evolution Society’s annual meeting at Amherst. Kermyt G. Anderson received his Ph.D. in anthropology from the University of New Mexico in 1999. He is currently a Mellon Postdoctoral Research Fellow at the Population Studies Center of the Institute for Social Research at the University of Michigan. His current research examines the relationship between family structure, parental investment, and children’s educational and employment outcomes in South Africa.     

Fertility and population policy in two counties in China 1980-1991

A survey of women in two highly developed rural counties of China, Sichuan and Jiangsu Provinces, was carried out in late 1991, to gain information about demographic and economic change between 1980 and 1990. Three separate surveys were conducted: the first a questionnaire administered to married women aged 30-39, eliciting information about childbearing and contraception, as well as the social and economic background of the respondents; the second, focus group interviews emphasizing the motivation for childbearing. Official information about the selected villages, townships and counties was also collected. National level data in 1987 show that individual reproductive behaviour in China failed to conform to a universal, effectively implemented, population policy. They imply either a spatial range of policies, or great diversity in the demand for children, or perhaps a combination of both. Such diversity in reproductive behaviour is also found in the study area. The purpose of the analysis was to examine the diversity in reproductive behaviour and contraceptive practice, and to discover whether differentials are influenced by area, or else exist between individuals within areas. If the former, then the explanation may be found in differences in policy formulation and implementation between areas: and if the latter, to demand for children, or else differential application of policy restrictions. The main findings were that: (1) the explanation of the pattern of fertility and contraceptive use is to be found at the individual level (within locations) rather than in policy differences between administrative units; (2) the association between income and number of children is negative, as is that between income and the propensity for uniparous women to remain unsterilized. The theory that privilege may be exercised to gain concessions from birth planning cadres is therefore not supported; (3) ideal family size differentials are largely absent, showing that social (education) and economic (income, occupation) characteristics are not responsible for differences in reproductive motivations, and implying that the nature of the demand for children is very different from that in most rural areas of the Third World; (4) data on ideal family size by sex of the existing offspring indicate only a weak preference for sons. The low demand for children, and the weak son preference, may both be explained by the social acceptability of uxorilocal marriages, and of village endogamy, together with the prohibitive costs of children, and especially of sons. This partly results from the expense of education, but most mothers emphasize marriage costs. It is speculated that the circumstances responsible for the escalating costs of children in the two countries are likely to pertain in growing areas of the country, with the privatization of education and health services, the declining support of collective institutions, and the replacement of this function by kinship networks. These on-going changes imply that any policy of reproductive restriction for the purposes of population control is likely soon to meet with diminishing resistance; and it may later be rendered unnecessary in the eyes of government officials, as fulfilled reproductive intentions lead to a fertility level below replacement level.     

Differential paternity in Libya

Libyan census and vital statistics data from 1973 are compared with genealogical records from Utah males born between 1830 and 1834 as representative of populations not using any method of fertility control. The Libyan vital statistics data contained 97% of paternity by age, but only 85% of maternity by age. These missing data were distributed pro rata, and all data were corrected for errors in reporting. Polygamous unions were excluded because polygamy is relatively rare in Libya. The Utah data were from 185,000 genealogies of the Genealogical Society of Utah. The Libyan child-woman ratio (number of children under age 5 per number of women aged 15-49) is 1112.9, compared to 850 in Morocco, suggesting that Libya is experiencing an increase in fertility, in child survival and probably in quality of statistics. The total fertility rates for females were 11.1 for Libya and 11.2 for Utah; the total paternity rates were 14.3 and 13.7, respectively. Male rates are higher because of remarriage after divorce or death of wives. Age-specific paternity rates are tabulated and graphed: The major difference between the 2 populations is the concave shape of the curve for Libyan men under 30. Age at marriage is late, but marriage is virtually universal for Libyan men over 30. Age-specific paternity rates by occupation show apparent lack of fertility regulation in traditional occupations like farming and sales. There is evidence of some parity-related fertility control in professional and administrative workers. Production workers have a high peak in fertility around age 27.5 and 32.5, and a dip occurring at older ages. These figures can be explained by education, since education is required for professional occupations, and older professionals were trained in the West. Production workers took advantage of rapidly expanding education in Libya late in their youth, postponing marriage. Libya's pronatalist policy forbids sale of contraceptives and provides child allowances, free education, health care, subsidized housing and social security. This paper indicates the utility of paternity data where statistics on maternity are unavailable.     

Cessation of childbearing in the absence of contraception in Sri Lanka

In data from the Sri Lanka Fertility Survey, 1975, the cessation of childbearing is examined among women who have never used contraception. The sample consisted of 6810 currently or previously married women, 57% of whom reported that they had never used contraception. Cessation of childbearing is studied according to age and marriage duration. The variables analyzed are age at last birth, proportions infertile during the last 5 or 7 years, and the infertile open interval. The duration of breastfeeding is taken into account where necessary, and the contraceptive users and nonusers are compared where appropriate. Non-users tend to cease childbearing early, and therefore are infertile for longer periods during their marriages. It is probably age of the 1st child that influences decisions on future fertility. Among women aged 45-49 who married before age 20 and continued in their 1st marriage, mean age at last live birth in non-users, was 34.5 years, about 2 years earlier than in those who had used contraception. Non-users who married at any age below 30 years cease childbearing well below age 40. The proportion not currently pregnant and infertile over the past 5 years increases with marriage duration among the fertile non-users in each age group. When age at last birth and the duration of breastfeeding in the open interval are taken into account and the reference period is increased to 7 years, the period of infertility increases with marriage duration among nonpregnant non-users below age 45. The proportion of women who were currently not pregnant and had remained infertile over the past 7 years is higher among the older non-users whose 1st child was born more than 10 years ago.     

Deformed nasal septa and relaxed selection

Abstract

Demographic studies undertaken in several Andean countries have found that women residing at high altitudes have significantly fewer live births than do their low altitude counterparts. This reduction has been explained as being due to various factors: the debilitating effects of hypoxia upon the reproductive system; the effects of sociocultural factors which vary with altitude and which affect reproductive behavior; and errors in data collection. In order to examine the validity of some of these hypotheses, the fertility of a group of 906 Bolivian women residing at low, medium, and high altitudes was examined. The women were selected from the lower socioeconomic strata and reported never having used any method of contraception. A detailed analysis of the fertility of these women showed no significant altitude‐related differences in the number of live births. However, as a result of significantly higher childhood mortality rates at altitude, there was a significant reduction in numbers of living children. The results of this study suggest that the collection and analysis of census data that ignores socioeconomic differences within a population or differences among census units in neonatal or early childhood mortality may bias or complicate the study of the impact of altitude on human fertility. Although the present research does not prove that hypoxic stress does not affect the reproductive system, the results suggest that if altitude does reduce fecundity, the reduction is not great and is likely to be shown only through studies of reproductive physiology.     

The delineation of fertility strategies in a tribal population of India: the Koyas of Koraput District,Orissa

Demographic data collected for a tribal population of India, the Koyas of Koraput District, Orissa, were examined in light of 2 models of reproductive behavior associated with the economic value of children: the replacement effect and son survivorship motivation. Both models are united in the concept that infant/child mortality affects subsequent fertility. The database consists of retrospective fertility histories of Koya women who had completed their reproductive period. The total number was 260, with the total offspring numbering 1407. 2 distinct cohorts of women were formed for the purpose of analysis, separated only by the criterion of offspring survival: women who had experienced infant child mortality (129 women with 739 children); and women who completed their reproductive period without suffering offspring loss of this nature (132 women with 668 children). The cohort without child loss had a mean parity of 5.10, lower than the average parity of 5.73 recorded for the cohort whose reproductive histories included at least 1 infant/child death. Age specific marital fertility and birth interval analyses indicated that this differential was because of biological, not behavioral, factors. The age pattern of fertility of females suffering offspring mortality failed to demonstrate a high rate of childbearing in the later age intervals of the reproductive period, a characteristic pattern of couples attempting to "replace" lost offspring. Birth interval analysis pointed to biological "interval effect," whereby infant/child mortality caused a cessation of lactation and hence a shortening of postpartum amenorrhea. Computer simulation further indicated that the higher fertility differential of the cohort experiencing offspring loss still did not result in high son survivorship values. The findings agree with earlier studies indicating that for predemographic transitional populations, economically motivated fertility strategies are ineffectual.     

Educational homogamy lowers the odds of reproductive failure

Assortative mating based on education is a common phenomenon. We investigated whether it affected parameters of reproductive performance such as childlessness, offspring number and age at first marriage. On the basis of the US census from 1980 (n = 670,631 married US couples), we find that the proportion of childless individuals is usually minimal in women married to a husband of the same educational level. This holds particularly true in the highest and the lowest educated women. Educational homogamy is also associated with a lower average age at first marriage. No obvious effect of educational homogamy on a woman''s average offspring number is found, where mean offspring number generally increases both with decreasing woman''s and decreasing husband''s educational attainment. We conclude that educational homogamy reduces the likelihood of reproductive failure.     

Natural selection at the MJD locus: phenotypic diversity, survival and fertility among Machado-Joseph Disease patients from the Azores.

Machado-Joseph Disease (MJD) is an autosomal dominant neurodegenerative disorder of adult onset, associated with the expansion of a (CAG)n tract in the coding region of the causative gene, localized on 14q32.1. Machado-Joseph Disease shows non-Mendelian features typical of other triplet repeat disorders, including clinical heterogeneity, variable age at onset and anticipation. Three phenotypes have been proposed (clinical types 1, 2 and 3). Type 1 is associated with early age at onset and a high repeat number of the CAG sequence, and Types 2 and 3 have later onset and lower numbers of CAG repeats. This paper investigates whether there is selection against the MJD gene, acting through differential survival. nuptiality and fertility associated with clinical type and age at onset. The study sample comprised 40 MJD patients from the Azores (Portugal) having fully documented reproductive histories and known dates of death. The proportion of married patients of each clinical type increased from 0.22 among Type 1 patients, to 0.40 in Type 2 and 0.95 in Type 3. Age at onset and length of survival were also associated with marital status, with the married cases having later mean age at onset and longer mean survival time. In the whole sample, clinical type was associated with fertility, with significantly fewer children born to Type 1 patients. Among married patients clinical type was not associated with age at marriage, reproductive span or number of children. No reduction of fertility was detected among married patients in whom the onset of MJD was below the age of 50. The authors' interpretation of these results is that the high-repeat CAG haplotypes associated with early age at onset and clinical Type 1 are selected against through reduced survival and fertility. The fertility component of selection is mediated by nuptiality rather than marital fertility.     

Biological and behavioral determinants of fertility in Tierra del Fuego

The reproductive history of 182 women in postreproductive life or near menopause from the Chilean part of Tierra del Fuego was traced back by means of familial interviews. These postmenopausal women represent the population since almost the beginning of the settlement, and their reproductive years were spent on the island. Path analysis was applied to analyze fertility determinants of these women and to propose a complex model of interconnections among factors. The reproductive history of these women is characterized by a long fertile span, a short childbearing period, and low fertility. Age at menarche is relatively late, and the age of the women at first birth is mainly determined by their late age at marriage. The use of contraception is related to both spacing and stopping behaviors. The late age of women at marriage, the rhythm of conception, and practices of contraception are proposed as the main determinants of fertility in Tierra del Fuego.     

Polygyny and fertility in Morocco: the effect of marital instability

This paper analyses the relationship between polygyny and female fertility in the province of Marrakech (Morocco) taking into account the effect of the previous marital instability (number of marriages) and the possible association between female sterility and type of marriage. In the analysed population, polygyny increases in the small towns. Polygynously married women have a higher level of education and show a higher percentage of use of contraceptive methods than the monogamously married ones. Although polygynously married women initially show lower fertility, multivariate analysis carried out on the group of women between 35 and 49 years old show that there are no significant differences in fertility between monogamously and polygynously married women when the effect of the previous marital instability is considerer. Female sterility does not determine marital instability, although it does determine a significant increase in polygynous marriages among the women without children.     

A Cultural–Historical Perspective on the Depressed Fertility Among the Matrilineal Moso in Southwest China

Family systems exert substantial influence on the demographic processes of populations. The Moso in Southwest China feature an institutionalized visiting sexual system, which differs from marriage, and a grand household system without conjugal units. Moso women enjoy virtually unlimited reproductive autonomy. On the basis of surveys of 127 Moso households conducted in 1988 and 1989, this paper analyzes the fertility experience of Moso women during the twentieth century and the mortality rates of the Moso population in the PRC era. Moso fertility patterns are characterized by a late age at first birth (median 23 years), long interbirth intervals (median 3 years), and a high rate of childlessness (16%). We conclude that this pattern of low fertility is an outcome of the unique Moso cultural practices. For the cohorts of Moso women born between 1905 and 1929, pathological sterility caused by STD is likely to have depressed fertility as well.     

Factors in the incidence of childlessness in Canada: an analysis of census data

Due to the difficulties involved in attempting to determine the relative proportion of involuntary childlessness and of voluntary childlessness in a given population, many investigators insist that the problem can only be examined in small-scale studies using intensive psychological interviewing techniques. A method for assessing the causes of childlessness in a population using census materials is described. If a distinction is made between psychological and physiological causes, instead of between intentional and unintentional causes, it is possible to assess the relative importance of these causes using large-scale investigation techniques. Physiological causes include all physiological conditions which produce sterility. Psychological causes include both psychosomatic infertility and the voluntary decision not to have children. The method involves using the minimum rate of childlessness in a population group known to place a high value on fertility as an estimate of the rate of physiological childlessness in the population. This estimate is then subtracted from the childlessness rate observed in other population groups in the same society in order to determine the degree of psychological childlessness in these other population groups. This method was used to assess the causes of childlessness in Canada. Census materials were used to determine the minimum rate of childlessness among rural women in Quebec. Since this group is primarily Catholic and places a high value on children, its childlessness rate provides an estimate of the proportion of sterile couples in the population. This rate was then subtracted from the childlessness rates for urban Canadian women, and the remainder provided an estimate of psychological childlessness among urban women. Age of marriage was controlled for since rural women married at younger ages. Study findings were 1) the proportion of childlessness among urban women declined over the years as 15.2% of the women over 45 years old were childless while only 10.8% of the women, aged 30-44, were childless; 2) the proportion of physiological childlessness declined from 6.6% among women over 45 years old to 4.6% among women, aged 30-44; and 3) approximately 50% of the cases of childlessness among urban women were due to psychological factors. These findings do not support the contentions of some investigators that 10% of the population is sterile and that psychological childlessness is rare. Study findings were presented in tabular form.     

Reproductive life of some Gujar women of Punjab

Data was collected on current age, age at menarche, marriage age, maternal age at 1st birth, age at the birth of last child, age at menopause, total number of conceptions, live births, stillbirths, abortions, dead children and living children for a sample of 150 Gunjar women of Punjab, India, during September and October 1977 to study their reproductive life. The women ranged in age from 45-55 years. The mean age at menarche was 14.90 years for the sample. The median age at menopause was 46.20 years. The mean age at marriage of the present sample was 12.56+-2.50 years; the mean age of the mother at the birth of her 1st child was 16.85 years; and the mean age at the birth of the last child was 38.68 years. The average number of conceptions was 7.2; the average number of live births of these 150 women was 6.90. The fertility of this population was natural as they were not using any family planning method.     

Community-level education accelerates the cultural evolution of fertility decline

Explaining why fertility declines as populations modernize is a profound theoretical challenge. It remains unclear whether the fundamental drivers are economic or cultural in nature. Cultural evolutionary theory suggests that community-level characteristics, for example average education, can alter how low-fertility preferences are transmitted and adopted. These assumptions have not been empirically tested. Here, we show that community-level education accelerates fertility decline in a way that is neither predicted by individual characteristics, nor by the level of economic modernization in a population. In 22 high-fertility communities in Poland, fertility converged on a smaller family size as average education in the community increased—indeed community-level education had a larger impact on fertility decline than did individual education. This convergence was not driven by educational levels being more homogeneous, but by less educated women having fewer children than expected, and more highly educated social networks, when living among more highly educated neighbours. The average level of education in a community may influence the social partners women interact with, both within and beyond their immediate social environments, altering the reproductive norms they are exposed to. Given a critical mass of highly educated women, less educated neighbours may adopt their reproductive behaviour, accelerating the pace of demographic transition. Individual characteristics alone cannot capture these dynamics and studies relying solely on them may systematically underestimate the importance of cultural transmission in driving fertility declines. Our results are inconsistent with a purely individualistic, rational-actor model of fertility decline and suggest that optimization of reproduction is partly driven by cultural dynamics beyond the individual.     

Desire for sons and subsequent fertility in rural India. A 20-year longitudinal study

This paper compares the desired fertility of rural Indian women in 1987 with their actual fertility in 2007. Seventy-one respondents who stated definite fertility intentions and had fewer children than desired in 1987 were re-interviewed 20 years later, as part of a larger study. The results indicated that these women had fewer children than intended and stopped childbearing once they reached, or approximated, their desired number of sons. The majority had been sterilized, indicating broad acceptance of lower fertility among rural women and the success of India's family planning efforts, although the practice of sex determination seems also to have played a role. These findings echo those of an earlier longitudinal study of reproductive intentions and outcomes in the same community, demonstrating the persistence of son preference in determining reproductive behaviour, even in the context of low overall fertility. The paper concludes with a discussion of the policy and programme implications of the study's findings.     

Premarital fertility in Namibia: trends, factors and consequences

Premarital fertility, defined as fertility before first marriage, was found to be highly prevalent in Namibia. According to data from the 1992 and 2000 DHS surveys, the proportion of premarital births was 43% for all births, and 60% for the first birth. This seemed to be primarily due to a late mean age at first marriage (26.4 years) and low levels of contraception before first marriage. Data were analysed using a variety of demographic methods, including multiple decrement life table and multivariate logistic models. Major variations were found by ethno-linguistic groups: Herero and Nama/Damara had the highest levels of premarital fertility (above 60%); Ovambo and Lozi had intermediate levels of premarital fertility (around 40%); Kavongo and San appeared to have kept a more traditional behaviour of early marriage and low levels of premarital fertility (around 20%). The largest ethno-linguistic group, the Ovambo, were in a special situation, with fast increasing age at marriage and average level of premarital fertility. Whites and mixed races also differed, with Afrikaans-speaking groups having a behaviour closer to the average, whereas other Europeans had less premarital fertility despite an average age at marriage. Ethnic differences remained stable after controlling for various socioeconomic factors, such as urbanization, level of education, wealth, access to mass media, and religion. Results are discussed in light of the population dynamics and political history of Namibia in the 20th century.