Guide on Gender Analysis of Census Data Full Draft of 6 December 2012 Contents
Country Example 1: Male Fertility in Norway
Download 4.06 Mb.
|
- Navigate this page:
- 4. Tabulations
- Table 3: Cambodia 2008 - Average numbers of children ever born classified by current age of the mother and highest level of education completed by the mother
- Table 4.A: Cambodia 2008 - Average numbers of children ever born classified by current age of the mother and major religious groups
Country Example 1: Male Fertility in Norway Below are male and female age-specific fertility rates (per 1,000) in Norway (2009): Ages Women Men 15-19 9.5 2.5 20-24 61.4 27.7 25-29 128.1 86.3 30-34 127.2 117.6 35-39 58.3 76.7 40-44 10.2 30.1 45-49 0.5 9.4 50-54 0.0 3.2
The dots in recommended Tabulation 3 indicate an age which may vary from country, generally 15 years or 12 years, sometimes 10 years. These tables can obviously only be generated in countries that have this information in their census (see the Text Box preceding Section 3). 83. The Principles and Recommendations also include the following additional tabulations for population censuses:
Of these, only 4 and 5 can be compiled in most censuses as the first three require information that relatively few censuses collect. 84. The table below, generated from the 2008 census of Cambodia, illustrates the cross-tabulation of fertility data by educational levels. Rather than using live births by sex within the 12 months preceding the census, the table shows the number of children ever born. The ideal procedure is to combine both (children ever born and children born during the past 12 months) to compute actual accumulated fertility rates, but because the number of children ever born is easier to understand, the table has been left in this format. What it shows is that in Cambodia fertility levels are fairly uniform for women with up to lower secondary education, but that fertility levels decline substantially as women complete their secondary education. Table 3: Cambodia 2008 - Average numbers of children ever born classified by current age of the mother and highest level of education completed by the mother
85. Beyond the additional tabulations suggested by the Principles and Recommendations, there may be others of potential relevance for gender issues. In countries where a significant proportion of unions are informal, consensual or polygamous, it may be relevant to tabulate the basic fertility data by type of union. Ideally, this information should be compiled by duration of the union, if available, rather than only by the woman's age. Consensual unions are usually associated with higher fertility than formal marriages (Henriques, 1979), even after controlling for other factors. Marital instability may play a role in increasing fertility rates as women feel the need to have at least one child with each new partner (Chen, Wishic and Scrimshaw, 1974). The latter, however, may be difficult to investigate using census data because the census generally does not provide any information on marital histories or even information on whether a woman has been married or in union before her present union. 86. Fertility levels vary by income in ways that are gender-specific. Paternal income tends to be positively associated with fertility, whereas maternal income is generally inversely correlated with income. The association is clearer when one of the variables is kept fixed while the other varies. This is understandable, given that high-income women suffer greater loss of income by having children. On the other hand, a higher income of the husband, given a certain income level of the wife, makes it more attractive to have children. 87. Given the social consequences of childlessness for women, it is recommended to prepare a table of women by number of children (or, at a minimum, women with children or without children) by 5-year age groups and marital status. Apart from the fact that such a table may identify the women at risk of social ostracism, it may allow the detection of actual trends, such as the higher incidence of divorce among women without children. The latter must be interpreted with caution, however, because the relationship may also go the other way, as women who divorced at an early age have had less time to become mothers. 88. Polygamous unions generally have lower female fertility rates than monogamous unions, due to lower coital frequency, especially older spouses, but they tend to have higher male fertility rates (Anderton and Emigh, 1989; Garenne and Van de Walle, 1989; Lardoux and Van de Walle, 2002; Shaikh, Aziz and Chowdhury, 1987). There are, however, certain forces that work in the opposite direction and that may bring about different results in some cases. For example, polygamy is associated with women's low status and inequality within marriage, which is often further exacerbated by large age difference between husbands and their second or third wives. One consequence of male dominance and unequal husband-wife interaction within a polygamous marriage is lower contraceptive use, mediated by the husband's disapproval of and lack of spousal communication about family planning (Hogan et al., 1999). It has also been observed that fertility may vary considerably between the first and later wives (Bean and Mineau, 1986). In addition, mothers depend on their children in later years, while fathers can receive support from younger wives. The number of children ever born, cross-tabulated by age and marital status (monogamous vs. polygamous union) should show which way these differences end up going. 89. A significant proportion (40-60 per cent) of polygamous unions in the Arab countries involve widows and divorced women who find it difficult to remarry as a first wife and therefore accept the status of second wife (Chamie, 1986). Unfortunately, census data normally do not contain information about the marital status of women before their present union, so that this relationship cannot be investigated. What can be investigated is the percentage of polygynous unions by level of education of the husband. Less educated men are more likely to be in polygynous unions, but some caution is called for as the relationship is partly explained by age effects. Older men are more likely to be in polygynous unions and they are also more likely to have a lower level of education. Therefore, it is best to tabulate the relationship by age groups, to separate the age effect from the educational effect. 90. Early marriage creates the conditions for early pregnancy and higher fertility rates. In societies where premarital childbearing is not socially and culturally accepted, a rising age at first marriage may play a crucial role in the transition from high to low fertility levels (Maitra, 2004). A study by UNICEF found that high levels of fertility were associated with the prevalence of child marriage in 50 countries (UNICEF, 2005). Women who had several children were significantly more likely to have been married before age 18 than women with no children. The relation between early marriage and fertility can be studied by cross-tabulating the age at first marriage, if available, with the number of children ever born, by age group or cohort. This will show the differences of fertility for women of the same age, or age group. In addition, the effect of early marriage on early pregnancy can be measured by tabulating the number of children ever born for young married women, by age. 91. Another tabulation of potential interest is the association between religion (when available) and fertility. Pronatalist ideologies, associated with different religious traditions, can have stances against the use of contraception (or certain types of methods) and abortion. When people have a strong attachment to religious communities, religious directives on gender roles and sexuality are likely to influence their fertility behaviour, although the extent to which religious convictions influence behaviour varies greatly. Verifying whether religious affiliation is associated with fertility rates and family size helps to understand cultural values with respect to gender, moral communities and social constraints toward women’s reproductive rights. The precise influence of religion is subject to a certain amount of controversy. Goldscheider (1999), for instance, maintains that Moslim women in Israel have higher fertility than the Jewish population not because of any specific prescriptions of the Moslim faith with respect to fertility and the use of contraception, but because of its views on the nature of familial relationships and the segregated social of roles of women which tend to emphasize their roles as mothers and spouses. Jejeebhoy (1995) has collected a lot of evidence that shows a negative relationship between the autonomy of women and fertility levels. Obermeyer (1992), on the other hand, does not accept the notion that Moslim women have less autonomy than women of other religions in similar contexts or that, on account of this lesser autonomy, they have higher fertility. Religion is sometimes used as an explanatory variable for contraceptive use (e.g. Addai, 1999 in Ghana; Adsera, 2006 in Spain), but because the census does not contain any information on contraceptive use, such studies are limited to data from the DHS or other types of fertility surveys and cannot be replicated with census data. 92. As an example, consider Table 4.A, which shows how the number of children ever born in Cambodia (2008) varies by the age of the woman, her area of residence (Phnom Penh or elsewhere in the country) and religion. Islamic women clearly have the highest fertility. In Phnom Penh, Christian women have the lowest fertility, but in the rest of the country their fertility is the next highest, after that of Islamic women. A study by Adsera (2006), on the link between marital fertility and religion in Spain in women aged 15-49 for the period between 1985 and 1999, found that fertility was particularly high among women in minority religious communities, such as conservative Protestant denominations and Muslims. It was lower among women in inter-faith marriages than among women in homogamous unions, particularly in unions where the husband was not Catholic. The study was based on two rounds of the Spanish Fertility Survey (SFS), but it would be possible to replicate it using census data. Table 4.A: Cambodia 2008 - Average numbers of children ever born classified by current age of the mother and major religious groups
Download 4.06 Mb. Share with your friends:
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||