From gender issues to gender statistics on demographic composition of population: illustrative examples
||Sources of data
|Does the sex distribution of births suggest a prenatal sex selection?
||Number of births by sex
||Civil registration system
|Are women or men in the working ages overrepresented in certain geographic areas, in urban areas or in rural areas?
||Population by sex, age, geographic areas, and urban/rural areas
|Are women overrepresented among older persons?
||Population by sex and age
- + Gender issues
- Several countries in the world have a sex ratio at birth significantly higher than expected, suggesting the practice of prenatal sex selection. Sex ratio at birth is determined primarily by biological factors. More boys than girls are born in all populations. The ratio usually varies from 103 to 107 boys per 100 girls born (United Nations, 1998). However, in the last decades, there has been a rise in the sex ratio at birth in several countries, mostly located in Asia, suggesting a prenatal sex selection to the disadvantage of girls (United Nations, 2010a). In countries with high sex ratio at birth, the population groups more likely to practice prenatal sex selection consider sons much more valuable than girls; have a relatively low fertility; and a significant proportion of women have access to technologies that allows them to know the sex of their foetus, making prenatal sex selection possible (UNFPA, 2010). Birth order, sex of previous births and total number of desired children are also important. For example, parents who prefer to have sons are more likely to practice prenatal sex selection if they expect to have only one child or if only girls were born so far in their family.
Gender differences in migration and mortality shape the sex and age composition of active age populations. Overall, women outnumber men in the older age categories (United Nations, 2010a). However, because of specific migration and mortality patterns, the overrepresentation of women at national or sub-national levels may start at earlier ages (United Nations, 2010a). More working-age women than men may live in rural areas or in certain regions of a country. In some countries men are more likely than women to migrate temporary from rural to urban areas or from one region to another, for work (United Nations, 2000b; 2008b). Men may also be more likely than women to migrate to other countries, although, lately, geographical patterns of female and male migration have become more similar (United Nations, 2010a).
Significant sex-differences in adult mortality may also contribute to the over-representation of women below the age 60. Overall, at the same age, men have a higher mortality than women, mainly due to biological factors. In some countries, the gender gap to the disadvantage of men becomes larger because social factors such as occupational risks, heavy drinking and smoking, war and conflicts affect more the survival of men than of women (United Nations, 2010a). In other countries, however, a number of factors increasing the vulnerability of women to infectious diseases or higher levels of maternal mortality may counterbalance the biological advantage of women and reduce the gender gap in mortality (WHO, 2009).
Since women live longer than men, they outnumber men at older ages, and this imbalance increases rapidly with age. Older ages are associated with changes in marital status, living arrangements, wealth, and health status that may affect women and men in different ways. Older women are more likely than older men to be widowed or divorced (United Nations, 2009a). In developed countries, older women are more likely than older men to live by themselves in one-person households, to have lower pensions and be at higher risk of poverty (United Nations, 2002; 2005; 2010a). In developing countries, older women with no pension benefits have to continue working for income while taking care of their husbands or grandchildren (United Nations, 2002). Many old women and men, especially among those with mental and physical impairments, become victims of abuse and violence (United Nations, 2002). Gender issues related to older ages are becoming more and more important as the share of older age population in total population is increasing, as a result of declining fertility and increasing life expectancy, a phenomenon called population ageing. This phenomenon started in the more developed regions but now is taking place in the less developed regions as well.
- + Sources of data
Population censuses collect data on sex and age characteristics of all individuals. These data can be used to estimate demographic composition of various groups of population at the time of the census. For population estimates in between the censuses, these data need to be combined with data on births, deaths and migration by sex and age either collected in the population census or from other sources.
Data on sex and age of young children (under age 1 or under age 5, for example) from population censuses may be used to estimate female and male births, if sex-disaggregated data on infant and child mortality is available. In any case, sex and age structure of young population provided by censuses can be used to assess the quality of data on female and male births as provided by censuses or other sources of data.
In countries lacking a timely and reliable system of vital statistics, population censuses often collect data on recent births (in the past 12 months) by sex.
Civil registration and vital statistics systems can provide data on live births by sex.
Population registers can provide data on composition of population by sex and age.
Household surveys can be used as a source of data on sex distribution of population in various age groups. Demographic and health surveys and fertility and family surveys may provide data on recent births, in the last 12 months, or retrospective birth histories during a longer time interval.
- + Conceptual and measurement issues
Female births may be more severely underreported than male births, in countries where women have a lower status. Registration of vital events may be incomplete and selective by sex. Some household surveys or censuses may also have sex bias in reporting of recent live births. It is important that quality of data on female and male births is assessed based on multiple sources providing data on births by sex, number of young children (under age 1, under age 5, or under age 6 in cases where age heaping is suspected) by sex and age, and number of children ever born by sex of the child and age of mother. Also, it is important that data from multiple sources and their quality are analysed over time. In general, it is expected that data quality improves over time with concurrent reduction of sex bias reporting. When this is the case, a substantial increase in the sex ratio at birth, compared to past levels and the normal range, is indicative of some prenatal sex selection.
Misinterpretation of sex ratio at birth may occur when sampling errors are not taken into account when analysing data based on surveys. A ratio of boys to girls of 112, for example, may suggest a prenatal sex-selection when the data are coming from a civil registration system considered complete; in the case of a survey of a small sample, however, it may be well within the confidence interval for the standard number of 107 boys to girls. Therefore it is necessary that countries calculate confidence intervals for survey data on sex ratio at birth.
In some population censuses or surveys, female members of the household may be more likely to be underreported than male members. Recording of household members in population censuses and household surveys may be sex-biased such as when the practice is to record first all the male members before all the female members. In addition, some sex-selective underenumeration may occur in countries or groups of population where women have a lower status. While a common problem in reporting the members of a household is the omission of infants, in some countries girls may be more likely to be underreported than boys.