PUTTING GRNDER INTO STATISTICS
There are many reasons for collecting data segregated by sex. Sex-disaggregated data is more often referred to, perhaps through some misplaced political correctness, as gender-disaggregated data. In fact collecting data disaggregated by gender whilst much more nuanced and useful for some purposes is also much more difficult & generally data is only sex-segregated. For purposes of this page however the terms will be used interchangeably and we will assume (just for once) that everyone takes on the gender performance indicated by their biological inheritance.
We shall also be reflecting on why statistics…mere numbers (or ‘lies, damn lies & statistics’ – according to Benjamin Disraeli, British Prime Minister 1804-81)…are not enough for many purposes, and need to be complemented by more qualitative data.
Why do we need sex or gender-disaggregated statistics? Gender experts need them to identify actual gender inequalities and to advocate for change. It is useful to know that despite getting the first woman into parliament in 1919 only 19.5% of MPs in the UK are women today(ipu.org). Even when numbers are acknowledged to be unreliable or inadequate the actual numbers are an essential advocacy tool as for example the much circulated figures of incidence of 5,000 honour killings per annum (unifem.org), of which according to some estimates as much as one fifth may take place in Pakistan (gendercide.org)
Sex-disaggregated data is needed by planners and policy-makers to provide, for example, health and family planning services catered to women and men; though it has been argued that in some areas of the heath system more nuanced targeting is needed for example in adjusting medicines for common illnesses to the usually different size and body weight of men & women; or to giving equal weight to research on the different needs of women and men.
Educational planners need to provide school facilities for boys and girls, need to see who drops out of school; which disciplines are chosen by girls and boys respectively. This statistical analysis can also lead to interventions of course; in case of drop-out of girls for marriage an increase in the legal age of marriage as well as public education might be two of the strategies tried. If girls are being streamed into soft areas by their school teachers as is often the case some re-training of teachers might be indicated as well as the establishment of vocational training courses which attempted to redress this bias by encouraging both women and men into non-traditional occupations. In some countries, for example, more men are being encouraged to teach primary school in order to provide positive male role models to young boys.
It may seem obvious to most people that data should reflect sex differences and we are used to seeing the parallel columns of ‘men’ & ‘women’ numbers. But in many countries sex-segregated data has been weak in many areas. The strongest areas are health and education where sex/gender differences are very easily understood. In health system data for example a large difference in the numbers of girls and boys born can indicate that advances in technology have made possible the suppression of foetuses of the least preferred sex, usually girls.
In the 1998 the Palestine Central Bureau of Statistics (pcbs.org) analyzed deaths of infants and children according to sex and discovered that although neonatal ( 0-28 days old) mortality was higher for boy babies (as is true globally) there was much greater attrition of girls between birth and the age of 11 months due to poorer nutrition, care, and medical attention. This was a stark demonstration of the existence of ‘son-preference’.
Traditionally sex-disaggregated data has been weak in the economic field on the assumption perhaps that the economy is gender neutral in its effects and that economic behaviour is not conditioned by social factors.
There is now however growing recognition that women’s work in the care/reproductive economy (as distinct from the ‘productive’ economy where men dominate (!)) or in the informal sector needs to be counted and valued in some way.
One of the instruments used is the Time-use survey which measures how women and men respectively spend their time over a given period. In principle this makes it possible to allocate a monetary value to unpaid activities which women provide such as child-care, marketing, cleaning, gardening & labour on the family farm by estimating the commercial cost of these activities performed by someone hired for the purpose.
(see more in unece.org)
Other topics where sex-disaggregated data are traditionally lacking include decision-making in the household and at local level, individual and household income, domestic violence and male fertility
(see more in openlibrary.org)
As noted earlier ‘hard’ statistical facts even when disaggregated by sex need to be supplemented by more qualitative data if we are to understand and to address the problem of boys or girls dropping out of school at different stages; or differences in access to credit, or patterns of saving, or participation in the IT sector as between men and women. This statistical knowledge of ‘what’ should lead to further probing into ‘why’ which will require us to go beyond mere numbers.
The private sector and in particular marketing experts have long recognized the absolute imperative of market research and of using some traditional qualitative anthropological methods - such as having a long chat over a cup of something appropriate - to round out the statistical picture. For the modern retail market indeed sex-disaggregated data is definitely not enough as is clearly indicated by their preoccupation with ‘pink purchasing power’ a term generally used to refer to the important market segment of gay and lesbian consumers.