In the US, more women than men die from heart disease complications; less consequentially, more female than male workers complain about office buildings being too cold during the summer. In Nigeria, farm households led by women are much less likely than those led by men to use modern agricultural inputs, such as fertilizers and farm machinery. In the low income neighborhoods of Lima, Peru, travel time and childcare demands stand in the way of businesswomen’s ability to complete business training.

Sub-standard data on women and girls is the common thread tying these seemingly unrelated issues together. The extrapolation of data collected on males to females without acknowledging sex differences helps explain the first two events. Gender biases in measurement instruments connect the latter two – surveys bias the identification of “head of household” towards men and fail to adequately record unpaid and domestic work, mostly done by women and girls.  

In our last blog post on gender-related data, we focused on indicators that are “ready to measure” and can help to kick start the process of collecting more and better data on women and girls. In this post, we turn to the more deeply-entrenched problems that will have to be addressed in order to ensure that women and girls, the work they do, and the obstacles they face are counted equally going forward. Encapsulated in a recent Data2X article, “What is Wrong with Data on Women and Girls?,” these problems can be grouped into two overarching categories: no data and bad data.

Problem Number 1: No Data

There are still areas where we simply have no data on aspects of women’s and girls’ lives. These include the following:

  • The fact that less research is done on women’s health conditions explains the first two examples cited above. The lack of data on aspects of women’s health means that men’s metabolic rates are used as the standard to set office temperatures and men’s experiences of heart disease are used to diagnose and treat women.
  • Because the household is often used as the basic unit of analysis by economists, intra-household dynamics negatively affecting women’s lives (including domestic violence and the unequal distribution of household resources between men and women) are not well-documented.
  • There are gaps in the sex-disaggregated data we do have; 80% of a total of 126 countries regularly produce sex-disaggregated statistics on education, and 65% to 70% produce statistics on sexual and reproductive health and fertility, but only 30-40% of countries produce statistics on informal employment, unpaid work, and violence against women.

Problem Number 2: Bad Data

“Bad data” is data that systematically misrepresents reality, particularly in ways that make women and girls appear to be more dependent and less productive than they actually are; this helps explain the last two examples we cited above. Women who head farm households are underreported in the statistics and bypassed by farm extension services. Training programs have been built on the mistaken assumption that women have plenty of free time, backed by limited data on women’s time intensive work schedules. Biased measures produce this bad data in a number of ways:

  • Instructions in widely used household surveys guide interviewers to identify males as heads of households when putting together household rosters.
  • Male members of households often respond to surveys on women’s and girls’ behalf.
  • Even when women respond themselves, traditional survey instruments don’t always reflect the work they do. Survey questions privilege formal and full-time jobs: those that are more often performed by men. As a result, much of women’s paid work (which is often seasonal, informal, or categorized as a secondary occupation) isn’t reflected, and unpaid work (such as housework, fetching fuel and water, childcare and eldercare) is not counted. 

What does it mean and what should we do?

The consequences of the lack of data about certain aspects of women’s and girls’ lives—and the bad data regarding other aspects—are significant and varied, and affect women’s well-being and opportunities in society.

So what can be done? The national and international actors we highlighted last time (the Global Partnership for Sustainable Development Data, the World Bank, and others) should start with the 20 “ready to measure” indicators we discussed, but by no means stop there. Survey instruments need to be revised to include women and girls and the work they do. Intra-household dynamics need to be examined in order for issues like domestic violence and the distribution of household resources to be better understood. Finally, women themselves need to be surveyed about the work they do and the aspirations they have rather than relying on what the male head of household says. 

The pay-off of addressing these challenges is substantial. The more women and girls are reflected accurately in data, the more their roles in society are recognized and valued, and the closer we get to achieving gender equality – and really leaving no one behind.