“Zombie facts.” We’ve all heard them. Many have unwittingly used them: fictional data points cited, recited, and recycled so often that, like zombies in a science fiction movie, it seems they simply refuse to die. In order to take action against gender inequality, let’s get the data straight.

A shopgirl in Pyin-Oo-Lwin, by Thomas Cheatham for the World Bank Group
A shopgirl in Pyin-Oo-Lwin, by Thomas Cheatham for the World Bank Group
 This Op-Ed originally appeared on


Gaps in what we know about women and girls are vast and real. Existing data tell us far too little about women’s workplace conditions and wages, ownership and control over land and housing, economic empowerment, financial inclusion, poverty, and time use.

Let’s  have a look at some of the most persistent “zombie facts” about gender so we can lay them to rest once and for all.


“Women are 70 percent of the world’s poor.”

False. Poverty is generally measured in terms of income or consumption at the household rather than individual level, so we cannot yet calculate separate poverty rates for men and women. Females may be overrepresented among the poor if “female-headed” households are poorer than “male-headed” households—but the category of headship poses its own problem (see third bullet)—or if intra-household distribution in is unequal (for instance, in some places, women may have less access to household resources than men).  They may also be poorer than males when we consider dimensions of well-being other than income, such as education and health.


“Female-headed households are the poorest of the poor.”

False. Data from 35 nationally representative surveys for 20 countries analyzed by the UN Food and Agriculture Organization (FAO) show that “female-headed” households are more likely to be poor than “male-headed” households. We cannot generalize. Female headed households are not a homogeneous group; data limitations make it impossible to distinguish systematically among households headed by women who are single, widowed, or divorced and those associated with an adult male who supports the family through remittances and social networks.  These demographic differences matter for poverty status.


“Female- and male-headed households are appropriate representative categories.”

Again, false. The categories of male- and female-headed households are problematic because they comprise fundamentally different living and demographic units. “Male-headed households” conflates those households with males and no female adults with households comprising households with both adult men and women. Female-headed households are almost always defined as those without an adult male present. These definitions, further, vary among countries, making cross-country comparisons problematic: Some countries define “head” to mean the eldest member of the household while others define it as “primary breadwinner” or “primary decision-maker.” Until individual level data become available, work-arounds to headship include categorizing households by economic type (e.g., those with a single male or single female breadwinner with and without dependents versus dual-earner households with and without dependents, or by sex composition of the adult members (e.g., those with a majority males, majority females, or equivalent number of each sex.)


“Women produce 60-80 percent of the world’s food.”

False. This statistic cannot be supported in any rigorous way. Different definitions of “food” and “production” make estimation difficult. And as Cheryl Doss points out, because food production requires a variety of resources such as land, labor, and capital—often controlled by men and women working cooperatively—breaking down male and versus female contributions and output just isn’t possible. Women are also broadly more time-constrained than men because of the differing roles they play in families and societies, making the statement even more improbable.


“Women own less than 2 percent of land worldwide.”

False. The global community lacks sufficient data on individual level asset ownership and control to draw any such conclusion. Most data are collected at the household level only—not the individual level.


“Women and girls are more likely than men and boys to be undernourished.”

Once again, false. Available evidence doesn’t support this statement, and generalizing broadly is ill-advised. The limited evidence that is available suggests women and girls may be more malnourished in Asia, but this doesn’t hold true in much of sub-Saharan Africa.


You are aware of other  “zombie facts” on gender equality. Please share these in the comments section below.


The opinions expressed on this blog are those of the authors and do not necessarily reflect the official position of their institutions or of AFD.

I subscribe to the ID4D newsletter

Once a week, I receive the latest blog posts!