Melinda Gates announced a new $80 million Bill and Melinda Gates Foundation commitment to global data collection in a May 17 address at the Women Deliver conference in Copenhagen. Over three years, the foundation’s efforts will focus on filling gaps in data about women’s unpaid labor, improving the accuracy of data around land and property ownership, and using that data to inform civil and government decision-makers about the effects of their existing programs and recommend areas for improvement.

“We can’t close the gender gap until we first close the data gap,” Gates said in her speech. In many places around the world, Gates continued, data about women’s work, pay, births, and deaths are missing – and “where the data does exist, quite honestly, it’s sexist.” That is, by leaving out the work of women (which includes unpaid labor like care work and subsistence farming that supports others’ moneymaking work), or failing to disaggregate numbers by gender, our existing data about women’s lives are undervaluing women’s participation in the economy.

It also reinforces stereotypes about men as producers and women as reproducers. “The way information is collected embodies many of the gender biases that exist in society,” Ruth Levine, director of global development and population at the Hewlett Foundation, told Slate. (The Hewlett Foundation funded one of the first major gender data initiatives, Data2X, which will harness the Gates Foundation’s new investment.) As a result, governments and agencies know a lot about how, when, and how often women have sex and babies and a lot about how, where, and for what pay men work, but little data has been collected on men’s health or women’s work. As a case study, Gates brought up Uganda, where farming accounts for 23 percent of the country’s GDP, and 80 percent of farmers are women. There, when data collection expanded to include subsistence farming, the reported workforce grew by 700,000 people. “If the reality of your life is not represented in the data used to devise programs and policies, then you are not represented in the data,” Levine said.

Continue reading on Slate.