Today we speak to Elizabeth Stuart, Research Fellow at the Overseas Development Institute and lead author of a recent report, The Data Revolution: Finding the Missing Millions, about its findings. She warns that many countries are lacking important data on poverty, health and education.

DD: What do you mean by “missing millions”? Which countries are the most affected by this absence of data?

ES: Many of the numbers that we think of as facts are in fact estimates based on broad assumptions, extrapolations and imperfect models. For example, at the global level there are up to 350 million people who have been missed out of household surveys—and therefore uncounted—either by design, because the surveys don’t include migrant populations, people living in institutions or the homeless, or in practice, because of the difficulty in conducting surveys in insecure areas.

At the national level, there are significant data gaps in many developing countries. The problem is particularly acute in sub-Saharan Africa, where more than 40% of countries have not had a survey in seven years.

DD: What other data lags did you pinpoint? Are you saying all statistics are faulty?

ES: No, not all statistics are faulty, but I think it’s fair to say that most development statistics are. By that, I mean the data that are tracked on the main issues of development, such as health, education and economic growth. Child mortality data are normally considered to be the most robust among the development numbers, but even here two thirds of the countries that account for the vast majority of child deaths do not have registries of births and deaths and 26 have no child mortality data for the past six years, so have to rely on estimations based on flawed models.

There’s a whole range of things beyond the Millennium Development Goal issues that you’d think we measure accurately, but we in fact don’t. These include pretty basic things, such as the number of people who live in cities, the number of girls who are married before the age of 18 and the size of the sub-Saharan African economy.

DD: Why is it so important to collect quality data?

ES: The most obvious reason is that data are needed to measure progress. It is problematic that we’re in the process of agreeing a new set of development goals and targets that, under present conditions, we would not be able to tell had been achieved. But more importantly, it’s vital that governments know their people if they’re going to be able to deliver services to them in a systematic way. Of course, most countries in Africa have been able to eradicate guinea worm and China has been able to reduce poverty without perfect data, but if you want to provide health care, education and social protection—especially to those most in need—then governments need to know who they are, what they look like and what their requirements are. In other words, data are not just about measuring progress, they are also about delivering the policies to achieve that progress.

DD: How do we plug the gaps to bring on the data revolution, as you call it?

ES: In some countries around the world this is already happening. For example, in Colombia, the Ministry of Agriculture is working with a local cooperative of rice growers and a research institute to use big data analytical techniques to reduce risk for small-scale farmers. Because of climate change, rice yields had been reducing dramatically. But by analysing data from the farmers themselves, and combining it with weather data, annual rice surveys and agronomic experiments with changing the sowing dates, the cooperative has been able to give pinpoint accurate advice on precisely the best window to plant, and the exact variety to plant. This saved farmers more than US$ 3.5 million in just one harvest alone—and it means that they no longer need to rely on multinational companies for planting information. One farmer reported that he trusted the advice because it was “based on his own data”.

Beyond this, the expansion in the use of mobile phones and the ability to machine count anything (such as different roof types, as a proxy for poverty—an experiment that’s happening in Uganda) means that the revolution really is opening up to the masses. In Liberia, 10 000 wells and water points have been photographed and mapped using Android phones, giving the government a live planning tool for water provision.

DD: How will this data revolution change the power dynamics between people, governments and the private sector?

ES: Done right, the data revolution could completely change the way that citizens and governments interact. Data could be used to track service provision, enable citizens to reallocate local budgets and better participate in democratic processes—perhaps for the first time. However, of course we don’t have a theory of change yet and impact will not be automatic. It’s most likely that data will actually deliver change when it’s used in a context where the government is already minded to reform.