Household Surveys

Is predicted data a viable alternative to real data?

The primary motivation for predicting data in economics, health sciences, and other disciplines has been to deal with various forms of missing data problems. However, one could also make a case for adopting prediction methods to obtain more cost-efficient estimates of welfare indicators when it is expensive to observe the outcome of interest (in comparison with its predictors). For example, consider the estimation of poverty and malnutrition rates. The conventional estimators in this case require household- and individual-level data on expenditures and health outcomes. Collecting this data is generally costly. It is not uncommon that in developing countries, where poverty and poor health outcomes are most pressing, statistical agencies do not have the budget that is needed to collect these data frequently. As a result, official estimates of poverty and malnutrition are often outdated: For example, across the 26 low-income countries in Sub-Saharan Africa over the period between 1993 and 2012, the national poverty rate and prevalence of stunting for children under five are on average reported only once every five years and once every ten years in the World Development Indicators.

Also available in: Español | Français
This is the first of three blog posts on recent trends in national inequality.

Inequality has featured prominently in the public debate in recent times. Media outlets highlight the apparent surge in the incomes of the richest, many books have been written on this issue, and numerous academic studies have attempted to assess the nature and magnitude of inequality over time. Most studies of inequality focus on the extent of inequality within a country; this makes sense since most policies operate at this level, too. Despite the attention this issue has received, it has been constrained by the quality of data on inequality. Household surveys collected by national authorities around the world are the most readily available source of data on inequality. However, compiling and harmonizing household surveys from different countries is extremely difficult as they are not always collected consistently or frequently enough. It is also well-known that household surveys often fail to capture the top tail of the distribution, as we will discuss in more detail in a future blog.

Also available in: Español

John Grunsfeld, former NASA Chief Scientist and veteran of five Space Shuttle flights, had several chances to look down at Earth, and noticed how poverty can be recognized from far away. Unlike richer countries, typically lined in green, poorer countries with less access to water are a shocking brown color. During the night, wealthier countries light up the sky whereas nations with less widespread electricity look dim.

Dr. Grunsfeld’s observation might have important implications. Pictures from satellites could become a tool to help identifying where poverty is, by zooming in to the tiniest villages and allowing a constant monitoring that cannot be achieved with traditional surveys.

Most of you probably buy lunch during the week, but can you recall what you ate yesterday? How about last week, did you snack in the afternoon? How much did you spend? Answering these questions is not as easy as you think, is it? Food consumed away from home (‘FAFH’ in short) represents an increasing share of food consumption around the world, caused by various factors including increasing urbanization, female labor force participation and evolving food systems that have made food availability easier.

The latest World Bank estimates suggest that the percentage of the developing world’s population living below $1.25 a day declined from 52% in 1981 to 22% in 2008. While this indicates that there is still a long way to go in poverty reduction, the progress is encouraging. Moreover, we now also appear to be in a much better position to make such statements. The numbers above, by my colleagues at the Department Research Group, are based on over 850 household surveys for nearly 130 developing countries, representing 90% of the population of the developing world. By contrast, they used only 22 surveys for 22 countries when the first such estimates were reported in the 1990 World Development Report.

One is always grateful to see attention paid to the quality and quantity of household data available to study poverty. It is a subject dear to my heart and to my colleagues in the Living Standards Measurement Study (LSMS ) in the World Bank. In sub-Saharan Africa, as a recent Global Dashboard post titled “What do we really know about poverty and inequality?” by Claire Melamed points out, there is still a dearth of data, even after years of government effort and international support. But there are data -- in some countries lots of data -- so it’s worth highlighting what is there. Today I wanted to add some nuance to the discussion of income and assets raised by Claire and, probably more importantly, steer people to some new data that will, we hope, excite the most blasé of you out there.

Household Surveys

Is predicted data a viable alternative to real data?

Inequality in the typical country in the last 25 years – a strong increase followed by a recent decline

Can a picture from space help to measure poverty in a Guatemalan village?

More people in the developing world are eating out. Measuring this well could change our understanding of poverty and inequality

To measure poverty over time, we need to do more!

Welfare, Assets, Data Availability and the Living Standards Measurement Study