The EquityTool was designed to balance methodological rigor with practicality and simplicity. Therefore, we have simplified the Demographic and Health Survey wealth indices and turned them into an easy-to-use web application. An peer reviewed journal article was published in March, 2016 with the full details of our method in Global Health: Science and Practice. We’ve also written a brief summary below.

DHS Wealth Index

The EquityTool is based on the wealth indices generated from Demographic and Health Surveys (DHS) and other similar surveys.

DHS are large scale national surveys, supported by USAID. The surveys include questions about the respondent’s household – the assets owned and the characteristics of the household. Here are some examples of the questions often used to create the wealth index:

There are many asset and household characteristic questions in the DHS for each country. The data are used to generate a measure of wealth using a method called Principal Components Analysis. Through this process, each respondent in the DHS dataset is given a wealth index score that indicates their wealth relative to the rest of the population. We are able to then put the whole dataset in order from poorest to wealthiest and see the range of scores corresponding to each 20% segment of the population. This way, we can separate the population into wealth quintiles.

This score is related to, but distinct from, a household’s income. The wealth index does not measure income, which is an absolute measure. Instead, the wealth index is a relative measure, placing each person’s asset-based wealth in relation to others in the population. Asset-based wealth measures are thought to be more stable than income because asset ownership fluctuates less than income.

The DHS can give us a wealth index and wealth quintiles for the national population, as well as the urban and rural populations separately. In other words, we can see what an urban respondent’s wealth is relative to other urban people.

Detailed information about the DHS wealth index can be found on the DHS website here.

Simplifying the DHS wealth index

The EquityTool uses a simplified version of the latest DHS wealth index for each country. See the list of countries for which we have an EquityTool, and the survey used to create it in this table. The simplification involves finding the questions that are most important to the wealth index. We then ensure that the simplified version agrees with the quintile assignment results produced by the full wealth index at a specified threshold (kappa>=0.75).

Our simplified indices first remove any questions which were shown to be difficult to answer in our facility based pilot tests. For example, questions about the amount of land owned, or number of livestock owned, are removed. We also permit limiting the number of response options for questions with a long list of choices.

Click here to see an example of one of the simpler sets of questions.

Considerations when using the EquityTool

Like any measure, the EquityTool has limitations which any user should consider. Measurement of wealth, or socio-economic status, can be done in other ways as well. Contact us for support.

Relative, not absolute, wealth: The wealth index is a measure of relative wealth rather than income, so the EquityTool will not tell you the proportion of respondents that live below specific absolute poverty lines. It should be noted that, in some countries, people in the middle or second highest quintile may still be poor in absolute terms, even though they are wealthier than most people in the country. A good example of this is Madagascar, where about 80% of people are thought to live below the World Bank poverty line of $1.90 per day. This means that people in the bottom four quintiles are living below the World Bank poverty line.

Reliability: the simplification of the wealth index is designed to ensure that the EquityTool is quick and easy to implement. The simplified wealth index the EquityTool uses agrees to the original data in terms of assigning people to the bottom two quintiles, the middle quintile or the top two quintiles with a kappa statistic of 0.75. In other words, the EquityTool will usually put people into the same quintile that the original DHS wealth index would have, but it is possible that it will put some people into a different quintile. If this happens, it is almost always one quintile above or below what the original wealth index would have put them in. Please see specific country fact sheets for more details.

Age of surveys: the wealth index is based on asset ownership and household characteristics. These things can change over time. For example, it may be that only the top quintile owned refrigerators in 2010, but in 2015 the top two quintiles own refrigerators. We are prioritizing surveys from DHS phase VI or later (in other words, surveys from 2010 and later). We are currently analyzing the effect of using surveys that are up to 5 years old for measuring relative wealth.

Reference population: the wealth index is based on DHS and other national surveys. It tells you how wealthy a person is relative to the rest of the country, or relative to the urban population in the country. If you would like to compare your respondents to a specific sub-group in the country, such as the population of a specific region, then you may need to run an analysis on the DHS dataset to create a scoring system for that specific group. If you would like guidance on how to do this yourself, or assistance with doing analyses, please let us know via our contact page or support [at] equitytool.org.