no image

Measuring changes to poverty levels over time

Johanna Ryan
• VisionFund International
• United Kingdom
• 09/19/17

To all users: I would be really interested to hear from users how they are measuring changes to poverty over time.  If there is no perfect solution yet, then can we collaborate to create something?  

-Johanna Ryan, Social Performance Director, VisionFund International

no image
This is a great question, Johanna. From my experience working with ultra poor families in graduation programs, it is difficult to measure changes to household poverty by using the PPI. At extreme levels of poverty, the change is PPI scores is very small to indicate any impact. Rather, I have used more holistic measures such as the MPI. I would be curious to hear what others have done.
Hello Rahul - Thanks for your feedback. Our experience is that indicators which are most sensitive to changes in poverty for the ultra-poor are often not the easiest for enumerators to verify. That is a trade-off which we constantly need to manage as we determine which indicators should be included in the tool. Our PPIs are built using a country's National Poverty Lines. Indicators in a PPI are therefore chosen such that they are most relevant to households whose per-capita /per-adult-equivalent daily consumption expenditure is "around" this line. We are however in the process of making a change to the PPI's methodology which we believe will enhance its ability to predict poverty (both at a point in time and over time), at lines lower or higher than this one. Response options for the same indicators will be allotted different points depending upon the poverty line that you plan to use. This is because the level of association of an indicator to poverty may not be the same across all poverty lines. What our change will mean in practice is that the "weight" of an indicator to the final PPI score will vary depending on the poverty line used. Our paper on the PPI's new methodology will be released later this year. We will also provide users with detailed advice on what this means in terms of changes to the way they are currently administering the PPI. Please stay tuned. Also, we intend to add an additional "lower" 2011 PPP poverty line - the $1.00/day line to all PPIs going forward.
no image
As you’ve noted, the question of measuring changes to poverty or wealth over time is a complicated one. Solutions are dependent upon the type of wealth measurement tool being used, and a careful consideration of a program’s goals. Notably, does the interest lie in assessing individual wealth movement or changes to a community? Second, over what period of time are changes being measured, and with what acceptable level of uncertainty? We know that wealth generally increases over time, and the goals of assessing the change in wealth are very relevant to determining the method used. If you have a fixed benchmark, such as the $1.90/day poverty line, then it would be possible to see changes in the population in relation to that line. However, you would not be able to see changes to individuals. If you were looking at a relative measure of wealth, then comparing respondents to an old benchmark population will lead to under-estimating the poverty rate in your survey. The best place to start is in considering what VisionFund’s goals are in assessing poverty changes, such as change in a cohort versus an individual. Then, one may develop a more detailed strategy to meet those goals. Dr. Nirali Chakraborty, Metrics for Management
no image
Hey Jonathan, We have developed an automated system of transferring responses from an Android form to our central MIS where it is processed and scores are generated. We're able to organize poverty levels and other information fairly quickly. We'll be happy to share our learnings with you. My email address is Regards, Jasim