“Can you drive a wheel? Can you drive a door?” As pictures of pieces of a car were passed around the room, these are the questions our trainers asked. Of course, the response was a resounding “No.” Just so, the trainers explained, “The PPI is like a car, you can only drive it if you have all the parts in place.”
The most common questions during training are usually around specific PPI indicators and how they are chosen, why they are chosen, and if they can be changed. However, the PPI is not just a compilation of random questions; each question is carefully chosen through a statistical logit regression process, based upon the national survey and the correlating strength of the questions, to determine poverty likelihood. PPI trainers use exercises like the one about the car to show how those indicators are chosen, Trainees learn that the PPI is the sum of its parts, not to be broken apart. This said, it is common practice in the PPI development process and it is absolutely vital that we obtain input from institutions working on the ground to determine if there are large concerns with any of the indicators and, if so, to consider putting in a different indicator that is statistically relevant.
These exercises go a long way. They help assure participants that this is an objective process based on real data that drives the indicator selection process: “It’s not a person, it’s the machine that does the math.”
To learn more about how the PPI is developed visit our FAQ section!
Preeti Wali is Communications Officer at the Grameen Foundation Social Performance Management Center (SPMC). She is based in Washington, DC.