Transforming behavioural measurements for evaluation through new methods

The Ipsos Africa Centre for Development Research and Evaluation, specialises in Monitoring, Evaluation Research and Learning for development organizations, and comprises 40 professional researchers and evaluators based all across Sub-Saharan Africa, with regional management based in Kenya and Uganda. Our goal is to be the leading Africa-led evaluation team, and to help reach the SDGs by solving challenges for our partners and clients in the development sector using evidence. Monitoring and evaluation work is a big part of what we do, and is critical for learning about how to make change happen.

One area which is particularly challenging, and often very central, is measurement and understanding behaviour; this is notoriously difficult to accurately do using conventional recall-based research methods.

As humans we often simply don’t remember precisely what we did, we rationalize it and why we did it after the fact, we overclaim or under-claim our behaviours depending on social acceptability, we don’t want to talk about certain behaviours, or our behaviour is so ingrained in us that we are not even aware we are doing it. So, you can see why it’s hard to measure by asking us!.

That’s why we incorporate a range of methods, techniques, and data sources in our evaluation work. We have a huge range of potential tools in Ipsos, from the deep personalized approaches to span the gap, like participant observation through to use of anonymized aggregated mobile data to track movements. For example, on the latter, it is possible to track waiting times in health clinics over time using this data, based identifying which phones are not usually stationed at the clinic, then measuring average lengths of time for those who come and go.

A huge number of fascinating and very illuminating alternative data sources are coming together to help us understand ourselves better, and combined with in-depth and most importantly, participatory methods, those mean we have a very powerful evaluative tool box.