Skip to main content

Community-based versus data- driven targeting: Experimental evidence from the DRC

https://docs.wfp.org/api/documents/WFP-0000177053/download/
This impact evaluation examined how two targeting approaches – one predominantly data-driven and the other community-based – performed when selecting households. The study was designed and implemented by WFP’s Office of Evaluation in partnership with the World Bank’s Development Impact Department between 2023 and 2024, with support from WFP’s Country Office in the DRC.

The DRC faces one of the world’s most severe and protracted humanitarian crises. As needs increase and funding declines, WFP is under growing pressure to sustain its humanitarian response by targeting assistance effectively and equitably. While data-driven methods based on a proxy-means test (PMT) are considered the standard for targeting in development contexts, they are often difficult to implement in humanitarian settings like the DRC, where data collection is challenging and the security situation is volatile.

The evaluation found that a simpler, community-driven approach achieved comparable precision and welfare outcomes to a more advanced data-driven targeting method at a much lower cost. Although community-based targeting selected fewer households for assistance, it achieved similar food security and welfare outcomes to PMT Plus for the target population (including beneficiaries and non-beneficiaries).

At the same time, community-based targeting's lower coverage rate resulted in significant programme savings. This indicates that there may be an “optimal coverage level” for a community receiving humanitarian aid. As most targeting methods assess vulnerability at the household level, they often overlook community dynamics such as the sharing of food and resources.