Impact evaluation in Eastern DR Congo.

We are undertaking a randomized control trail to study the impact of Tuungane – a large community-driven development (CDD) project implemented by the IRC and CARE International in the Eastern Democratic Republic of Congo. The coverage area is the size of France, but then without any roads, bridges, etc. In addition to using public lotteries to generate a treatment and control group of around 2,500 villages each, we are using a randomized intervention to examine the impacts of gender quotas. The project has a population coverage of approximately 1.8 million and takes places between 2006 and 2011.

The baseline survey took place in 2006/7 in over 600 villages (see baseline survey and report). Almost a hundred enumerators will be conducting the final part of the evaluation from mid-2010 to end-2011. The goal of the CDD project is to improve good governance. Because it is unlikely that surveys will give reliable data, we introduce a behavioral measures in 560 villages. We implement an unconditional cash transfer project (called RAPID) and then observe whether Tuungane treatment communities behave different than control communities. All the material can be found here.

The preparations for the final part of the survey have taken more than two years. I’ve visited the Congo four times since the summer of 2009 and lived there for more than a year.  This preparation included, among others: writing do-files to clean databases, visiting Congolese villages and conduct focus groups to check our final measures, training people and pilot draft surveys, writing the design documents, training almost a hundred enumerators, organizing equipment, and joining the teams in the field to check the quality of implementation.

One of the most interesting parts of this project is the fact that one at times has to wear different hats. One has to be a diplomat with local partners and governments. A teacher for the almost hundred enumerators. A manager for all those different teams that have to keep to a schedule and that have (a lot of) equipment needs. A salesman at times to sell a new part of the survey to implementing partners. And, of course, a scientist to think through and design the surveys and related material.



Conditions for the teams in the DRC are tough. Infrastructure is bad to non-existent. Photo 1 looks impressive but at least it is a road where a 4×4 can be used. Most villages can only be reached by motorbike or by foot. Source: Friend at IRC. Because of continued fighting between government soldiers, MONUSCO and rebel groups, security is also a big problem. Photo 2 is me in Sud Kivu. Source: Own camera. For the behavioral measure 560 villages will be visited four times. Photo 3 is an enumerator in the province Maniema introducing project RAPID – Recherche-Action à travers un Projet d’Impact dans le Développement. The evaluation is technology-heavy with a 100+ PDAs, 100+ solarchargers, satellite phones, digital cameras, GPS-devices, and more than 10 laptops to synchronize the PDA data. This not only makes us print less paper, it also allows us to verify the collected data almost in real-time and be in contact with the teams – even if they are not in coverage areas. Source: Project cameras.