Data collection in the Democratic Republic of Congo can be daunting. We lack data for much of a country that is 2/3 the size of Western Europe. See the map below with the big grey space- that’s an area where we have no data and it’s probably about the size of France. The last census was in 1984 and a proposal for the next census has landed in the middle of a serious political controversy.
But things are moving- especially on the mobile network side. Network coverage is expanding- a private sector source told us that his company is building 1,000 cell phone towers per year. There are four major cell phone companies operating in DRC, and all territories now have at least some coverage. Mobile penetration remains low but is expanding- currently around 35% of people are estimated to have phones. We started our first mVAM pilot in DRC- in Goma collecting data from Mugunga III IDP camp (blog, video) and market prices in Goma (blog).
CAID- A promising government collaboration
But we thought, DRC is also an exciting case for national mobile data collection. It turns out a new, young, dynamic government team under the Prime Minister’s office thinks so too. The Cellule d’Analyses des Indicateurs du Développement (CAID), or Center for the Analysis of Development Indicators has the atmosphere of a fast-paced Congolese start up. And innovation is necessary if CAID is going to fulfill its crucial mission: addressing DRC’s vast data collection challenges and ensuring the various data that is collected is available openly on a central website.
For the CAID, the first priority is filling in that grey emptiness. They set up an impressive data collection system, recruiting agents in each of the 145 territories to collect data. And they were already using mobile technology! As their agents went out to collect data, they often used an SMS system to send back results.
So for us at WFP, working with CAID was a perfect match. The WFP Country Office in DRC has made partnering with the government to enhance technical capacity a priority. So we at WFP were thrilled to meet with CAID early this year and brainstorm ideas of how remote mobile data collection could support their efforts to put in place a data collection system covering the whole country. Their agents have lots of work as they try to collect data for all the different development sectors for their territories- and territories are huge.
Designing mKengela- a national mobile monitoring system
As we brainstormed, the idea for a national mobile monitoring system, mKengela or mobile “alert” in Lingala, started to take shape. To lighten their workload and ensure a steady flow of data, CAID liked the proposal to monitor market prices and gather household information by phone calls. At WFP, we thought great- we can monitor 145 markets or one per territory throughout the country. But CAID sent their agents out to collect the phone numbers from 1350 traders in 435 markets! CAID is ambitious and ready to try new things, including 3 markets per territory to get a more meaningful national price monitoring system.
If you read our blog on collecting market prices in Goma, you’ll remember that units of measurement can make or break your price data collection. If you don’t ask people in local units, you won’t get good information. But you also then have the difficult task of converting various local units to the metric system. Well CAID solved this problem. Max, the statistician at CAID, looked through all the price data that CAID had. Their agents had collected prices in local units from traders and then weighed the amount in kg or liters. So Max had the list of almost all local units used and the corresponding metric conversion. When there was some variation in the metric weight for a local unit, Max took the mean weight. Voilà- about the best solution to the local units of measurement challenge that you could ask for.
mKengela will also have a household survey component. We are planning on collecting information on food security indicators- household consumption and coping- from 5,200 households in 26 provinces. For the moment, the network and cost constraints make a household survey at the territory level a little out of reach. So we plan to pilot the survey at the province level first.
Strong Partners, Strong System
This of course would not be possible if we didn’t have tech-oriented donors and an excellent call center to work with. We received funding from USAID and the Belgian Development Cooperation for scaling up mobile data collection in DRC. In terms of a call center partner, we have struggled to find a good private sector partner in some countries. But in Kinshasa, Congo Call Center, created by two entrepreneurial women, is extremely professional and experienced. Their operators asked us thoughtful questions during the training, helping us further improve the questionnaires.
So momentum is growing around mKengela in DRC. We’re excited about this new partnership with CAID and all the opportunities that come from setting up a national monitoring system. We’re planning to keep partnering with CAID- next priority, working together to improve data management.