Blog entry originally posted in November 2013 on the Humanitarian Innovation Fund website.
In this month’s entry we explore the results of the face-to-face survey in the eastern Democratic Republic of Congo, and provide an update on the process of getting our call centres set up.
Earlier this month, our colleagues in the Democratic Republic of Congo finished the face-to-face assessment that serves as the mVAM baseline. These face-to-face assessments allow us to understand the profiles of our respondents, and request permission to call prospective survey respondents in the future. The assessment was carried out in a camp hosting internally displaced persons near Goma. A total of 333 households were visited. The methodology followed standard WFP guidelines for camp settings.
In addition to the core set of questions on household demographics, food consumption, and coping strategies, the respondents were asked if they would like to participate in the monthly mVAM voice surveys. Some 90% (300 households), expressed their willingness to receive phone calls from WFP. This is the same proportion of households that agreed to participate in central Somalia, the other location where this survey method will be piloted.
According to the GSMA’s report on mobile telephony, unique mobile phone penetration is below 20% in the Democratic Republic of Congo – well below the 30% average for Sub Saharan Africa, and below Somalia.
The findings of the face-to-face survey show that only 24% of respondents in the camp near Goma own a mobile phone. Another 33% reported not having a phone, but that another member of the household owned one. These figures explain why basic mobile phones should be distributed to all who had signed up for the mVAM surveys in Goma. Note that this is different to the approach in Somalia, where high mobile phone ownership rates make phone distribution unnecessary. During the monthly data collection rounds, it will be interesting to compare response rates under these two scenarios; how much of an issue does using WFP-provided phones become (re-charging, lost/stolen phones etc.). Further insights to the results from the face-to-face survey in the Democratic Republic of Congo follow below.
Households participating in the mVAM surveys have an equal number of female and male headed households, 50.3% and 49.7% respectively, with an average of 5.5 members in the household. The great majority (89%) of mVAM respondents have been displaced and in the camp since 2009, with 41% arriving in 2012 when conflict in North Kivu increased. Almost all (94%) of respondents said they had received food assistance during the past month.
As in Somalia, we are curious to know whether the statistics we will collect through voice calls reflect the situation of the population at large. One of the key indicators we will collect through our calls is the reduced coping strategies index (rCSI), a quick and simple indicator that reveals how households manage or cope with shortfalls in food consumption.
The mean rCSI for the households that volunteered to participate in mVAM was 24.57, compared to 25.42 for the general population. A difference in means test, shows that there is no statistically significant difference (p=0.49) between the two groups. The rCSI result of the mVAM respondents is also reflective of the overall IDP population that was surveyed, thanks to the high percentage of households that agreed to participate. We realise that it will be a challenge to ensure that all of these households respond to our calls.
A mean rCSI of almost 25 is high, in keeping with high rCSI values commonly observed during previous surveys in the same area. A high CSI indicates that people are using a lot of coping strategies to deal with the lack of food or resources to purchase food. The most frequent strategies used were consuming less preferred or less expensive foods, reducing the number of meals in a day and eating smaller quantities of food.
Expenditures on food and debt
The respondents’ main income sources are daily labour (25%), petty trade (15%) and agriculture (14%). A huge amount, 83% on average, of monthly household expenditure is spent on food. The amount is even higher than with the case of our mVAM survey respondents in Somalia, who spent an average of 76% of their monthly expenditure on food. At the time of the survey, 65% of the respondent households were in debt and 85% of households had accumulated debt in the past 12 months. In three times out of four, the reason for getting into debt was to purchase food.
Similar to the baseline in Somalia, this assessment showed that the prospective mVAM survey respondents generally owned very few assets, the most common ones being a hoe/axe/machete (42%), mobile phone (24%) and radio (14%). Only very few respondents owned productive assets such as chicken (8%) or goats (3%). It’s acknowledged that the distribution of mobile phones in asset-poor settings raises challenges – this was discussed with the community and WFP partners at the camp and will be monitored during the project.
Updates on other developments
This month we also conducted three training sessions, training a total of 14 WFP staff in headquarters, the Democratic Republic of Congo and Somalia. The training involves showing users how to use Verboice and set up a SIP channel to do calls. After about an hour, colleagues in Rome and in the field were able to place phone calls through a SIP channel, using Verboice. This confirms our earlier insights about the user friendliness of the software.
The recruitment of our call centre operators is still ongoing in both countries; operators should be on board by January. In the meantime over the next few weeks, we will be drafting the operator’s manual. The document will describe how to manage the process of placing live and automated calls. We’ve also just finished buying the modems and servers that we need to set up for the Interactive Voice Response system for both countries.