Use of government surveys has limitations. Reliance on the government’s nationally representative EMICoV (Enquête Modulaire Intégrée sur les Conditions de Vie de Ménages) survey led to issues with data quality and sampling. First, randomization was constrained by the initial EMICoV sampling frame and the related expansion of EMICoV 2006 sample to additional villages. Data from some survey villages had to be dropped due to treatment villages with no control groups in the same commune and 20 treatment villages which could not be located in the EMICoV sample due to mismatch in village names. There was also a limited set of land and agricultural questions in the EMICoV, which required the World Bank to use 2011 World Bank survey estimates instead of panel estimates using the EMICoV data. Collection of detailed land and agricultural data was also challenging for the data collection firm, including elements like GPS plot measurements as the survey was not focused on land and agriculture. There were also data quality issues from data entry by the Benin National Statistics Institute which administered the survey, INSAE (Institut National de la Statistique et de l’Analyse Economique). In the end the World Bank did not utilize the EMICoV data from 2006 or 2010 and instead collected two rounds of follow-up data directly. MCC has faced similar issues in government run surveys in the past as well as having issues obtaining that data. Close coordination or oversight is necessary and not solely paying for additional sampling or modules.
Official websites use .gov
A
.gov website belongs to an official government
organization in the United States.
Secure .gov websites use HTTPS
A
lock (
) or https:// means you’ve safely connected to
the .gov website. Share sensitive information only on official,
secure websites.