By Alberto Zezza, World Bank
On the occasion of the 45th Session of the United Nations Statistical Commission, FAO and the World Bank organized, with support from the Global Strategy, a pre-conference event and a seminar aimed at launching an international effort to improve the quality of food consumption data in household surveys.
An assessment carried out by the World Bank, FAO, and the International Household Survey Network (IHSN) on 100 household surveys world-wide, found great variety across surveys in data collection methods and pointed to many areas where survey design and questionnaires could be improved. The primary objective of these surveys is usually to measure poverty, derive consumption patterns needed for the calculation of consumer price indices, or provide input for the compilation of national accounts.
These data are increasingly being re-purposed and used to calculate food security indicators, compile food balance sheets, plan and monitor food-based nutrition interventions, serve the information needs of the private sector, and for other research activities. What makes a survey dataset “fit for purpose” is specific to each of these particular uses. Small improvements can sometimes lead to a significant increase in reliability and thus to great improvements in measurement accuracy.
A number of UNSC members (including Ethiopia, Indonesia, Mongolia, and Peru and several institutional members) have expressed interest in playing an active role in a proposed medium-term program of work which will entail validating different survey design options to evaluate their impact on data quality. Work on two pilot tests has already started in Indonesia (BPS) and Peru (INEI), both with technical and financial support from the Global Strategy and the Living Standard Measurement Study survey team of the World Bank. The ultimate objective of this work program is to develop, validate and promote scalable standards for the measurement of food consumption in household surveys.