Considering food consumption modules in survey design
A new paper co-authored by EDI’s Co-Founder Joachim De Weerdt along with Jed Friedman, Kathleen Beegle and John Gibson addresses food consumption data, measurements, response error, and how findings from a randomised survey experiment in Tanzania can influence survey design. This most recent paper builds on Joachim’s previous work and relates to EDI’s experience in tackling non-standard units in consumption modules.
A recent post from the World Bank’s Impact Evaluations blog series focusses on measurement and is appropriately titled ‘What do we measure when we measure food consumption?’ The post highlights a number of papers recently published in a special issue of Food Policy entitled “Food Counts” and helpfully draws on some key challenges to be considered in survey design. EDI’s researchers can relate to these challenges, such as the impact of diary accounts versus memory recall, individual consumption versus household consumption and how broadening food group categories in surveys can significantly limit analysis.
At EDI, we are continually improving survey design. The developments in surveybe capability and function means that we constantly push the boundaries of how data is collected. Together our Research Team and Technical Team continue to evolve modules, such as food consumption modules, so that the end user of any dataset can be confident in the quality of findings and subsequent work published is accurate. After all, quality is what underpins our core values.
Friedman, J., Beegle, K., De Weerdt, J., Gibson, J. 2017. “Decomposing response error in food consumption measurement: Implications for survey design from a randomized survey experiment in Tanzania” Food Policy Vol 72, 94-111