Measuring food security using respondents' perception of food consumption adequacy

Food security is a complex and multi-dimensional phenomenon. As such, its measurement may entail and benefit from the combination of both 'qualitative-subjective' and 'quantitative-objective' indicators. Yet, the evidence on the external validity of subjective-type information is scarce, especially using representative household surveys. The aim of this paper is to compare information on self-perceived food consumption adequacy from the subjective modules of household surveys with standard quantitative indicators, namely calorie consumption, dietary diversity and anthropometry. Datasets from four countries are analyzed: Albania, Madagascar, Nepal and Indonesia. Simple descriptive statistics, correlation coefficients, contingency tables and multivariate regression show that the 'subjective' indicator is at best poorly correlated with standard quantitative indicators. The paper concludes that while subjective food adequacy indicators may provide insight on the vulnerability dimension of food insecurity, they are too blunt an indicator for food insecurity targeting. An effort towards developing improved subjective food security modules that are contextually sensitive should go hand in hand with research into how to improve household survey data for food security measurement along other dimensions of the phenomenon, particularly calorie consumption.