Food Insecurity Measures: Experience-Based versus Nutrition-Based Evidence from India, Bangladesh, and Ethiopia

Using data from three household surveys in Bangladesh, Ethiopia, and India, this report compares two commonly used measures of household food insecurity: a measure of caloric consumption (reported undernourishment) and an experiential measure. In the second measure, a single affirmative response to whether the household experienced certain conditions or behaviors due to insufficient food designates a household as food-insecure (experiential-based measure). The authors demonstrate that a significant share of households categorized as undernourished because their caloric consumption is below 2,100 calories do not report experiencing any form of food insecurity. This finding is robust across different experiential food security metrics and different contexts. For India, which used a single indicator of experienced food insecurity, the experiential measure had the least overlap with the caloric consumption measure compared with the measure used in the other two country case studies. Although the measure from the Ethiopian survey, which contained nine experiential questions, had the most overlap with the caloric consumption measure of food insecurity compared with the measures used in India and Bangladesh, there was still substantial misclassification of food security status among households. These findings suggest that even if the overall prevalence of food security is similar when estimated with experiential and other measures of food security, experiential measures appear to be classifying a different subset of the population as food-insecure.