Estimation of Prevalence of Food Insecurity in Sri Lanka

Sri Lankan Journal of Applied Statistics, Vol (14-1)

Information on populations that are food insecure at national and regional level, and changes in food security over time, are important in formulating policy and development interventions for food security. Sri Lankas’s food security assessments have been based on composite indices constructed using variables that are direct or indirect determinants of food security. Such composite indices help compare relative food insecurity across geographic areas, and over time, but limit the assessment of the magnitude of the prevalence food insecurity. In this study using deprivation of food energy intake as an indicator of food insecurity, prevalence of food insecurity is measured as the food insecure proportion of populations in specific geographic areas at specific points in time. The probability density function of per capita food dietary energy intake is assumed to follow a lognormal distribution and prevalence of food insecurity is estimated as the proportion of the population whose per capita dietary energy intake per day is unsatisfied in relation to a recommended minimum level. The method was applied to primary data from the National Household Income and Expenditure Surveys conducted in the years 2006-2007 and 2009-2010. The derived food insecurity estimates were validated comparing with official estimates of expenditure poverty, nutritional outcomes and also with the findings from food security assessments based on composite indices, available in the literature. The analysis of prevalence estimates reveal insights on regional differences in food insecurity and changes over time, yielding important implications for policy.