Confirmatory factor analysis to validate a new measure of food insecurity: perceived and actual constructs

Abstract

Food crises in 2008 raised the need for more timely and effective policy instruments to monitor and fight food insecurity. FAO developed the Food Insecurity Experience Scale (FIES), based on the individual experience in quantity and quality of food. Compared to the traditional, macro-based indicators, FIES presents important advantages: FIES makes it possible to identify the characteristics of food insecure people and allows to analyse food insecurity also in rich and developed countries. The main purpose of the paper is to validate the FIES, in order to assure policymakers that the tool is what they need to design meaningful development efforts. In this study, the methodological properties of FIES were evaluated on a sample of 150 thousand people, all over the world. The analysis of frequency distributions, the measurement of internal consistency, as well as an exploratory factor analysis and a cluster analysis were conducted. A principal component analysis-multiple linear regression allowed to rate the effect of each item on each factor. A confirmatory factor analysis was applied to test the hypothesis that a relationship exists between the observed variables and their underlying latent constructs. External validity was evaluated by a micro-econometric analysis of FIES in relation to extreme poverty and other relevant socio-economic characteristic of the population. The results show that FIES presents a good level of reliability and internal consistency. However, two distinct latent constructs were identified and analysed: a subscale measuring ‘perceived’ aspects of food insecurity and a subscale related to ‘actually experienced’ activities.