Search

The aim of this study is to put forward an approach designed to calculate and sum up discrepancies between the actual food acquisition or intake and any standard or norm. Based on secondary analysis of cross-sectional data from the Mexican National Survey of Household Income and Expenditure, our proposed method to produce classes of entropy-based Diet Adherence Indices generates a Food Basket Adherence Index (FBAI) for Mexico City (2129 households). Findings suggest that it is possible to measure and decompose diet adherence using a cross entropy measure. Using food expenditure data and a normative food basket for Mexico City results, we show households' deviations from the suggested norm for different food groups. The average FBAI was 0.44, far above the minimum score of 0 which would indicate full adherence to the normative food basket. Our measure has a distribution wide enough to detect meaningful changes and distinguish between groups with known differences, providing important new insights on the linkages between home food environments and income distribution, and food insecurity and household distribution. [ABSTRACT FROM AUTHOR]

The Latin American and Caribbean Food Security Scale (ELCSA) is widely used in the region to classify population groups with presumably distinctive levels (mild, moderate or severe) of food (in)security. This paper puts under scrutiny the validity and utility of the resulting population groups using data from the Mexican National Survey of Household Income and Expenditure 2018. The analyses use scalogram analysis, Item Response Theory and Latent Class Analysis to look at the properties of the groups derived from the ELCSA. The paper shows that ELCSA's capacity to actually distinguish between four levels of food insecurity is rather weak. The analyses suggest the ELCSA reliably captures severe food insecurity scores but not milder forms of food deprivation. A key objective of the ELCSA is to identify different ranges/groups of food insecurity, and the main implication of our findings is that it is time for a thorough makeover of the scale. [ABSTRACT FROM AUTHOR]