The process of adapting and validating a perceived household food security scale in a poor community.

This paper describes the process of modifying and validating a hunger index developed in the United States by Wehler et al (1992). It is part of a research whose main objective is to develope and validate a simple method that measures both quantitative (food sufficiency) and qualitative (female self-perception) dimensions of household food insecurity. In a pilot study, the original instrument was modified from a 2 point 8 item to a 4 point 12 item scale. Precision measured with Alpha Chronbach's coeficient was high (0.871) suggesting consistency in the scale's items. The instrument was applied to a sample of 238 poor and very poor households in a peri-urban barrio of Caracas. To determine overall internal validity of the scale the relationship between possible economic, social and behavioral determinants and food security level measured with the scale, was analyzed. Construct validity of the scale was established with factor and principal components analysis. Finally, with multiple regression analysis evidence is presented for overall validity of the scale. Four determinants: predictors of food sufficency score, monthly income per capita, social class, and number of children in the household predict in the expected direction self-perceived food security level (R-2 = 0.343). Results suggest that this instrument, together with an abbreviated measure of food sufficiency, based on strategic foods may be a valid, precise, and simple method for identifying and monitoring households that suffer from some degree of food insecurity in poor urban communities.