摘要：This paper reports on two methods used for identifying alternative indicators of chronic and acute food insecurity. A need for alternative indicators exists since many of the "benchmark" or "gold standard" indicators (such as household income or dietary intake) are too cumbersome to be of practical use in food aid targeting. The ideal alternative indicator should be statistically reliable, yet straightforward to collect and analyze. The study uses data collected in four villages in the Indian Semi-Arid Tropics to illustrate two methods for identifying the alternative indicators. A qualitative methodology included ethnographic case studies of at-risk households, participatory mapping of vulnerable households within a community, food charts, and seasonality charts. The quantitative methods included both economic and nutrition surveys. The data were collected over three rounds in 1992-93 from 324 households in south-central India. For the qualitative work, we used both the villagers' perceptions of food insecurity as well as the ethnographers' observations to generate a list of indicators for these areas. Triangulation among the various qualitative methods was used to validate the indicators suggested. For the quantitative study, we used statistical methods to test the strength of association between each indicator and six benchmark measurements of food security. The benchmark measurements were derived from dietary recall, anthropometric, and blood data. The dietary data were used to generate a benchmark for chronic and acute households' food insecurity. The anthropometric data were used to construct benchmarks of chronic and acute preschooler food insecurity. Finally, serum measures of vitamin A and iron adequacy were used to generate benchmarks of household micronutrient insecurity. We tested a core set of alternative indicators against each of these benchmarks. The majority of the alternative indicators were drawn from a review of the food security literature as well our own qualitative work in the study sites. Other indicators were included as they represent information that is typically available in secondary data sets collected by governments and research institutions.
