COMMUNITY SYSTEMS FOUNDATION
Despite agreement that evaluation is integral to development programs, substantial diversity exists on the best methods of evaluating nutri- tion interventions.
Sahn, David E.|Pestronk, Robert M. · 1980

Abstract
This report, based on the present practices of several multilateral, bilateral, and private donors, reviews these methods. The authors propose standardized evaluation terms and concepts for examining nutrition education, supplementary feeding, food fortification, multisectoral, and planning programs. Presented are two models of nutrition programming -- A.I.D.'s logframe, outlining a causal progression of events at the input, output, purpose, and goal levels; and the Nutrient Flow Model, depicting nutrition as a function of activities such as food production, trade, loss, and distribution. Linked to each of these models, respectively, are process and impact evaluations. For process evaluations (often conducted along with impact evaluations), information is collected on pre-existing indi- cators, project components, and intervening events in order to reveal nutritional improvements attributable to a program. Impact evalua- tions seek to determine improvements in overall nutritional status. Various impact evaluation designs (e.g., situation assessment, before/ after, recurrent institution cycle, comparison groups) and threats to their internal and external validity are reviewed. The authors also discuss alternative evaluation methodologies which combine aspects of both evaluation types to assess such program aspects as effort, adequacy, appropriateness, etc. They also differentiate operational program evaluation (analysis of a program's specific impact and performance), the subject of this report, from basic research (knowledge gained using scientific methods and controlled experi- mentation) and evaluative research (field testing applications of existing knowledge). Anthropometric, biochemical, clinical, and dietary nutritional measures are examined in terms of objectivity, coverage, directness, completeness, and precision. Also presented are 70 abstracts of published evaluations, emphasizing methodology over outcomes, and brief descriptions of the programs and approaches of several donors involved in nutrition programs. A 171-item bib- liography (1955-80) is appended.
Connected topics
Classification
USAID DEC