From 1994 to 1998, a total of 73 rainfed lowland rice fields (RLR) (>0.3 ha each) in Cambodia were studied to determine which pests affect yields and how cropping practices affect pest levels. From 1997 to 2000, this information was used to generate testable hypotheses for crop protection research and to help prioritize the integrated pest management (IPM) research program of CIAP. The data collection methods of Savary et al (1996) were adapted to Cambodian conditions. "Pests" included insects, weeds, and diseases. Farmer interviews (Jahn et al 1997b), pest collections, and practical considerations (e.g., ease of recognizing the pest or its damage) were used to determine which pests to include in the study. Pesticides were not used in any of the fields in this study.
Data were gathered at four crop development stages: tillering, booting, milk, and maturity. The yield of each field was estimated by averaging the weights of three randomly selected 2 x 5-m2 harvested areas, converted to t ha-1 and adjusted to 14% moisture. Pest incidence was recorded from 10 hills chosen haphazardly from each field. Weed infestation was measured as the percentage weed cover in three 1-m2 areas that included sampling hill number 3, 6, and 9. The analysis proceeded in five steps: (1) determination of average injury levels of each pest for each crop stage, (2) categorization of variables (i.e., pest, cropping practices, and yield) into classes, (3) testing for independence of paired variables (i.e., pest levels, cropping practices, and yields) in contingency tables, (4) clustering of cropping practices, yields, and pest profiles, and (5) development of contingency tables and correspondence analysis. Data were analyzed with Excel and STAT-ITCF.