RESUMEN
PIP: Data from a Sri Lanka national sample survey -- 3597 households stratified on the basis of development program areas -- were analyzed to compare impacts of 3 national development programs and their combinations upon the occupational and income status of females and males in Sri Lanka. These programs, implemented over the last 30 years, are guaranteed price schemes that develop markets for agricultural produce, land settlement schemes that include irrigation, and rural electrification. To date, no attempt has been made to assess the gender-specific socioeconomic impacts of these individual programs and their combinations. It was hypothesized that the utilization of development program outputs will exert a gender-differential impact upon occupational and income status, but the magnitude and direction of the impacts remain to be determined. Path analysis was applied to estimate the model for each development program and their mixes for males and females separated. A multistage stratified sampling design was utilized. All of the development programs and their mixes exhibited significant effect of educational attainment upon participation in nonagricultural occupations. Rural electrification (RE) was the only program whose effect was positive; in combinations with education it accounted for 15% of the variation in occupation. Among the programs that were negatively related to male participation in nonagricultural occupations, the most important predictors were the land settlement (LS) and guarantee price scheme (GPS) programs. Each program contributed to over 1/5 of the variation in occupation net of educational attainment. RE was the only program that was not significantly related to female participation in nonhousehold occupations. All of the remaining programs exerted a positive effect upon occupation. 3 of these programs -- RE + LS, GPS, and LS + GPS -- were of almost equally high importance in predicting participation of females in nonhousehold occupations, and in combination with female education accounted for a little over 20% of the variation. All of the development programs with the exception of GPS exhibited a significant positive relationship with female income. Among males, all programs with the exception of GPS and LS exhibited significant indirect effects upon income. The magnitude of total development program effects upon income was gender specific in Sri Lanka.^ieng