The purpose of the study is to examine the variation in knowledge and usage of contraceptive methods across Bangladesh villages. The main hypothesis is that the variation can be explained by three sets of factors measured at the village-level : development programmes, family planning programme efforts, and given environmental and socio-economic conditions. Data are drawn from the Bangladesh Fertility Survey and the 1974 Bangladesh Population Census. The three sets of factors taken together explain a greater proportion of the variance in knowledge and usage of contraceptive methods than each of the sets taken singly or in paired combination. Knowledge of clinical contraceptive methods is found to be affected more by development programmes than by either family planning or environmental and socio-economic conditions. Knowledge of non-clinical contraceptive methods, on the other hand, is affected more by given environmental and socio-economic conditions. While both development and family planning variables have independent and approximately equal effects on ever use of contraception, each of them separately is not likely to produce as much contraceptive usage as would both of them jointly. The policy implication of this finding is that if both development and family planning programmes are provided to the villages, the impact on fertility may be maximized.