The main focus of this paper is on the relative merits of macro and micro data in providing realistic and reliable information on the economic activities of women in rural Java. The two types of data would seem to have a symbiotic relationship: large-scale surveys tend to provide a simplified view of complex realities, and village studies can demonstrate where over-simplification occurs, thus acting as an impulse to the generation of increasingly more useful census and survey data. On the other hand, census data can provide the framework for assessing the degree of representativeness of village studies. The two, in interaction, should be able to generate increasingly more useful large-scale data, with village studies continuing to 'flesh out' the framework provided by census and survey data. Such a framework is provided in the latter part of the paper where 1971 census data are used to investigate the extent of information available on the economic activities of women in rural Java. Limitations of these data are considered, and some suggestions regarding areas where improved data are necessary are made.