112 SOCIAL SCIENTIST participation rate ranges from a high of 50 per cent for Thailand to a fow of 6 per cent for Bangladesh. As a matter of fact. East and South-East Asia have relatively higher rates of participation than South Asia, although India (20 per cent) is relatively better off than Bangladesh and Pakistan (6 per cent). One reason for this variation could be the accepted problem of conceptualisation and measurement, due mainly to a gender bias of the data collection agencies which tend to overlook women's contribution to economic activity. In India the CSO and the NSS have their specific deviations from international standards in enumerating women's work. Whereas it is relatively simpler to apply international standards to those who are employed in market production, identification and measurement in the non-market producing areas in most developing countries adds to the difficulty, since such production is located within the household. There is also the general problem of cultural perception. Rigid definition of gender roles is prevalent. Domestic chores are a part of the housewife's domain and are excluded from the purview of gainful employment. The CSO and NSS subscribe to such rigid definition of gender roles, defining as non^work that activity of women that forms a part of their domestic work. Again, they use a relatively short reference period, ignoring seasonality, which results in under-enumeration of women. The CSO and the NSS have recently started to use the concept of marginal worker to net such participation, although the conceptual bias in such mutually exclusive terms such as main and marginal is obvious, since women generally come under the marginal category. The ILO has worked out an alternative methbdology based on an experimental household survey, where all processing household activities are included whether or not they were performed by the producer household (Anker, 1988). However, the studies in this volume use the standard methodology. In contrast, the experience of japan and South Korea indicate that output and employment levels in the household sector are superseded by those in the secondary and tertiary sectors with the development of the labour market. There are also distinct changes in the female work force in terms of age, status, industrial distribution and improved skills. It is interesting to note that whereas in totality there was only a marginal increase in the strength of the women workers, there was a transformation of their activity from the primary to the secondary and tertiary sectors, resulting in improved wages, health and education. Equally useful is the overview of the female labour force in India. CSO and NSS data indicate a concentration of women in agriculture; a concentration of women in low income occupations like agricultural labour, construction and domestic work. What is interesting is the reminder that this was not always so, in comparing the participation rate from 1900 to 1950. One reason is perhaps the process of de-industrialisation in the rural areas under the colonial regime, which