Several articles have examined the status of women in developing countries. But while numerous studies have attempted to estimate the extent of wage discrimination against women in the west, few such studies have been conducted for Third World cOWltries. The only such published study for Pakistan appears to be Ashraf and Ashrafs (1993) examination of the gender earnings differential for Rawalpindi City. Given the very restricted sample (based on just one city in Pakistan) and the relatively dated data (from 1975) used in that study, a need clearly exists for a more comprehensive examination of earnings by gender. This article represents the first intensive look into male-female earnings differentials, not only for Pakistan as a whole, but for each of its four provinces individually as well. Gender earnings differentials have also been calculated for a number of industrial subgroups. Computations have been made for the years 1979 and 1985-86.1 This allows the trend in the gender earnings gap to be tracked over that period. Data from the Household Income and Expenditure Surveys are used in the estimations for both the Oaxaca (1973) model as well as the more recently developed Cotton (1988) and Neumark (1988) models. A new twist is the incorporation of correction for selectivity bias in the sample data which strengthens the methodological underpinnings of the model.