In this paper we contrast two approaches to the measurement of women’s work applied to the same population of ever-married women. These women were interviewed on two separate occasions – first during the 2003 Interim Egypt DHS, and again during the Slow Fertility Transition (SFT) survey conducted in 2004. The DHS uses a standard keyword question to measure work, while the SFT employs an activities list question format. We argue that the widely-used keyword approaches to measuring women’s work underestimate the level of female labor force activity, as shown by the DHS-SFT comparison and supported by analysis of data from the 1998 Egyptian Labor Market Survey. We demonstrate that the activities list approach captures a wider range of economic activities among women, while
allowing us to document multiple jobs held simultaneously by respondents. Furthermore, we find that keyword questions disproportionately exclude poor and poorly educated working women from the labor force. Survey approaches to the measurement of women’s work must be revised if we are to fully account for women’s contributions to family welfare and national accounts, and if we are to understand the relationship of work to other variables and processes of interest to social scientists