According to the figures from the latest round of the European Community Household Panel survey carried out in France in autumn 2001, in the preceding four weeks one employee in ten was absent from work for at least one day on sick leave or for other reasons. This proportion barely changed between 1994 and 2001. Slightly more women than men were absent from work, which applied for overall absences as well as for sick leave alone. However, in terms of the number of bouts of illness or the total number of days' sick leave taken over the last twelve months, there was hardly any difference between men and women. Regression analysis using cross-sectional (2001) and longitudinal data (1998-2001) has been carried out to assess the effects of individual factors on the probability of being absent from work, looking at men and women separately. Three factors play a decisive role: health, level of job satisfaction and, for women, the constraints of reconciling family and work. For men, a negative relation can be observed between the estimated hourly wage and the probability of taking sick leave, at one time or another, over the last twelve months. For women, on the other hand, the probability of being absent hardly seems to be related to their level of pay.