Alba, AlfonsoLópez-Mourelo, ElvaUniversidad Carlos III de Madrid. Departamento de Economía2017-11-222017-11-222017-11-012340-5031https://hdl.handle.net/10016/25937We use a sample of social security records containing work histories and sick leave episodes to investigate gender differences in the incidence and duration of absence from work due to sickness in Spain. For sick leave incidence we apply a competing risk model to a panel of newly employed workers who can be followed for two years until an episode of sick leave occurs or the job ends. For the duration of sick leave spells, we estimate a Weibull model. We distinguish between sick leave due to occupational illness or injury and sick leave due to common disease or accident. This distinction is important because only for the latter women have higher incidence and longer duration than men. In this respect, the presence of children under 3 years of age in the household becomes a significant explanatory factor.application/pdfengAtribución-NoComercial-SinDerivadas 3.0 EspañaSick leaveGender differencesCumulative incidenceCompeting risksProportional hazard modelworking paperJ14J28J81DT/0000001594