This paper studies in two organizations, one public and the other private, using the cross-sorting method, the association between socio demographic variables, organizational variables and other variables related to the work, which are independent variables and the absenteeism referred to as dependent variable.

The test of the hypotheses formulated in our model shows that the majority of the hypotheses are validated in the public structure, whereas the hypotheses relating to age, hierarchical level and job satisfaction are not validated in the private structure, which constitute  paradoxical

résults.