This field study of employees (A' = 1,292) of a public utility company had several objectives: (a) to identify that portion of employee absenteeism that is avoidable; (b) to compare this measure of avoidable absenteeism with more traditional measures, and (c) to examine the relationship of the measure of avoidable absenteeism with a selected absence policy-a waiting-time provision whereby employees were compensated differently for absenteeism. The avoidable absenteeism measure was more robustly related to predictor variables than were more traditional absence measurements. The results indicate that absenteeism policy is a far better predictor of avoidable absenteeism than any individual or attitudinal factor singly in or combination. Also, a relatively small percentage of employees were responsible for the vast majority of avoidable absenteeism.In this research, we compared the variance in several criterion absence indexes-including a measure developed here, which we refer to as avoidable absence-explained by a number of predictor variables. In addition, we examined the explanatory power of an absence-policy provision relative to a series of demographic and attitudinal predictor variables. We first present a brief overview of absenteeism as an organizational concern and then discuss organizational policy as a robust predictor of actual absenteeism and measurement issues in absenteeism research.