The use of an accelerometric activity monitor as an exposure assessment tool in ergonomic studies was evaluated by examining the extent to which physical differences in grocery cashiers and general merchandise clerks were reflected in the recorded data. Five medium sized retail supermarkets and one large combination store were considered. Twenty two grocery cashiers and ten general merchandise clerks, ranging in age from 18 to 55 years, were evaluated. All were right handed and free of musculoskeletal complaints, and all had at least 6 months experience in their current job. A unit, consisting of a piezoceramic motion transducer (accelerometer), a clock, and memory storage, was strapped to the left and right wrists and the left ankle of each of the workers. Two portable video cameras were used to record the actions of the cashiers. Monitoring was performed for a 3 to 6 hour period. Work activities were observed and recorded at random intervals throughout each monitoring period. The results indicated that both jobs required near continuous hand movement. Activity levels recorded at both wrists were higher during work than break periods. Dominant wrist activity was greater than nondominant wrist activity for all task types. Clerks had greater ankle activity than cashiers during work periods. Analysis indicated that wrist activity was higher during scanning and bagging periods, intermediate during tendering of money and miscellaneous activities, and lowest during periods of waiting for customers. The authors states that this approach entails less time and effort, yields more precise data, and is more sensitive to subtle activity differences than traditional observational methods. This approach can be used to facilitate future epidemiological studies of the relationship between work activities and the development of upper extremity musculoskeletal disorders.