This paper describes the assessment of physical activity in chronic fatigue syndrome (CFS) and investigated the following questions: Do patients with CFS have low levels of physical activity; is there a relationship between actual level of physical activity and fatigue; can self-report measures adequately assess actual level of physical activity; what is the role of cognitions with respect to physical activity; and are results with respect to physical activity specific to CFS? Three different types of activity measures were used: self-report questionnaires, a 12-day self-observation list, and a motion-sensing device (Actometer) which was used as a reference for actual activity level. Fifty-one patients with CFS, 50 fatigued patients with multiple sclerosis (MS), and 53 healthy subjects participated in this study. Although none of the self-report questionnaires showed high correlations with the Actometer, questionnaires that require simple ratings of specified activities were related to the Actometer and can be used as acceptable substitutes, in contrast to instruments that require general subjective interpretations of activity that had low or non-significant correlations with the Actometer. Actometer results showed that CFS patients and MS patients had similar activity levels and both groups were significantly less active than healthy subjects. Compared to MS patients, CFS patients were more likely to indicate that they had been less active than other persons they knew. Activities which patients expected to result in higher fatigue levels were less frequently performed. Patients with CFS had significantly higher scores on this measure than MS patients and healthy subjects. Low levels of physical activity were related to severe fatigue in CFS but not in MS. In conclusion, although CFS patients have similar low activity levels than MS patients, there are also important differences between both groups: in CFS cognitive factors are more prominently involved in producing the low activity levels than in MS and in CFS patients activity level is related to fatigue but not in MS.