The assessment of physical activity is one of the most important methodological issues in research into physical activity and cancer risk. A sedentary Western lifestyle has been observed to influence biological mechanisms promoting development of certain types of cancer. At present the totality of evidence supports a protective effect against cancers of the colon and probably the breast, while further data concerning carcinoma of other cancers are required. Thus, physical activity represents a powerful public health measure for reducing cancer risk. Studies of the association between physical activity and cancer risk have used a great variety of methods, but have most often included work and/or leisure time activity. Questionnaires are the method most often used and various components of physical activity such as type, frequency, intensity and lifetime physical activity have been recorded. However, the measurements used when assessing physical activity have been hampered by lack of accuracy as regards validity and reliability, missing information on the various components of physical activity and sparse information of lifetime exposure, and often no repeat assessments in cohort studies. Discrepancies between studies elaborating the association between physical activity and site-specific cancer risk may be explained through real differences or lack of information on the various components of physical activity (type, intensity, duration) and incomplete information about the cancer type studied (localization, histological type). The complicated nature of the variable physical activity, combined with incomplete understanding of the pathogenesis of most cancer and lack of knowledge regarding possible biological mechanisms operating between physical activity and cancer, warrants further studies. In these studies methodological improvements in measuring physical activity, combined with inclusion of physiological markers (heart rate, energy balance, hormonal levels, etc.) reflecting the variety of physical activities performed are of particular interest. Assessing biomarkers and intermediate steps for site-specific cancer risk may give us further insight into the relation between physical activity and cancer that will be of enormous interest for public health recommendations.