A cohort study was conducted to investigate the construct validity of the domains of the EORTC QLQ-C30 (European Organization for the Research and Treatment of Cancer Quality of Life Questionnaire). Ninety-six patients undergoing chemotherapy at the Kingston Regional Cancer Centre, Kingston, Canada were given the EORTC QLQ-C30 and two out of four comparison scales; the Sickness Impact Profile, the McGill Pain Questionnaire, the General Health Questionnaire and the Cancer Rehabilitation Evaluation System, to complete during their clinic visits. Using the multitrait-multimethod matrix to examine relationships, the Spearman ranked correlation coefficients of similar and dissimilar domains between the EORTC QLQ-C30 and the comparison scales were compared. The EORTC QLQ-C30 domains of pain and physical and role functioning performed well in that they converged and diverged appropriately with the other instruments. The domain of psychological functioning had strong associations with domains thought to be dissimilar (social interaction and fatigue). Social interaction and financial impact diverged across dissimilar domains. This lack of specificity may relate to question wording. The symptom domain had strong associations with other domains across the comparison scales. This study shows that the domains of the EORTC QLQ-C30 have substantial construct validity in the multidimensional assessment of the quality of life of cancer patients and identifies where further work is required.