Choosing a measure instrument for a study raises the question of whether instruments designed for the same purpose produce the same results. We investigated this question for two instruments designed to measure subjective quality of life (QOL) in cancer clinical trials: the Functional Living Index-Cancer (FLIC) and the Quality of Life Questionnaire Core module (QLQ-C30). These were administered concurrently to 98 cancer patients. Four patient groups were defined: (1) well, no chemotherapy (n = 23); (2) adjuvant chemotherapy (n = 24); (3) stable disease, active chemotherapy (n = 24); (4) progressive disease (n = 27). Both instruments have global, role, social, emotional, pain, and nausea scales; QLQ-C30 also assesses physical function, cognitive function, and fatigue, while FLIC assesses hardship. Correlation analysis indicated convergent validity for the global, role, emotional, pain and nausea dimensions, but not the social dimension. Both instruments indicated that groups 1 and 2 had better QOL than group 4 in at least one dimension. However, different dimension-specific results meant that qualitatively different conclusions would have been drawn if either instrument had been used singly: FLIC indicated that group 1 had better role function that group 4 and suffered less hardship and that group 1 suffered less nausea than group 3, while the QLQ-C30 data indicated that group 2 had better physical function than group 4. The only consistent result was for pain: both instruments indicated group 4 had more pain than either groups 1 or 2. Thus the choice of QOL instrument for use in a particular trial will affect both the results and conclusions. It is important, therefore, to consider carefully which instrument is most likely to detect important differences relevant to the patients' lives in that setting.