OBJECTIVES: To test a null hypothesis that the measurement performance of the Functional Assessment of Cancer Therapy - Breast (FACT-B) was constant across breast cancer patients with different educational background.  METHODS: Totally 280 Singaporean breast cancer patients responded to a questionnaire package that included the FACT-B at baseline and follow-up. Quality of life scores were assessed in relation to Eastern Cooperative Oncology Group performance status and self-rated change in quality of life for the purpose of examining validity and responsiveness to change. Intraclass correlation coefficient (ICC) was estimated in patients with stable situations to assess test-retest reliability. RESULTS: The mean FACT-B score decreased monotonically in relation to poorer performance status in the higher (A-level or above) but not lower educated group. The FACT-B score was more responsive to self-rated change in quality of life in the higher than lower educated group (P<0.01 and P>0.10, respectively). Larger ICC of FACT-B (P<0.05) was observed in patients with higher education than those with lower education.  CONCLUSIONS: Measurement performance of the FACT-B was sub-optimal among lower educated individuals.