Arm/shoulder problems and lymphedema are common in patients treated for breast cancer. The Kwan's arm problem scale (KAPS) is a 13 items self-rating form developed for examination of such problems. Since no such testing has been done, we wanted to examine the psychometric properties of the KAPS. At a median of 47 months (range 32-87 months) post-surgery, 256 stage II breast cancer survivors filled in the KAPS. Findings from physical examination of arm/shoulder function, assessment of lymphedema, and established measures of quality of life (QOL), mental distress and fatigue were used to study convergent and discriminant validity. Factor structure and internal consistency were tested in split-half samples. The total KAPS as well as its two subscales showed high internal consistency (alpha > or = 0.92). A two factor solution explained 70% of the variance in the total KAPS score. Convergent validity was shown as the total KAPS score explained 32.5% of the variance in abduction of the ipsilateral shoulder. Discriminant validity was shown as the total KAPS score explained only 6.7% of the variance in abduction of the contralateral shoulder. Concurrent validity was confirmed since the total KAPS explained 54.8% of the variance in arm symptom subscale of breast cancer-specific QOL. Area under the curve was 0.80 for problems of using the arm at a cut-off of the total KAPS score of 21.5. The results show that the KAPS has good psychometric properties supported by high reliability and considerable convergent and discriminant validity.