BACKGROUND: In order to evaluate the family-centeredness of paediatric oncology services, a psychometrically sound measure of family-centred services is needed. We performed a comprehensive evaluation of the psychometric properties of the 20-item Measure of Processes of Care (MPOC-20) in parents of children undergoing treatment for cancer at five paediatric oncology centres in Canada. METHODS: The sample included 411 parents (80% response rate). Exploratory factor analysis was used to determine the best way to group the items into scales. Psychometric tests were used to examine data quality, targeting, internal consistency reliability, within-scale construct validity and known-groups validity. RESULTS: Exploratory factor analysis identified two factors: a summary measure of family-centred services and a scale measuring activities that meet parents’ general informational needs. Scores spanned the entire scale range, floor and ceiling effects were low, and the sample distribution was not unduly skewed. Scales showed acceptable internal consistency reliability (Cronbach’s alphas > or =0.93). Known-group hypotheses supported the scales’ ability to differentiate between groups hypothesized to differ. Moderate effect sizes were found when MPOC-20 scale scores for parents and for children with good quality of life were compared with those with poor quality of life. CONCLUSIONS: The MPOC-20 is the only evaluated instrument currently available to measure family-centred services in paediatric oncology. Paediatric cancer programmes can now use this tool to determine parental perception of the extent to which services are family-centred.