OBJECTIVES: To establish a model that estimates a preference-based measure of social participation for children between 8 and 18 years from the Handicap Scale for Children (HSC) classification system. METHODS: A sub-sample of the social participation vignettes of the HSC classification system was valued on a VAS scale by a representative sample of 249 persons of the general Dutch population aged 18 years onwards. Several models based on a full state (econometric) method were considered in order to extrapolate the available valuations for the sub-sample to all possible vignettes. RESULTS: The best fitted model assumes that the VAS scale and the scale of the five dimensions of the HSC (mobility, physical, daily activities, social integration and orientation) are linear, includes an interaction for mobility with physical independence, orientation with mobility, physical independence, daily activities and social integration, and accounts for clustering by respondents. The measurement properties of the weighted scoring of the best fitted model improves upon models which assume an ordinal scale of the five dimensions, do not include interactions and/or do not account for clustering. The model that simply sums up the levels of each dimension provides systematic errors for the preferences. CONCLUSIONS: We obtained a preference-based measure for social participation for children between 8 and 18 years that can be used for assessing need, for quality assurance and for evaluating interventions on a group level aiming to increase social participation in children with chronic illnesses.