Abstract
Ryff’s Psychological Well-being Scales (PWS) were developed with 120 items and have been used in versions with 84, 54, 42, 24, and 18 items, all divided into the same six dimensions of eudaimonic well-being: self-acceptance, positive relations with others, autonomy, environmental mastery, purpose in life, and personal growth. The use of different combinations of the original 120 items, in different countries and with sociodemographic variability, makes analysis of the reliability and validity of the PWS a complex task. The longer scales have better internal consistency, while the shorter scales have better factorial validity. There is no consensus on the optimal version of the PWS, or on which populations it can be used. We examined scale reliability and factor loadings for the 42-item version in a multi-generational sample from a study of well-being associated with theatre audience involvement (N=581). Participants had similar demographic characteristics to the Ryff development sample. We found comparable internal consistency (α) between older and middle-aged groups, with slightly poorer reliability in the youngest (and smallest) group; average alphas were .71, .78, and .77, respectively, for the three groups. The scales performed with reasonable, but not ideal, factorial validity for the 6-factor model; NFI= .777, CFI= .836, RMSE= .063, PClose= .000, CMIN/DF= 3.089. A single factor hierarchical model did not improve fit. Results suggest that the Ryff scales are reliable across age groups but that there may be factor solutions that are superior to the original Ryff six factors.