Both laypersons and social scientists typically assume that psychological well-being or happiness is a response to objective circumstances or events. The present study contributes to recent literature showing that stable individual differences are more useful than life circumstances in predicting well-being. Responses to items from the General Well-being Schedule were examined for 4942 men and women surveyed in a follow-up of a national sample. Results showed substantial stability for well-being scales for total group and demographically defined subgroups, and stability coefficients were as high for those who had experienced changes in marital or employment status or state of residence as for those who had not. These findings point out the need for caution in interpreting well-being scores as indices of the quality of life, because well-being is strongly influenced by enduring characteristics of the individual.