Abstract

The hedonic view on well-being, consisting of both cognitive and affective aspects, assumes that through maximizing pleasurable experiences, and minimizing suffering, the highest levels of well-being can be achieved. The eudemonic approach departs from the concept of a good life that is not just about pleasure and happiness, but involves developing one-self, being autonomous and realizing one's potential. While these approaches are often positioned against each other on theoretical grounds, this paper investigates the empirical plausibility of this two dimensional view on subjective well-being. The interrelations between common measures such as the General Health Questionnaire, the CES-D inventory of depressive symptoms, the satisfaction with life scale and the eudemonic CASP scale are examined in a confirmatory factor analysis framework using the third wave of the English Longitudinal Study of Ageing (ELSA). A multidimensional structure of well-being, distinguishing cognitive, affective and eudemonic well-being, is shown to be the best fitting empirical solution. This three dimensional second order structure is neutral to gender in its measurement. A lower influence of feeling energetic on self-actualisation, and of somatic symptoms of depression on affective well-being was noted for respondents in the fourth age in comparison to respondents in the third age. These small measurement artefacts underline that somatic symptoms of later life depression should be distinguished from mood symptoms. Two main social facts are confirmed when we compare the different forms of well-being over gender and life stage: men tend to have a higher level of well-being than women, and well-being is lower in the fourth age than in the third age. Although the three measures are very closely related, with high correlations between .74 and .88, they each have their specific meaning. While affective and cognitive well-being emphasize the use of an internal yardstick to measure well-being, the eudemonic perspective adds an external dimension. As each measure has an own story to tell, we advocate the use of these multiple assessments of well-being.