[Measuring psychological well-being in the elderly]

The results of a comparison of measures assessing psychological well-being in a random sample of 354 elderly people in the city of Groningen are presented: Positive Affect and Negative Affect of the Affect Balance Scale, Positive Well-being (comprising three subscales: General Positive Affect, Emotional Ties and Life Satisfaction), and the Seven Point Satisfaction Rating. None of the studied instruments is based on a clear conceptual model of well-being. The results show reasonable psychometric properties of most measures (internal reliability, skewness, test-retest reliability). Exceptions are the low internal reliability of Positive Affect and the low test-retest reliability of the (I item) Seven Point Satisfaction Rating; the same results were previously reported. Furthermore, the test-retest reliability of the short scales is lower. As expected, the correlation between Positive Well-being and Positive Affect is strong; the same holds for the correlation between Positive Well-being and General Positive Affect. The last one is suggested as alternative for the subscale of the Affect Balance Scale. The results of multiple regression analysis showed that indicators of physical and social well-being explained most variance in General Positive Affect.