Normative life satisfaction: measurement issues and a homeostatic model

Previous studies have established that, when life satisfaction is measured over a range from 0 to 100 the mean scores of Western populations average 75 ± 2.5. The consistency of these data has given rise to the idea that life satisfaction may be held under homeostatic control. This paper further investigates this hypothesis by examining the distribution of data within populations. It also examines these data with a view to elucidating various methodological issues regarding life satisfaction measurement. In terms of the methodological issues it is concluded that measurement is best achieved using bi-directional Likert scales with at least 11 choice points. It is also determined that the life satisfaction of Western populations did not change over the decade 1980–1990, and that data derived from college students cannot be validly employed as proxy general population data. In terms of data distribution, it is calculated that the normal range of life satisfaction within Western populations lies within the range 40 to 100. The consistency of this non-normal distribution is argued to be further evidence that life satisfaction is held under homestatic control, and a descriptive model is proposed.