The Ranking Measure of Life Satisfaction: A Constructive Critique

Life satisfaction (LS) is used as a welfare-measure in many different ways to inform about what makes people better off and even to quantify tradeoffs between different things like putting a monetary value on environmental goods. All these welfare analyzes rely on the ability of the measurement method to reliably identify preference (or satisfaction) rankings. We demonstrate a potential and systematic identification failure of the currently used measurement method (level measure), which asks people to state their LS level on a bounded and discrete scale. We then propose a new measurement method (the ranking measure), which directly asks the individual to give a preference ranking of the current situation against a past situation. This ranking measure will allow us to test the ability of the level measure to capture preference rankings of two consecutively experienced situations. First results from a similar measure in the German Socio-Economic Panel from the years 1984-87 suggest that there is a systematic bias between the two measures. Our ranking measure will be included in the GESIS Panel in February this year. It will include not only general LS but also satisfaction with specific areas of life. With this paper we contribute to a better understanding of life satisfaction measures and the nature of their measurement biases by formally connecting the measurement method with a theoretical welfare concept and by then testing the derived hypotheses empirically with survey data.