Quantification of Happiness Inequality
Author: Kalmijn, Wim , wim.kalmijn@upcmail.nl
Department: Faculty Social Sciences
University: Erasmus University Rotterdam, Netherlands
Supervisor: Prof. Ruut Veenhoven
Year of completion: 2010
Language of dissertation: English
Keywords:
Quantification
, Happiness
, Inequality
, Measurement
Areas of Research:
Logic and Methodology
Abstract
Happiness is “the degree to which an individual judges the overall quality of his/her life-as-a-whole favorably”. Since this is something that people have in mind, it can be measured using single direct questions. Such questions are commonly used in surveys of the general population in nations. Responses are rated on a scale with a small number of either numerical or verbal categories.
The core subject of this dissertation is how to convert responses in the sample into information on the happiness distribution in the population. Estimated parameter values of these distributions quantify the inequality both within and across nations. Three specific methodological problems are addressed in that context:
1. How to identify the most appropriate statistic to quantify the inequality of happiness in samples from nations (the standard deviation);
2. How to combine the happiness level and the inequality of a nation into a single joint measure, which is presented as the “Inequality-Adjusted Happiness” (IAH);
3. How to make responses to different questions on happiness comparable using the scale interval method.
In this latter approach, happiness in the population is postulated to be a latent continuous variable on the [0, 10] interval. This interval is partitioned into subintervals; each of them is linked to one of the categories of the scale of measurement. In the case of verbal scales, the positions of the boundaries between the subintervals are estimated by a panel of native speakers. Their judgments are then applied to the frequency distributions observed in survey studies in their country for estimating the mean and standard deviation on scale 0-10 of latent happiness distribution in the population.
