『幸福度』は政策科学のために測定可能か?

抄録

The survey on “happiness” is now popular not only in academic research but also in governmental surveys. Some unresolved issues, however, are appearing on the surveys and the implications in policymaking. On the one hand, it is not rare to find some contradictions in the survey results. Those contradictions seem to result from lack of literacy in data collection and analysis, among others, or of the representativeness of data that, in an ideal situation, is guaranteed by statistical sampling and scientific scaling. On the other hand, even if the methodological issues of data collection and data analysis are solved, more substantive question remains: Should we really measure “degree of happiness” by public opinion survey for policymaking? In other words, can the statistical distribution of responses on “happiness,” on a numerical scale such as 10 point-scale, be a scientific measure of people's wellbeing by public opinion survey? This article provides some comments from the viewpoint of basic statistics and cross-national survey research. These comments would recommend that researchers and policymakers be sensitive to the distinction between the surface structure represented by response distribution in public opinion survey and the deep structure of people's minds.