Now that governments are beginning to think that measuring well-being through self-reports is a good idea, it’s high time to discuss how best to do this. In a recent interview for the International Journal of Wellbeing, Nobel Laureate Daniel Kahneman has expressed fears that governments will insert only one of the widely-used subjective well-being questions (e.g. How satisfied are you with your life as a whole? On a 1-5 Likert scale). He worries that this kind of measure won’t be sensitive enough reveal anything interesting, even in the face of dramatically changing traditional economic indicators. Kahneman’s main concern is that both politicians and the public will view the insensitivity of such measures as a reason to reject all future use of subjective measures of well-being. So, how can governments ensure that they ask the right question from the start?
For starters, one question will simply not be enough. As Martin Seligman discusses in his forthcoming book, Flourish, there is considerable evidence in favour of the idea that there are several distinct and roughly equally important dimensions of well-being. He outlines the following five aspects of well-being as being worthy of inclusion in a dashboard of well-being indicators: positive emotion, engagement, meaning, positive relationships, and accomplishment. The truth of the matter is that more research needs to be done to finalise both the list of distinct aspects of well-being and the most accurate and effective way to measure them with one question. So, the question becomes: how can governments ensure that they ask the right questions from the start?
I propose a ten-year international collaborative effort to answer this question. The importance of this issue means that this should not be considered excessive. Governments in several countries in Europe and around the world already conduct longitudinal panel surveys and some of them even use these surveys to collect data on both economic and subjective well-being indicators. The best of these kinds of surveys for our purposes are those for which each respondent, and everyone else in their household, has to complete the survey every three months or so for at least three years. The number and types of questions would have to be expanded, however. The generic objective economic and demographic questions would need to be accompanied by subjective versions of the same questions, subjective questions about the events that have recently impacted their lives, and large batteries of subjective well-being measures for each potentially important aspect of well-being.
If several governments around the world conduct this kind of survey (with the same questions) very useful information could be gained from them. Not only would we be able to assess how perceptions mediate the effect of objective economic changes on individuals and groups, we would also be able to see what kinds of objective economic and demographic factors affect the various aspects of well-being and if these results are cross-culturally robust. The combination of surveying the same participants and those they live with over time provides more definitive information on how what happens to people and those around them affects their well-being.
Furthermore, the measures of the components of well-being could be refined for accuracy and sensitivity during this process. The questions could be updated and improved with each new round of the surveys, until the ten years is up. At the end, individual governments could decide which of the measures of particular aspects of well-being are most relevant to their country. They would know if the measures were already robust enough to be used as a question in the census or if they needed further refinement and could then act accordingly.
If this kind of approach is adopted, and wellbeing dashboards are incorporated into censuses, then political parties of the not-too-distant future could differentiate themselves by giving precedence to the promotion of certain aspects of well-being over others.