Search

There is a vast literature on the determinants of subjective wellbeing. International organisations and statistical offices are now collecting such survey data at scale. However, standard regression models explain surprisingly little of the variation in wellbeing, limiting our ability to predict it. In response, we here assess the potential of Machine Learning (ML) to help us better understand wellbeing. We analyse wellbeing data on over a million respondents from Germany, the UK, and the United States. In terms of predictive power, our ML approaches do perform better than traditional models. Although the size of the improvement is small in absolute terms, it turns out to be substantial when compared to that of key variables like health. We moreover find that drastically expanding the set of explanatory variables doubles the predictive power of both OLS and the ML approaches on unseen data. The variables identified as important by our ML algorithms - $i.e.$ material conditions, health, and meaningful social relations - are similar to those that have already been identified in the literature. In that sense, our data-driven ML results validate the findings from conventional approaches., Comment: The first three authors share joint first authorship. The displayed order of these authors is random and was determined by the AEA randomization tool, confirmation code oJsh ZMZJwhH

Which occupations are best for wellbeing? There is a large literature on earnings differentials, but less attention has been paid to occupational differences in non-pecuniary rewards. However, information on both types of rewards is needed to understand the dispersion of wellbeing across occupations. We analyse subjective wellbeing in a large representative sample of UK workers to construct a measure of "full earnings", the sum of earnings and the value of non-pecuniary rewards, in 90 different occupations. We first find that the dispersion of earnings underestimates the extent of inequality in the labour market: the dispersion of full earnings is one-third larger than the dispersion of earnings. Equally, the gender and ethnic gaps in the labour market are larger than data on earnings alone would suggest, and the true returns to completed secondary education (though not to a degree) are underestimated by earnings differences on their own. Finally, we show that our main results are similar, and stronger, for a representative sample of US workers. [Funding for this report was provided by an Écoles Universitaires de Recherche, EUR (University Research Schools, EUR) grant.]

The COVID-19 pandemic has led many governments to implement lockdowns. While lockdowns may help to contain the spread of the virus, it is possible that substantial damage to population well-being will result. This study relies on Google Trends data and tests whether the lockdowns implemented in Europe and America led to changes in well-being related topic search terms. Using different methods to evaluate the causal effects of lockdown, we find a substantial increase in the search intensity for boredom in Europe and the US. We also found a significant increase in searches for loneliness, worry and sadness, while searches for stress, suicide and divorce on the contrary fell. Our results suggest that people's mental health may have been severely affected by the lockdown.

This report documents the multicultural education programs funded by the New York City Board of Education in 1989-90, the second year of the program. Although a formal evaluation was not conducted, program implementation was documented through interviews and surveys of district staff. An overall summary and descriptions of program components are given for 21 districts or high school superintendencies. Programs were required to achieve at least 1 out of 10 defined goals, in each of the following categories: (1) self-respect and ethnic pride; (2) respect for cultural diversity; (3) interpersonal skills and awareness of the negative consequences of bias; and (4) historical awareness of discrimination. Some activities were continuations of activities previously started, but many were new initiatives. Only a few programs met the stipulation that they meet the needs of special students, and only approximately half had activities that directly involved parents. Few districts made explicit efforts to infuse the concept of multicultural education into the whole instructional process, and many efforts were hindered by funding difficulties. These factors contributed to the lack of equivalence among programs. Recommendations for program continuation stress the need for fewer and better-defined goals with better evaluation. (SLD)