for comments on earlier versions. The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research. NBER working papers are circulated for discussion and comment purposes. They have not been peer-reviewed or been subject to the review by the NBER Board of Directors that accompanies official NBER publications.

for comments on earlier versions. Responsibility for opinions and errors remains with the authors. The views expressed herein are those of the authors and do not necessarily reflect the views of the National Bureau of Economic Research. NBER working papers are circulated for discussion and comment purposes. They have not been peer-reviewed or been subject to the review by the NBER Board of Directors that accompanies official NBER publications.

This paper presents a new public-use dataset for community-level life satisfaction in Canada, based on more than 500,000 observations from the Canadian Community Health Surveys and the General Social Surveys. The country is divided into 1216 similarly sampled geographic regions, using natural, built, and administrative boundaries. A cross-validation exercise suggests that our choice of minimum sampling thresholds approximately maximizes the predictive power of our estimates. The resulting dataset reveals robust differences in life satisfaction between and across urban and rural communities. We compare aggregated life satisfaction data with a range of key census variables to illustrate some of the ways in which lives differ in the most and least happy communities.

Strong versions of the set-point hypothesis argue that subjective well-being measures reflect primarily each individual's own personality and that deviations are temporary. International migration provides an excellent test, since life circumstances and subjective well-being differ greatly among countries. With or without adjustments for selection effects, the levels and distributions of immigrant life-satisfaction scores for immigrants to the United Kingdom and Canada from up to 100 source countries mimic those in their destination countries, and even the destination regions within those countries, rather than those in their source countries, showing that subjective life evaluations are substantially driven by life circumstances and respond when those circumstances change. Résumé. L'immigration comme test de la théorie des seuils de bonheur : l'exemple du Canada et du Royaume-Uni. Les principaux tenants de la théorie des seuils de bonheur font valoir que les évaluations subjectives du bien-être reflètent essentiellement la personnalité de chaque individu, et que les dérogations à ces seuils ne sont que temporaires. L'immigration internationale offre un excellent moyen de mettre cette théorie à l'épreuve étant donné que les circonstances de vie ainsi que le bien-être subjectif varient considérablement d'un pays à l'autre. Avec ou sans ajustements relatifs

This paper presents a new public-use dataset for community-level life satisfaction in Canada, based on more than 400,000 observations from the Canadian Community Health Surveys and the General Social Surveys. The country is divided into 1215 similarly sampled geographic regions, using natural, built, and administrative boundaries. A cross-validation exercise suggests that our choice of minimum sampling thresholds approximately maximizes the predictive power of our estimates. Our procedure reveals robust differences in life satisfaction between and across urban and rural communities. We then match the life satisfaction data with a range of key census variables to explore ways in which lives differ in the most and least happy communities. The data presented here are useful on their own to study community-level variation, and can also be used to provide contextual variables for multi-level modelling with individual life satisfaction data set in a community context.