@article{c359c7c1a6a24e19b7f0c9ca5c119bd4,
title = "Happy people or happy places? a multilevel modeling approach to the analysis of happiness and well-being",
abstract = "This article aims to add a regional science perspective and a geographical dimension to our understanding of substantive questions regarding self-reported happiness and well-being through the specification and use of multilevel models. Multilevel models are used with data from the British Household Panel Survey and the Census of UK population to assess the nature and extent of variations in happiness and well-being to determine the relative importance of the area (district, region), household, and individual characteristics on these outcomes. Having taken into account the characteristics at these different levels, we are able to determine whether any areas are associated with especially positive or negative feelings of happiness and well-being. Whilst we find that most of the variation in happiness and well-being is attributable to the individual level, some variation in these measures is also found at the household and area levels, especially for the measure of well-being, before we control for the full set of individual, household, and area characteristics. However, once we control for these characteristics, the variation in happiness and well-being is not found to be statistically significant between areas.",
keywords = "combining data, happiness, hierarchical models, multilevel modeling, well-being",
author = "Dimitris Ballas and Mark Tranmer",
note = "Funding Information: The British Household Panel Survey data were made available through the UK Data Archive. The data were originally collected by the ESRC Research Centre on Microsocial Change at the University of Essex, now incorporated within the Institute for Social and Economic Research. Census data were provided through the Census Dissemination Unit of the University of Manchester, with the support of the ESRC / JISC / DENI 1991 Census of Population Program. We are grateful to Danny Dorling, Andrew Oswald, and Charles Pattie for very useful comments on earlier drafts. We would also like to thank the editor and two anonymous referees for their very constructive comments. All responsibility for the analysis and interpretation of the data presented in this article lies with the authors. The author(s) declared no potential conflicts of interests with respect to the authorship and/or publication of this article. The author(s) disclosed receipt of the following financial support for the research and/or authorship of this article: UK Economic and Social Research Council (research fellowship grant number RES-163-27-1013). 1. “Sample of Anonymised Records” districts are individual 1991 local authority districts or amalgamations of districts so that no area has a population of less than 120,000 ( Marsh and Teague 1992 ) 2. See http://www.cmm.bristol.ac.uk/ 3. It is also worth noting that happiness studies that, as also argued by Clark and Oswald (2002) the results of ordered probits and OLS regression happiness models are qualitatively similar (also see Frey and Stutzer 2002 ) 4. In the remainder of the article we use the terms “well-being” and “happiness” to describe “subjective well-being” and “subjective happiness,” respectively. 5. Chi-square tests comparing the –2 log likelihood values between the null model and the models with explanatory models found that the latter represent significantly better fits than the null model. For instance, the difference between the –2 log likelihood value of the model with all explanatory variables and of the null model was 624.038 and the difference is statistically significant ( p < .0000; 40 degrees of freedom). ",
year = "2012",
month = jan,
doi = "10.1177/0160017611403737",
language = "English",
volume = "35",
pages = "70--102",
journal = "International Regional Science Review",
issn = "0160-0176",
publisher = "SAGE Publications Inc.",
number = "1",
}