Abstract
In terms of a composite well-being indicator, literature remains inconclusive regarding the appropriate weighting scheme to apply. Although condemned as arbitrary, equal weights remain popular and are applied in several global indicators. This paper examines whether the well-being level is sensitive to the underlying weighting scheme by comparing equal weights to non-paternalistic weights. Using a representative sample of 1431 Dutch speaking Belgians, we present a well-being index based on five dimensions: health, income, education, family life, and social life. The non-paternalistic weighting scheme is derived by asking respondents to think about the importance of the five dimensions to their well-being, and based on this importance allocate 100 points over the dimensions. We find that the underlying weighting scheme affects the well-being level of individuals who report low outcomes on some dimensions and high on others. We also find that the two schemes deem different groups of individuals to be in the bottom decile, affecting the beneficiaries of a policy targeting the worst-off. Since the well-being of respondents performing poorly on all five dimensions will be low regardless of the applied weights, we recommend the use of non-paternalistic weights to evaluate the well-being of respondents with a varying outcome across dimensions. This recommendation is based on the notion that well-being is intrinsically personal, and therefore is best evaluated by the individuals themselves.
Similar content being viewed by others
Notes
The authors describe the principal component analysis as a method that “exploits the correlation between the highly correlated variables of the HDI to provide a one-dimensional summary. The weights thus derived measure the contribution of each indicator to [HDI]” (Ngufack-Tsague et al. 2011, p. 186).
Deutsch and Silber (2005) provide an example of a society where owning a fridge is very common but owning a dryer is rare. Following their line of reasoning, a greater weight is assigned to owning a fridge since not owning one is a rare occurrence. As a result, an individual is deemed more deprived if he/she does not have a fridge. Owning a dryer, however, is assigned a smaller weight since it is less frequent and is considered rare.
Individuals were divided into groups of six age bands, two genders, and five religions.
Not to be confused with frequency-based weights, group-based weights take into account the perception of the sampled individuals regarding the importance of the dimensions. Frequency-based weights on the other hand simply consider the frequency of best/worst performance on life aspects.
Self-reported subjective well-being levels are often measured with a life satisfaction question (on a scale from 0 to 10, how satisfied are you with your life as a whole?).
The ranking remains the same even if the coefficients are not divided by their p value.
Following the recommendation of Sudman (1976) concerning the sample size of subgroups, very small groups (i.e. with less than 30 individuals) were disregarded.
We assume a constant marginal rate of substitution between dimensions such that this rate does not change with changes in the outcome levels. Although this assumption might not be realistic, in our case, the correlation between the outcome on a dimension and its PA weight is quite low as can be seen in Table 3. Therefore, our assumption of a constant MRS does not raise concerns.
Our interest lies within the weighting scheme, therefore we refrain from further discussing the choice of the aggregation form of the indices. However, it is crucial to point out that a weighted arithmetic sum is wildly used and accepted in the literature on composite well-being indicators (Bellani 2013; Foster et al. 2013).
The bootstrap procedure consists of drawing, a 1000 times, a new sample from our data and calculating the t-statistic obtained from a paired t-test examining the significance of the difference between mean WBPA and mean WBE. Using the bootstrapped t-statistic, we calculate the probability of observing a t-statistic higher than 1.96, which is the minimum t-statistic value needed to reject the null at a p value of 0.05.
The bottom 10% amounts to 143 observations, where the maximum well-being level observed is 600 points. Since our focus is the well-being level, it is not logical to exclude observations with a well-being of 600 points simply because they are not within the threshold of 143 observations. Consequently, accounting for all observations with a well-being level of 600 points or less, we end up with 151 observations rather than 143.
Interested readers can refer to Al-Ajlani et al. (2018) for an empirical analysis on the relationship between individual characteristics and the number of points allotted to each dimension.
References
Al-Ajlani, H., Van Ootegem, L., & Verhofstadt, E. (2018). What is Important for Well-Being? Social Indicators Research, 1–18.
Atkinson, A. B., Marlier, E., & Guio, A. C. (2017). Monitoring social inclusion in Europe: 2017 edition. Luxembourg: Publications Office of the European Union.
Bellani, L. (2013). Multidimensional indices of deprivation: the introduction of reference groups weights. The Journal of Economic Inequality, 11(4), 495–515.
Bellani, L., Hunter, G., & Anand, P. (2013). Multidimensional welfare: do groups vary in their priorities and behaviours? Fiscal Studies, 34(3), 333–354.
Bottomley, P., & Doyle, J. (2013). Comparing the validity of numerical judgments elicited by direct rating, and point allocation: Insights from objectively verifiable perceptual tasks. European Journal of Operational Research, 228(1), 148–157.
Bottomley, P., Doyle, J., & Green, R. (2000). Testing the reliability of weight elicitation methods: direct rating versus point allocation. Journal of Marketing Research, 37(4), 508–513.
Buchanan, J. M. (1991). The foundations for normative individualism. In The economics and ethics of constitutional order, chap 18 (pp. 221–229). Ann Arbor: The University of Michigan Press.
Campbell, A., Converse, P. E., & Rodgers, W. L. (1976). The quality of American life. Perceptions, evaluations, and satisfactions. New York: Russel Sage Foundation.
Chowdhury, S., & Squire, L. (2006). Setting weights for aggregate indices: An application to the commitment to development index and human development index. The Journal of Development Studies, 42(5), 761–771.
Cummins, R. A. (2003). Normative life satisfaction: Measurement issues and homeostatic model. Social Indicators Research, 64(2), 225–240.
Davies, R. (2014). Measuring well-being and progress Looking beyond GDP. European Parliamentary Research Service. Retrieved from http://www.europarl.europa.eu/EPRS/140738REV1-Measuring-well-being-and-progress-FINAL.pdf.
Decancq, K., & Lugo, M. (2013). Weights in multidimensional indices of wellbeing : an overview. Econometric Reviews, 32(1), 7–34.
Decancq, K., Van Ootegem, L., & Verhofstadt, E. (2013). What if we voted on the weights of a multidimensional well-being index? an illustration with Flemish data. Fiscal Studies, 34(3), 315–332.
Deutsch, J., & Silber, J. (2005). Measuring multidimensional poverty: An empirical comparison of various approaches. Review of Income and Wealth, 51(1), 145–174.
Diener, E. (2009). Subjective well-being. In E. Diener (Ed.), The science of well-being (pp. 11–58). New York: Spring.
Diener, E., & Diener, C. (1996). Most people are happy. Psychological Science, 7(3), 181–185.
Doyle, J., Green, R., & Bottomley, P. (1997). Judging relative importance: direct rating and point allocation are not equivalent. Organizational Behavior and Human Decision Processes, 70(1), 65–72.
Dworkin, R. (1981). What is Equality? Part One: Equality of Welfare. Philosophy and Public Affairs, 10, 185–246.
Fiske, S., & Taylor, S. (1984). Social cognition. New York: Random House.
Fleurbaey, M., & Blanchet, D. (2013). Beyond GDP. Oxford: Oxford University Press.
Foster, J., McGillivray, M., & Seth, S. (2013). Composite indices: rank robustness, statistical association, and redundancy. Econometric Reviews, 32(1), 35–56.
Frijters, P., Johnston, D. W., & Shields, M. A. (2011). Life satisfaction dynamics with quarterly life event data. Scandinavian Journal of Economics, 113(1), 190–211.
Gintis, H. (1974). Welfare criteria with endogenous preferences: the economics of education. International Economic Review, 15(2), 415–430.
Guven, C., Senik, C., & Stichnoth, H. (2012). You can’t be happier than your wife. Happiness gaps and divorce. Journal of Economic Behaviour & Organization, 82(1), 110–130.
Haisken-DeNew, J., & Sinning, M. (2010). Social deprivation of immigrants in Germany. Review of Income and Wealth, 56(4), 715–733.
Helliwell, J., Layard, R., & Sachs, J. (2012). World Happiness Report. New York: The Earth Institute, Columbia University.
Helmer, J., Senior, K., & Cunningham, T. (2014). The need for a comprehensive survey of youth wellbeing in the Northern territory. Applied Research in Quality of Life, 10(3), 509–523.
Joskin, A. (2017). What matters to Belgians? Analysis of the determinants of individual well-being in Belgium. Working paper 4-17. Brussels: Federal Planning Bureau.
Krosnick, J., & Alwin, D. (1988). A test of the form-resistant correlation hypothesis: Ratings, rankings, and the measurement of values. Public Opinion Quarterly, 52(4), 526–538.
Lee, H., Kim, K. D., & Shin, D. C. (1982). Perceptions of quality of life in an industrializing country: The case of the Republic of Korea. Social Indicators Research, 10(3), 297–317.
Lind, N. (2009). A Calibrated Index of Human Development. Social Indicators Research, 98(2), 301–319.
McCarty, J., & Shrum, L. (2000). The measurement of personal values in survey research. Public Opinion Quarterly, 64(3), 271–298.
Michaelson, J., Abdallah, S., Steuer, N., Thompson, S., & Marks, N. (2009). National accounts of well-being: Bringing real wealth onto the balance sheet. London: New Economics Foundation.
Mill, J. S. (1991). On liberty. London: Routledge.
Nguefack-Tsague, G., Klasen, S., & Zucchini, W. (2011). On weighting the components of the Human Development Index: a statistical justification. Journal of Human Development and Capabilities, 12(2), 183–202.
Nozick, R. (1974). Anarchy, state and utopia. New York: Basic Books.
OECD. (2011a). How’s Life?: Measuring well-being. Paris: OECD Publishing.
OECD (2011b). Compendium of OECD well-being indicators, available at: oecd.org/sdd/47917288.pdf
OECD (2013). OECD guidelines on measuring subjective well-being. Paris: OECD Publishing. Available at: https://doi.org/10.1787/9789264191655-en.
Roodman, D. (2012). The Commitment to Development Index: 2012 Edition. Washington, DC: Center for Global Development.
Schokkaert, E. (2007). Capabilities and satisfaction with life. Journal of Human Development, 8(3), 415–430.
Stapleton, L., & Garrod, G. (2007). Keeping things simple: why the Human Development Index should not diverge from its equal weights assumption. Social Indicators Research, 84(2), 179–188.
Stiglitz, J.E & Sen, A. & Fitoussi, J.P. (2009) Report of the Commission on the Measurement of Economic Performance and Social Progress (CMEPSP), p 14. Available at: http://ec.europa.eu/eurostat/documents/118025/118123/Fitoussi+Commission+report
Sudman, S. (1976). Applied sampling. New York: Academic Press.
Sugden, R. (2004). The opportunity criterion : Consumer sovereignty without the assumption of coherent preferences. American Economic Review, 94(4), 1014–1033.
Takeuchi, L. (2014). Incorporating people’s values in development: Weighting alternatives. Overseas Development Institute (ODI). Available at: http://www.odi.org/sites/odi.org.uk/files/odi-asset s/publicatio ns-opini on-files /9023.pdf.
Van de Kerk, G., & Manuel, A. (2008). A comprehensive index for a sustainable society: The SSI — the Sustainable Society Index. Ecological Economics, 66(2-3), 228–242.
Author information
Authors and Affiliations
Corresponding author
Ethics declarations
Conflict of interest
The authors declare that they have no conflict of interest.
Additional information
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Appendices
Appendix 1: Hedonic Weights
Appendix 2: Social Groups
Appendix 3: Distribution of Variance Across Different Levels of Absolute Well-Being Differences
Appendix 4
The figure above considers the point distribution of individuals who are considered to be in the bottom 10% according to WBE but not according to WBPA. One dimension is considered per individual, and the dimension is that with the lowest outcome. Five individuals report a minimum outcome value of one, while 12 individuals report a minimum outcome of five. From the figure above, we can see that respondents whose poorest outcome on a dimension has a value of one allocate either no points to this dimension or a maximum of five points, which is relatively low compared to the 100 points available for allocation. Twenty-five percent of respondents assign zero points to the dimension where they perform the poorest. The highest number of points received by a dimension with the poorest outcome is 20 points, which are distributed by only three respondents.
Appendix 5
Rights and permissions
About this article
Cite this article
Al-Ajlani, H., Van Ootegem, L. & Verhofstadt, E. Does Well-Being Vary with an Individual-Specific Weighting Scheme?. Applied Research Quality Life 15, 1285–1302 (2020). https://doi.org/10.1007/s11482-019-09733-0
Received:
Accepted:
Published:
Issue Date:
DOI: https://doi.org/10.1007/s11482-019-09733-0