Multidimensional well-being indicators have the potential to reduce the “bias” associated to monetary indicators. However, they face stringent data constraints. This paper studies the construction of indicators that strike a balance between (i) reliability in approximating conceptually sound well-being comparisons and (ii) simplicity of application and communication. The recommendations focus on globalmultidimensional poverty measures. The paper identifies three potential sources of improvements: “wasting” less data, better filtering the data, and further developing multidimensional analysis. Less information would be “wasted” by avoiding needlessly dichotomizing all the variables, using the available mortality data, and combining variables from separate surveys. To filter the data better, “equal weights” could be replaced by weights selected from external information on preferences. When the data permit, the unit of analysis should be switched from household level to individual level. Finally, multidimensional indicators should be used to help move beyond a suboptimal “dimension-by-dimension” approach to policy making.
Authors
- DOI
- https://dx.doi.org/10.1596/1813-9450-10800
- Disclosure Date
- 2024/06/12
- Disclosure Status
- Disclosed
- Doc Name
- Multidimensional Well-Being Measurement Practices : A Review Focused on Improving Global Multidimensional Poverty Indicators
- Originating Unit
- Off of Sr VP Dev Econ/Chief Econ (DECVP)
- Published in
- United States of America
- Series Name
- Policy Research working paper ; no. WPS 10800; PEOPLE;
- Unit Owning
- DECRG: Poverty & Inequality (DECPI),EFI-AFR2-POV-Poverty and Equity (EAWPV)
- Version Type
- Final
- Volume No
- 1
Table of Contents
- Introduction 4
- Well-being measurement in practice 7
- General notion of well-being 7
- Why aggregate across dimensions? 8
- Limits of the monetary approach 9
- Data constraints on non-monetary dimensions 10
- Multidimensional indices under data constraints 12
- Minimizing incorrect well-being comparisons 14
- Improving on current indicators 16
- Opportunities for improvement 17
- Criteria for selecting opportunities 17
- Identification not based on dichotomous variables 18
- Hybrid indices that integrate longevity 23
- Combining data from separate surveys 26
- Assuming the overlap between two types of poverty 26
- Imputing incomes into the non-monetary survey 29
- Total consumption 31
- Taking advantage of individual-level variables 33
- Beyond ad-hoc weights 34
- Improving multidimensional analysis 38
- Discussion of opportunities 40
- Research papers 41
- Appendix 44
- The global MPI and the World Bank's MPM 44