Essays on the relationship between income inequality and mortality.
| dc.contributor.author | Wu, Weilun | |
| dc.date.accessioned | 2024-05-29T02:25:57Z | |
| dc.date.available | 2024-05-29T02:25:57Z | |
| dc.date.issued | 2024 | |
| dc.description.abstract | This thesis systematically investigates the relationship between income inequality and mortality with a meta-science approach. Chapter 1 introduces the contested association between income inequality and mortality, respectively, through theoretical background and a critical review of the current literature. Chapter 2 replicates the study of Leigh and Jencks (2007). L&J find that the relationship between inequality and mortality is insignificant economically and statistically. In this chapter, I retrieve L&J’s missing data with the multiple imputation approach and re-analyze the authors’ specifications. I also extend L&J’s analysis with updated data to 2019. All my attempts indicate the replicability of L&J, providing robust evidence to support the insignificant inequality-mortality relationship. Chapter 3 reviews 1,000 published meta-studies from 10 different academic disciplines to understand the meta-analysis (MA) methodology and evaluate the established MA procedures. In this chapter, I highlight the challenges of multilevel meta-data and dependent standard errors associated with commonly used effect types in MA. For the following MAs, I advocate for using robust clustered standard error, a Three-Level meta-analytic model, and appropriate effect types to improve MA accuracy. I also advocate for standard MA procedures such as reporting I-squared values for effect heterogeneity, using meta-regression for identifying heterogeneity sources and addressing publication bias. Chapter 4 applies the above methodologies to an MA examining the income inequality-mortality relationship. This MA includes 84 studies and 1,008 Partial Correlation Coefficients (PCCs). Utilizing the Three-Level model with clustered robust standard error, my initial unadjusted results suggest a moderate, significant impact of inequality on mortality. However, after adjusting for publication bias, the impact appears small and statistically insignificant. Furthermore, Chapter 4 replaces PCCs with Fisher’s z, addressing the bias introduced by the dependency of PCC standard errors on effect sizes. This re-analysis confirms the previous findings with PCC, reinforcing the conclusion of an insignificant relationship between income inequality and mortality. Chapter 5 synthesizes the findings from each chapter and draws a conclusion. With consistent evidence, this thesis does not find that income inequality has a significant influence on mortality. | |
| dc.identifier.uri | https://hdl.handle.net/10092/107119 | |
| dc.identifier.uri | https://doi.org/10.26021/15367 | |
| dc.language.iso | English | |
| dc.language.iso | en | |
| dc.rights | All Rights Reserved | |
| dc.rights.uri | https://canterbury.libguides.com/rights/theses | |
| dc.title | Essays on the relationship between income inequality and mortality. | |
| dc.type | Theses / Dissertations | |
| thesis.degree.discipline | Economics | |
| thesis.degree.grantor | University of Canterbury | |
| thesis.degree.level | Doctoral | |
| thesis.degree.name | Doctor of Philosophy | |
| uc.bibnumber | in1360298 | |
| uc.college | UC Business School |