Exploring the Sources of Downward Bias in Measuring Inequality of Opportunity | |
Lara Ibarra, Gabriel ; Martinez Cruz, Adan L. | |
World Bank, Washington, DC | |
关键词: LIVING STANDARDS; EQUAL OPPORTUNITY; HOUSEHOLD SURVEY; POPULATION DISTRIBUTION; POPULATION EDUCATION; | |
DOI : 10.1596/1813-9450-7458 RP-ID : WPS7458 |
|
学科分类:社会科学、人文和艺术(综合) | |
来源: World Bank Open Knowledge Repository | |
【 摘 要 】
This study analyzes the extent ofdownward bias in the calculation of inequality ofopportunity for continuous outcomes such as income. Atypically recognized source of bias is the unobservedcircumstances as there is a limited set of variablesavailable in household and labor force surveys. Anotherpreviously overlooked source is the likely unobservablenature of top incomes. Using Monte Carlo simulations wherethe underlying inequality of opportunity is predetermined atvarious levels, the study presents three key findings.First, the omission of a relevant circumstance can bias theinequality of opportunity estimate by as much as 80 percent,depending on how much variation of the outcome suchcircumstance explains. Second, not observing the top 5percent of the income distribution can lead to downwardbiases of anywhere between 12 and 35 percent, and thecombination of missing the most favored population and evenone relevant circumstance exacerbates the bias of theempirical estimates. The third key result is that theestimated inequality of opportunity is strongly correlatedwith the amount of variation in the outcome variableexplained by the combination of circumstances (measured bythe R2). This result suggests that in empiricalapplications, the inequality of opportunity estimate can beroughly (and quickly) approximated using simple econometric techniques.
【 预 览 】
Files | Size | Format | View |
---|---|---|---|
Exploring0the00ality0of0opportunity.pdf | 977KB | download |