A study by Benjamin Smith of the University of Florida examines endogenous selection bias in cross-national statistical research. The research uses a causal model of post-colonial sovereignty on the Arabian Peninsula to illustrate survivorship bias, showing the effect of oil on autocratic survival becomes negligible when corrected for bias. The work motivates context-sensitive statistical modeling for causal inference with observational data.
Use Cases
- Studying survivorship bias in comparative politics based on the described causal model.
- Analyzing the relationship between natural resources and regime survival using the Arabian Peninsula case study.
- Developing statistical models to correct for endogenous selection bias in observational data.
- Teaching concepts of collider bias and selection on the dependent variable in social science research.
Strengths
- Focuses on a specific, well-defined methodological problem in political science research.
- Provides historical evidence to validate the proposed causal model for a specific region.
- Published as Open Access (green), facilitating academic use.
Limitations
- Column-level documentation is absent; field semantics must be inferred after download.
- Row count is unknown, which may limit suitability assessment.
- Last update date is unknown; freshness unverified.
Provenance
- Source
- University of Florida
- Collection Method
- Cross-national statistical research and historical case study analysis.
- Geography
- Arabian Peninsula (primary case study), with comparative global context.