Cross-section individual level data Impacts of Clustering and Estimator Choices on Estimated Coefficients and Standard Errors
| Estimation Method | ||
|---|---|---|
| OLS | FGLS (RE) | |
| Slope coefficient | 0.0108 | 0.0314 |
| Standard errors | ||
| Default | 0.0042 | 0.0199 |
| Heteroscedastic robust | 0.0042 | — |
| Cluster robust (cluster on state) | 0.0229 | 0.0214 |
| Pairs cluster bootstrap | 0.0224 | 0.0216 |
| Number observations | 65,685 | 65,685 |
| Number clusters (states) | 51 | 51 |
| Cluster size range | 519 to 5,866 | 519 to 5,866 |
| Intraclass correlation | 0.018 | — |
Notes: March 2012 CPS data, from IPUMS download. Default standard errors for OLS assume errors are iid; default standard errors for FGLS assume the random effects model is correctly specified. The bootstrap uses 399 replications. A fixed effect model is not possible because the regressor is invariant within states.