| Article ID | Journal | Published Year | Pages | File Type |
|---|---|---|---|---|
| 5058969 | Economics Letters | 2014 | 5 Pages |
Abstract
•We compare parameter estimates of earnings equations in the PSID and CPS.•The PSID estimates are statistically different from those in the CPS since 1996.•The PSID data substantially underestimate the returns to education in 1996–2007.•The PSID sample appears to be non-randomly selected on earnings.
Comparing earnings equations in the Panel Study of Income Dynamics (PSID) and the Current Population Survey, we find that the PSID considerably underestimates the returns to education during the 1992–2007 period. Non-random selection in the PSID sample appears to be the reason.
Related Topics
Social Sciences and Humanities
Economics, Econometrics and Finance
Economics and Econometrics
