Table 1.
Characteristics of the datasets included in the analysis
| Dataset | Birth cohorts | Data collection | N |
| OCG 1973 | 1908–1952 | Cross-sectional survey | 25,163 |
| NLS Young Men | 1949–1951 | Longitudinal survey | 1,132 |
| NLS Young Women | 1951–1953 | Longitudinal survey | 752 |
| PSID | 1954–1989 | Longitudinal survey | 7,978 |
| NLS72 | 1954 | School cohort survey | 9,637 |
| HS&B | 1962–1964 | School cohort survey | 18,805 |
| NLSY79 | 1962–1964 | Longitudinal survey | 2,259 |
| NELS | 1974 | School cohort survey | 10,337 |
| Add Health | 1977–1982 | Longitudinal survey | 3,850 |
| Chetty et al. (5) | 1981–1993 | Tax data | ca. 13 million |
| NLSY97 | 1980–1984 | Longitudinal survey | 5,254 |
| ELS | 1986 | School cohort survey | 9,990 |
| HSLS | 1995 | School cohort survey | 13,612 |
Add Health, National Longitudinal Study of Adolescent to Adult Health; ELS, Education Longitudinal Study; HSLS, High School Longitudinal Study.