Sensitivity of Results to Treatment of GED Holders
| GED Holders Have 12 Years Schooling | GED Holders Have Less Than 12 Years Schooling | ||||
|---|---|---|---|---|---|
| Sex–Age Adjusted | Not Sex–Age Adjusted | Not Sex–Age Adjusted | |||
| Fixed Effects (1) | Fixed Effects (2) | Fixed Effects (3) | Ten-Year Differences (4) | Ten-Year Differences 2SLS (5) | |
| Share population 11–64 that is immigrant; t–10 | 0.20** | 0.20** | 0.31** | 0.23** | 0.48** |
| (0.06) | (0.07) | (0.10) | (0.10) | (0.18) | |
| Share population 11–17 that is immigrant; t–10 | −0.45** | −0.18 | −0.15 | −0.26** | 0.46 |
| (0.10) | (0.16) | (0.15) | (0.12) | (0.41) | |
| Share population 18–64 that is immigrant; t–10 | |||||
| <12 years education | 0.88** | 0.44** | 0.56** | 0.51** | 0.97** |
| (0.15) | (0.13) | (0.18) | (0.18) | (0.31) | |
| 12 years education | −0.45 | −0.39 | −0.57 | 0.40 | 4.18 |
| (0.28) | (0.30) | (0.43) | (0.55) | (2.75) | |
| >12 years education | 0.67** | 0.53* | 0.81** | 0.05 | −3.30 |
| (0.28) | (0.29) | (0.40) | (0.52) | (2.34) | |
| Average age of natives | — | Yes | Yes | Yes | Yes |
| Other covariates | Yes | Yes | Yes | Yes | Yes |
| Observations | 343 | 294 | |||
Notes: The dependent variable is the share of native-born aged 21–27 who have completed 12 years of schooling; in Column 1 only this is adjusted at the individual level for age and sex. In Columns 1 and 2, GED holders are considered to have completed 12 years of schooling, while in Columns 3–5 they are not. Estimation is by weighted least squares: Column 1 weights k are the inverse of the squared standard errors on the state–year interaction coefficients in the individual regression, Columns 2–3 weights are the native population aged 21–27, Column 4–5 weights are 1/(1/kt+1/kt+10). All specifications include year dummies; fixed effects specifications also include state dummies. Other covariates are those in Table 2: unemployment rates, cohort size, BEA region trends, 1940 agriculture share trend, 1940 share white natives with less than 12 years education trend. The dependent variables are based on 1950–2010 data, adjusted in Columns 3–5 with official GED statistics, the independent variables on data from 1940–2000. The instruments in Column 5 are based on the 1940 distribution of immigrants from different regions (see text) and correspond to those in 2SLS 1 in earlier tables. Standard errors are clustered by state and reported in parentheses.
*** p < 0.01,
↵** p < 0.05,
↵* p < 0.1.