Going to War and Going to College: College completion

Going to War and Going to College: College completionTo attempt to control for the confounding effects of Korean Service on our estimates of the effect of World War II service on educational attainment, we add the fraction of a cohort serving during the Korean (but not World War II) to our time series regressions. Results are reported in Tables 6 and 7. Not surprisingly, estimates based on men born between 1923 and 1928 yield very imprecise estimates. Expansion of the time frame of analysis provides some leverage on estimating both effects independently. This brings more of the variation in the manpower demands of the Korean conflict into play, diminishing the problems with identification of both effects seen in the estimates based on the six-year interval. The range through the 1932 birth cohort includes men who would have turned 18 before the start of the Korean conflict, while the longer range includes all men who would have been eligible for Korean service.
The specifications reported in columns (1)-(6) in Tables 6 and 7 report results based on the assumption that service during the Korean conflict had similar effects on educational attainment regardless of birth cohort. Regardless of the cohorts used, we find that the magnitude of the effects of the Korean War and World War II service were quite similar in these specifications, on the order of between .3 and .4 years of college and a 7 to 10 percentage point advantage in college completion.
For all of the reasons enumerated on page 11, it seems more natural to assume that service had a larger effect for men born during the 1930s that it did for men during the 1920s. What is more, while it is possible that Korean War educational deferments, which were introduced in 1951, had a substantial effect on men born in the 1930s, men from the 1927-28 cohorts would have been among the first called up. To attempt to deal with the notion that the Korean conflict might have differentially affected cohorts born during the 1920s and 1930s, we include and interaction between the linear time trend and the variable indicating the fraction of each cohort serving during the Korean conflict. Results are reported in columns (7) and (8). In addition, because we expect the effect of the Korean War to rise and then level off for cohorts born after 1932 we also experimented with specifications that included a quadratic time trend interacted with the Korean War fraction. The results are reported in Column (9). While not very precisely estimated, the coefficients on these interaction terms are, in each case, of the expected sign.
Table 6 – Birth Cohorts

1923-28(1) 1923-29(2) 1923-30(3) 1923-31(4) 1923-32(5) 1923-38(6) 1923-32(7) 1923-38(8) 1923-38(9)
World War II 0.32(0.28) 0.34(0.15) 0.36(0.12) 0.40(0.10) 0.42(0.10) 0.40(0.06) 0.23(0.16) 0.36(0.06) 0.28(0.13)
Korean War 0.28(0.42) 0.34(0.21) 0.36(0.15) 0.42(0.12) 0.44(0.12) 0.39(0.04) 0.16(0.21) 0.35(0.05) 0.21(0.20)
Trend/10 0.26(0.06) 0.29(0.04) 0.29(0.04) 0.30(0.04) 0.32(0.04) 0.33(0.03) 0.30(0.04) 0.30(0.04) 0.29(0.05)
KW*Trend/10 0.19(0.13) 0.11(0.06) 0.44(0.38)
Trend/10 Squared -0.04(0.06)
KW*Trend/10 Squared -0.61(0.66)
N=R2 240.74 280.79 320.82 360.87 400.90 640.90 400.91 640.91 640.91

Table 7: Between cohort estimates of the effect of World War II service and Korean War service on college completion

Birth Cohorts
1923-28(1) 1923-29(2) 1923-30(3) 1923-31(4) 1923-32(5) 1923-38(6) 1923-32(7) 1923-38(8) 1923-38(9)
World War II 0.05(0.08) 0.07(0.05) 0.09(0.03) 0.10(0.03) 0.10(0.03) 0.10(0.01) 0.05(0.04) 0.09(0.01) 0.06(0.03)
Korean War 0.02(0.11) 0.06(0.06) 0.09(0.04) 0.10(0.04) 0.10(0.04) 0.10(0.01) 0.05(0.05) 0.08(0.01) 0.03(0.04)
Trend/10 0.07(0.01) 0.07(0.01) 0.07(0.01) 0.07(0.01) 0.07(0.01) 0.08(0.01) 0.07(0.01) 0.07(0.01) 0.06(0.01)
KW*Trend/10 0.05(0.03) 0.03(0.02) 0.15(0.09)
Trend/10 Squared -0.02(0.01)
KW*Trend/10 Squared -0.23(0.15)
N=R2 240.7 280.7 320.76 360.81 400.87 640.86 400.88 640.87 640.87