A regression of years of schooling S on an intelligence test score (ASVABC), mother's years of schooling (SM), father's years of schooling (SF), MALE, black race (ETHBLACK), Hispanic race (ETHHISP), and interaction terms for MALE and race, yielded the following results. The reference group containing the dropped categories is picked up by the constant term. This reference group is white female because MALE, ETHHISP, and ETHBLAC are explicitly included as regressors. The average effect of being male and white on years of schooling relative to the reference group is:
gen MALEBLAC = MALE*ETHBLACK
gen MALEHISP = MALE*ETHHISP
reg S ASVABC SM SF MALE ETHBLACK ETHHISP MALEBLAC MALEHISP
Source | SS df MS Number of obs = 540
Model | 1267.90571 8 158.488214 F( 8, 531) = 40.87
Residual | 2059.09243 531 3.87776353 Prob > F = 0.0000
Total | 3326.99815 539 6.17253831 R-squared = 0.3811
Adj R-squared = 0.3718
Root MSE = 1.9692
S | Coef. Std. Err. t P>|t| [95% Conf. Interval]
ASVABC | .1230092 .0108183 11.37 0.000 .1017574 .1442611
SM | .1742696 .0436279 3.99 0.000 .0885651 .2599742
SF | .1090157 .0324569 3.36 0.001 .045256 .1727753
MALE | -.5127105 .1874445 -2.74 0.006 -.8809342 -.1444868
ETHBLACK | .8454512 .3718442 2.27 0.023 .1149849 1.575917
ETHHISP | .5528601 .5842755 0.95 0.344 -.5949149 1.700635
MALEBLAC | -.2039688 .5413949 -0.38 0.707 -1.267507 .8595698
MALEHISP | 1.301367 .7767075 1.68 0.094 -.2244294 2.827163
_cons | 4.202037 .5900196 7.12 0.000 3.042978 5.361096
A. 1.34
B. 0.85
C. 0.55
D. -0.51