> RSSu RSSu [1] 128.2469 > ols2 summary(ols2) Call: lm(formula = lnwage ~ looks + union + goodhealth + black + married + south +

> RSSu<-sum(resid(olsu)^2)

> RSSu

[1] 128.2469

> ols2<-lm(lnwage~looks+union+goodhealth+black+married+

+ south+bigcity+smallcity+service+education+

+ experience,data= gendergap1100_)

> summary(ols2)

Call:

lm(formula = lnwage ~ looks + union + goodhealth + black + married +

south + bigcity + smallcity + service + education + experience,

data = gendergap1100_)

Residuals:

Min 1Q Median 3Q Max

-1.06625 -0.23567 0.01723 0.24534 1.09755

Coefficients:

Estimate Std. Error t value Pr(>|t|)

(Intercept) 0.4645818 0.0868923 5.347 1.09e-07 ***

looks 0.0416921 0.0157173 2.653 0.0081 **

union 0.1727979 0.0237067 7.289 5.99e-13 ***

goodhealth 0.1295061 0.0436830 2.965 0.0031 **

black -0.0916025 0.0407453 -2.248 0.0248 *

married 0.0564234 0.0236144 2.389 0.0170 *

south 0.0553891 0.0281896 1.965 0.0497 *

bigcity 0.1699972 0.0305498 5.565 3.31e-08 ***

smallcity 0.0976759 0.0243952 4.004 6.65e-05 ***

service -0.1494790 0.0251925 -5.933 3.98e-09 ***

education 0.0512549 0.0045721 11.210 < 2e-16 ***

experience 0.0114133 0.0009429 12.104 < 2e-16 ***

---

Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 0.3473 on 1088 degrees of freedom

Multiple R-squared: 0.283, Adjusted R-squared: 0.2757

F-statistic: 39.03 on 11 and 1088 DF, p-value: < 2.2e-16

> RSSr<-sum(resid(ols2)^2)

> RSSr

[1] 131.2417

> F<-((RSSr-RSSu)/2)/(RSSu/(1086))

> F

[1] 12.67983

> p=1-pf(F, 2, 1086)

> p

[1] 3.601902e-06

> anova(olsu, ols2)

Analysis of Variance Table

Model 1: lnwage ~ looks + union + goodhealth + black + married + south +

bigcity + smallcity + service + education + female + experience +

female:experience

Model 2: lnwage ~ looks + union + goodhealth + black + married + south +

bigcity + smallcity + service + education + experience

Res.Df RSS Df Sum of Sq F Pr(>F)

1 1086 128.25

2 1088 131.24 -2 -2.9947 12.68 3.602e-06 ***

---

Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

I don't know how to explain it in words.

Can you reject the null hypothesis above and accept the alternative hypothesis of a different intercept and/or partial slope for females at conventional significance levels?