The data set consists of information on 4400 full-time full-year workers. The highest educational achievement for each worker was either a high school diploma or abachelor's degree. Theworker's ages ranged from 25 to 45 years. The data set also contained information on the region of the country where the personlived, maritalstatus, and number of children. For the purposes of theseexercises, let

AHE = average hourly earnings(in 2005dollars)

College = binary variable(1 ifcollege, 0 if highschool)

Female = binary variable(1 iffemale, 0 ifmale)

Age = age(in years)

Ntheast = binary variable(1 if Region= Northeast, 0otherwise)

Midwest = binary variable(1 if Region= Midwest, 0otherwise)

South = binary variable(1 if Region= South, 0otherwise)

West = binary variable(1 if Region= West, 0otherwise)

Results of Regressions of Average Hourly Earnings on Gender and Education Binary Variables and Other Characteristics Using Data from the Current Population Survey
DependentVariable: average hourly earnings (AHE).
Regressor	(1)	(2)	(3)
College (X1)	5.90 (0.23)	5.92 (0.23)	5.88 (0.23)
Female (X2)	2.85 (0.22)	2.83 (0.22)	2.83 (0.22)
Age (X3)		0.31 (0.05)	0.31 (0.04)
Northeast (X4)			0.75 (0.32)
Midwest (X5)			0.65 (0.30)
South (X6)			0.29 (0.28)
Intercept	13.71 (0.15)	4.75 (1.13)	4.05 (1.14)
Summary Statistics
F-statistic for regional effects= 0	6.41
SER	6.77	6.72	6.71
R2	0.190	0.205	0.210
n	4400	4400	4400

For each of the three regressions, compute the p value of the test of the hypothesis that the coefficient on College is 0 against the two sided alternative that it is non zero.

• Is the college-high school earnings difference estimated from this regression statistiscally significant at the 5% level?
• Construct a 95% confidence interval of the difference.