1. (a) A regression of “hrly_wage” on “female” was computed using Gretl. The coefficient on “female” in context of “hrly_wage” is -2.51. Interpretation in the context of the problem is the difference in a person’s estimated mean amount of hourly wages (in dollars) between male and female is approximately -$2.51. This shows that there is a negative relationship with “female” and “hourly wage.” This implies a negative relationship between “hourly wage” and “gender.” Thus, hourly wages are less for females than males when it comes to gender.

Output from Gretl: coefficient std. error t-ratio p-value --------------------------------------------------------- const 7.09949 0.210008 33.81 8.97e-134 *** female −2.51183 0.303409 −8.279 1.04e-15 ***

Mean dependent var 5.896103 S.D. dependent var 3.693086

Sum squared resid 6332.194 S.E. of regression 3.476254

R-squared 0.115667 Adjusted R-squared 0.113979

F(1, 524) 68.53668 P-value(F) 1.04e-15

Log-likelihood −1400.732 Akaike criterion 2805.464

Schwarz criterion 2813.995 Hannan-Quinn 2808.804

(b) A regression of “hrly_wage” on “female” and “tenure” was computed using Gretl. The coefficient on “female” in context of “hrly_wage” and “tenure” is -2.09. Interpretation in the context of the problem is the difference in a person’s mean amount of hourly wages (in dollars) between male and female and tenure is approximately -$2.09. This shows that there is a negative relationship with “female” and “hourly wage” and “tenure.” This implies a negative relationship between “hourly wage” and “gender.” Thus, hourly wages are less for females than males when it comes to tenure.

Output from Gretl: coefficient std. error t-ratio p-value -------------------------------------------------------- const 6.13644 0.240055 25.56 4.01e-94 *** female −2.08651 0.295233 −7.067 5.07e-12 *** tenure 0.148746 0.0204344 7.279 1.24e-12 ***

Mean dependent var 5.896103 S.D. dependent var 3.693086

Sum squared resid 5749.681 S.E. of regression 3.315668

R-squared 0.197018 Adjusted R-squared 0.193948

F(2, 523) 64.16124 P-value(F) 1.20e-25

Log-likelihood −1375.352 Akaike criterion 2756.704

Schwarz criterion 2769.500 Hannan-Quinn 2761.714

(c) A regression of “hrly_wage” on “female,” “tenure,” “educ,” and “exper” was computed using Gretl. The coefficient on “female” in context of “hrly_wage” on “female,” “tenure,” “educ,” and “exper” is -1.81. Interpretation in the context of the problem is the difference in a person’s mean amount of hourly wages (in dollars) between male and female, tenure, # of years of education and years of experience is approximately -$1.81. This implies a negative relationship between “hourly wage” and “gender.” Thus, hourly wages are less than males when it comes to tenure, # of years of education and years of experience.

Output from Gretl:

coefficient std. error t-ratio p-value -------------------------------------------------------- const −1.56794 0.724551 −2.164 0.0309 ** female −1.81085 0.264825 −6.838 2.26e-11 *** tenure 0.141005 0.0211617 6.663 6.83e-11 *** educ 0.571505 0.0493373 11.58 9.09e-28 *** exper 0.0253959 0.0115694 2.195 0.0286 **

Mean dependent var 5.896103 S.D. dependent var 3.693086

Sum squared resid 4557.308 S.E. of regression 2.957572

R-squared 0.363541 Adjusted R-squared 0.358655

F(4, 521) 74.39801 P-value(F) 7.30e-50

Log-likelihood −1314.228 Akaike criterion 2638.455

Schwarz criterion 2659.782 Hannan-Quinn 2646.805

(d) A regression of “hrly_wage” on “female,” “tenure,” “educ,”