A researcher suspects gender discrimination in terms of salary at a bank. To test his hypothesis, he has run the two regressions in the attached spreadsheet. In "Regression 1", salary is the dependent variable, whereas in "Regression 2" he used the natural log of salary as the dependent variable. Which regression model should he choose?

1. The researcher should choose the model in "Regression 1" because of the higher R².

2. The researcher should choose the model in "Regression 1" because of the higher adjusted R².

3. The researcher should choose the model in "Regression 2" because of the lower standard error.

4.We cannot use adjusted R² or standard errors for model selection in this case, as the models have different dependent variables.

SUMMARY OUTPUT 3 Regression Statistics Regression Statistics Multiple F 0.535688 4 Multiple F 0.510894 R Square 0.286961 Regresssion 1 6 R Square 0.261013 Regression 2 Adjusted | 0.280005 6 Adjusted | 0.253803 7 Standard [ 9551.132 Standard | 0.203219 Observati 8 Observati 208 208 9 ANOVA 10 ANOVA 11 df SS MS df F gnificance F SS MS F gnificance F Regressio 2 7.53E+09 3.76E+09 41.25095 8.79E-16 12 Regressio 2 2.990233 1.495117 36.2033 3.43E-14 Residual 205 1.87E+10 91224120 13 Residual 205 8.466049 0.041298 Total 207 2.62E+10 14 Total 207 11.45628 15 16 Coefficient:andard Err t Stat P-value Lower 95%Upper 95%%ower 95.09pper 95.0% Coefficient:andard Err t Stat P-value Lower 95%Upper 95%ower 95.09pper 95.0% Intercept 27915.44 2791.203 10.00122 1.97E-19 22412.3 33418.59 22412.3 33418.59 17 Intercept 10.32847 0.059388 173.9146 1.4E-224 10.21138 10.44556 10.21138 10.44556 -11757 -6176.95 18 Female -0.18031 0.030109 -5.98867 9.35E-09 -0.23968 -0.12095 -0.23968 -0.12095 Female -8966.98 1415.108 -6.33661 1.46E-09 -11757 -6176.95 Age 446.6467 64.48431 6.92644 5.46E-11 319.5092 573.7842 319.5092 573.7842 19 Age 0.008837 0.001372 6.441094 8.29E-10 0.006132 0.011542 0.006132 0.011542 20 21 22 23