Question

Total compensation for teachers is equal to the sum of salary and benefits. You want to test whether the effect of an additional dollar in salary is the same as an additional dollar in benefits so you include the amount of benefits received (measured in thousands) as a regressor. The equation below shows the new model math10-β0 + β1tot comp thous + β2staff per thous + β3enroll thous + β4benef thous + SourceI Number of obs- 408 df MS 7.25 F (4, 403) Model 3010.16739 Residual 41807.0131 0.0000 0.0672 4 752.541847 Prob > F 403 103.739487 R-squared Adj R-squared- 0.0579 Total 44817.1805 407 110.115923 Root MSE 10.185 math10 I Coef. Std. Err t P>t [95% Conf. Interval] 681 1.195 051 -.248 .137 502 040 tot comp thous benefits thous I 4.98 0.000 4122026 2.183142 9496943 2076807 .1292791 1744199 2.38 0.018 1.30 0.195 -1.15 0.249 0264704 -.6711193 staff per thous l enroll thous I cons l 215 6.093 1.283 0.21 0.833 10.69479 13.26157 What is the effect of an extra $1,000 in teachers' salary on the share of students that pass the math test? [Remember, total compensation (tot_comp thous) is measured in thousands of dollars and is equal to the sum of teachers' salary and benefits.] 5. Do you reject the null hypothesis that the effect of an extra dollar in benefits is equal to the effect of an extra dollar in salary, against the alternative that their effects are different, at a 95% confidence level? How did you reach this conclusion? 6.

Answer 1

machers sels vven thaf

/.195 O. 5o2 2380 2. 1s3/