A recent study of the salaries of elementary school teachers in a small school district in Northern California came up with the following estimated equation:

Where:

SAL_i = the salary of the ith teacher (in dollars)

EMP_i = the years that the ith teacher has worked in this school district

UNITS_i = the units of graduate work completed by the ith teacher

LANG_i = a dummy variable equal to 1 if the ith teacher speaks two languages

EXP_i = the total years of teaching experience of the ith teacher

a. Make up and test appropriate hypotheses for the coefficients of this equation at the 5-percent level.

b. What is the functional form of this equation? Does it seem appropriate? Explain.

c. What econometric problems (out of irrelevant variables, omitted variables, and multicollinearity) does this equation appear to have? Explain.

d. Suppose that you now are told that the simple correlation coefficient between EMP and EXP is .89 and that the VIFs for EMP and EXP are both just barely over 5. Does this change your answer to part c above? How?

e. What remedy for the problem you identify in part d do you recommend? Explain.

f. If you drop EMP from the equation, the estimated equation becomes Equation 8.22. Use our four specification criteria to decide whether you prefer Equation 8.21 or Equation 8.22. Which do you like better? Why?