assignment

Econ 390-01

Midterm 2

Due date is Wednesday, Nov 11th at 11:59 PM. Late submissions are not accepted. Mistakes in submissions are not accepted

RULE: you can collaborate but there is a coefficient of 0.9 that will be multiplied by your score on this exam.

If you collaborate and do not inform me your score is 0.

Question 1: Import the “TeachingRating” data in RStudio.

a. How many variables and how many observations do we have in the data?

b. Provide a summary of all the variables: “Course_eval”, “beauty”, “female”, “minority”, “Nnenglish”, “intro”, “age”, “onecredit”. What they are; What their unit is?

c. What does the mean of female mean?

Question 2: Run a SLRM regression of “Course_eval” over “Beauty”. (Course-eval is the dependent variable)

a. What is the population equation?

b. What is the regression equation?

c. Is the coefficient of “beauty” significant? Preform the hypothesis testing? What is the null? What is the alternative? What does significance mean?

d. What is the 95% confidence interval associated with the slope?

e. What is the average of beauty in the data? (mean). Now according to your predicted equation, what is the predicted course-eval for a professor named Bob who has the average value of beauty? (find mean(beauty) and then insert that in the predicted equation on paper to find the predicted course-eval).

f. What is the effect of Beauty on course-eval? (the slope)

g. Do you think missing “age” variable in regression 1 would cause omitted variable bias? Why? If yes, is the bias upward or downward? Why?

h. Provide the scatter plot and the fitted line for regression 1 in your submission.

Question 3: Run a SRLM of “course-eval” over “female”.

a. What is the model/population equation?

b. What is the predicted equation?

c. Provide the scatter plot and the fitted line. Why the plot looks different than all the previous plots?

d. Do you think gender is a significant explanatory variable on course-evaluation? (check the significance of the slope). Does it mean male and female professors have the same course evaluation in the population?

Question 4: Run a MLRM regression of “course_eval” over “Beauty”, “female”, “minority”, “Nnenglish”, “intro”, “age”, “onecredit”.

a. What is the model/population equation?

b. What is the predicted equation?

c. How do you interpret (read) the coefficient of “Nnenglish” in this regression?

a. Bob is a male professor who has average value of beauty, he is not a minority, and he is a native English speaker. He teaches and introductory 3 credits course, and he is 30 years old. What is his predicted course evaluation according to your regression?

b. What is the percentage of evaluation that is explained by these regressors? (R-squared).

c. Which one/ones of the variables are statistically significant? Why? What does it mean?