Suppose that 5 years ago, a new job training program was introduced in Massachus

Question

Suppose that 5 years ago, a new job training program was introduced in Massachusetts that accepted 1000 applicants. For $200, participants would undergo a 6-week training program that would teach them about basic computer skills, resume building, interviewing, and job searching. Now, 5 years later, you wish to survey some of these participants to see if the training has helped them. You create a simple survey asking participants how much they thought the job training helped them in their professional life on a scale of 1 to 5 (with 5 being most helpful and 1 being least helpful), how much they currently earn per month, and how many jobs they are currently working.

You mail out 100 surveys to known participants, and after waiting several weeks you receive 32 responses. You wish to use the data on each participant’s income and the number of jobs he or she is working to predict how favorably a participant viewed the training (using the 1 through 5 scale above). If you estimate your model using OLS, do you believe that any of our Classical Linear Model assumptions would be violated? If so, explain which ones, why they are violated, and what potential problems that could pose for your estimation.   

Hint: Using an ordinal scale of the left-hand side of an OLS regression is not a problem.

Explanation / Answer

For a linear regression model:

1) number of observations(32>5) are more than the number of classifiers hence there should not be any problem of overfitting

Generally the following assumptions are made for a classical linear regression:

Suggestions:

1) The sample size can be increased to get a clear idea

Related Questions

Navigate

• Browse (All)
• Browse S
• Subjects

Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.