14. In one of the Stat 2 courses that I have taught, I gave a diagnostic test in

Question

14. In one of the Stat 2 courses that I have taught, I gave a diagnostic test in the first class. The average on the diagnostic test was 6.5 points out of 10. The midterm had an average of 77 points out of 100. I tried two methods of predicting the midterm score for someone in the class: (a) Ignore any information of the diagnostic, and just use the average of 77 as the predicted value for everyone. (b) Use the regression method with the midterm as response (), and the diagnostic score as the predictor The RMS error worked out to be about the means between the diagnostic score and the midterm score must be closest to: 0.1 0.5 0.9

Explanation / Answer

If the RMS error in both the cases have come out to be the same, this means that the error terms are all equal to 0. This can be deduced from the fact that the observed and predicted values are equal in part (a), which gives the residual value as 0. This happens in part (b) as well.

This tells us that the correlation coefficient must be closest to 0.9.

Related Questions

Navigate

• Browse (All)
• Browse 1
• Subjects

---

---

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