2. A. Comcast\'s customer service department asks its customers to rate their ov
ID: 3229489 • Letter: 2
Question
2. A. Comcast's customer service department asks its customers to rate their over-the-phone service on a scale of 1 20 immediately after their service has been completed. The company then matches each customer's rating with the number of minutes the person waited on hold. The following table shows the ratings and number of minutes on hold for 5 randomly selected customers followed by the regression printout. (16 points Minutes Rating 4 15 SUMMARY OUTPUT Regression Statistics 0.613571991 Multiple R R Square 0.376470588 Adjusted R Square 0.168627451 Standard Error 2.658320272 Observations ANOVA Significance MS Regression 12.8 12.8 1.811321 0.271022 21.2 7.066667 Residual Tot Standard Upper Lower Upper 95.0% Coefficients t Stat P-value Lower 95% 95% 95.0% rror 18 3.201041 5.62317 0.011122 7.812857 28.18714 7.812857 28.18714. ntercept Minutes (X) -0.8 0.594418 1.34585 0.271022 2.6917 1.091705 2.6917 091705Explanation / Answer
Solution:
Part a
The dependent variable is given as rating and independent variable is given as minutes. The regression equation is given as below:
Y = 18 – 0.8*X
Part b
The slope for the above regression equation is given as -0.8 which indicate the negative relationship. If there is one unit of increase in X, then dependent variable will decrease by 0.8. The intercept for the given regression equation is given as 18 which indicate the value of Y by assuming X as constant.
Part c
There is no sufficient evidence to indicate that X and Y are linearly related. The correlation coefficient is given as 0.6136 while the test for slope has p-value as 0.271022 which is greater than alpha value 0.05.
Part d
The sample correlation coefficient is given as 0.6136, which indicate considerable positive linear relationship or association.
Part e
The coefficient of determination or the value of R square is given as 0.3765. About 37.65% of the variation in the customer’s rating is explained by regression model, while 1 – 0.3765 = 0.6235, which means about 62.35% of the variation in the customers rating is explained by regression model.
Related Questions
Navigate
Integrity-first tutoring: explanations and feedback only — we do not complete graded work. Learn more.