Question

1. You are analyzing a customer satisfaction survey for clients in which (Overall) Satisfaction, Price Satisfaction and Reliability Satisfaction (and several other attributes) are measured on a 1-10 scale, 10 being the most favorable value for each variable. You fit a regression model of the form

Satisfaction= α+β1Price+β2Reliability+ε

and find that in this model the estimated coefficient of Price is positive, but that of Reliability, is negative. You give a preliminary talk for your business clients, in which they declare this result must be false and cannot be presented to management.

1. (2 pt) Why is this result unacceptable?
2. (4 pts) Why might this result have occurred, and how could you modify the model to develop a more acceptable result?

Answer 1

Answer:

Step(1)- (a)

The regression equation can be presented as : Satisfaction= α+β1Price+β2Reliability+ε

Where customer satisfaction taken as a dependent variable, price and realibility as a independent variable

As per qus. states that price having positive constant and reliability have negative constant that means there is postive association between price of goods and customer satisfaction i.e as increase in the price of goods satisfaction should be increase but practically this is not possible, and there is negative association between reliability of the product and customer satisfaction i.e lower the reliable of product greater the satisfaction .practically this is meaningless.

Generally every customer wants that maximum the relaibility of product at optimum investment. since the model is unacceptable.

Step(2)-

(b) it can be say, a combination of other independent variables relates strongly to the dependent variable , or you can take large sample size .