A marketing consultant would like to analyse whether an advertisement is recalled or not by a group of consumers, predicted by the number of hours of television they watch. He has received various pieces of advice from colleagues. Which should he listen to?
A) Logistic regression is not suitable with this predictor variable.
B) Logistic regression is not suitable as the dependent variable only has two categories.
C) Logistic regression is not suitable as only one predictor variable is postulated.
D) Logistic regression is perfectly acceptable for this model.

Question

Quizplus · Accepted Answer

Logistic regression is appropriate when the dependent variable is dichotomous, as is the case here, where the advertisement is either recalled or not recalled. Moreover, it is possible to use a single predictor variable with logistic regression, and the number of hours of television seems a reasonable predictor for advertisement recall. Therefore, logistic regression is an acceptable approach for this analysis.

A Marketing Consultant Would Like to Analyse Whether an Advertisement