Imagine we wanted to look at predictors of popularity of famous sports people. We might hypothesize that the type of sport played would predict how popular the sports person was. Imagine we recruited 1000 participants and asked them to rate out of 10 (10 = They are the best sports person in the world, 0 = They are the worst sports person in the world) how much they liked 100 famous sports people from 10 different types of sports (10 sports people from each category). Could we analyse these data using regression, and if so how?
A)Yes.To do this we would create 9 dummy variables for the 'type of sport played' variable.
B)Yes.To do this we would need to create one coding variable for the 'type of sport played' predictor variable.
C)No.We could not conduct a regression because our categorical predictor is made up of more than two categories.
D)No.We can only analyse this scenario using ANOVA.

Question

Quizplus · Accepted Answer

To include the categorical predictor variable "type of sport played" in a regression analysis, we need to create dummy variables for each category. Since there are 10 different types of sports, we would need to create 9 dummy variables to avoid perfect multicollinearity (when one predictor variable is a perfect linear function of one or more other predictor variables). This would allow us to run a multiple regression analysis to examine the relationship between the predictors (including the dummy-coded categorical variable) and the outcome variable (popularity ratings).

Imagine We Wanted to Look at Predictors of Popularity of Famous