Multicollinearity For the high school female athletes data file, regress the maximum bench press on weight and percent body fat.
a. Show that the F test is statistically significant at the 0.05 significance level.
b. Show that the P-values are both larger than 0.35 for testing the individual effects with t tests. (It seems like a contradiction when the F test tells us that at least one predictor has an effect but the t tests indicate that neither predictor has a significant effect. This can happen when the predictor variables are highly correlated, so a predictor has little impact when the other predictors are in the model. Such a condition is referred to as multicollinearity. In this example, the correlation is 0.871 between weight and percent body fat.)
Already registered? Login
Not Account? Sign up
Enter your email address to reset your password
Back to Login? Click here