Question

Which of the following is not a research question that should be investigated with logistic regression?

	a.	Prediction of group membership
	b.	Strength of association between criterion and predictors
	c.	Differences between groups
	d.	Interaction between predictors

I believe it is C, but I want to make sure I am correct.

Answer 1

Answer:

a)

Logistic regression is used to a prediction when the outcome is dichotomous.

Therefore, this is a TRUE statement

b)

Strength of association between criterion and predictors is very important in logistic regression. Strength of association measures multicollinearity.

Therefore, this is a TRUE statement

c)

Differences between groups should not be investigated with logistic regression. We can not measure the differences by logistic regression.

d)

Interaction between predictors should be investigated with logistic regression.

Therefore, Differences between groups are not a research question that should be investigated with logistic regression.

Option C is correct.