Lecture 9D: Logistic Regression

Question 1

When is it appropriate to use logistic regression?

Answer 1

to predict the probability of an event occurring for a nominal variable with 2 categories only (dichotomy)

Question 2

What is the dependent variable in logistic regression?

Answer 2

A nominal variable with only 2 categories (dichotomy).

Question 3

What is the relationship between logistic regression and linear regression?

Answer 3

Conceptually similar, but logistic regression deals with a dichotomous dependent variable.

Question 4

What does logistic regression generate?

Answer 4

A regression coefficient (B) that can be transformed into an odds ratio (OR).

Question 5

How is the odds ratio calculated in logistic regression?

Answer 5

OR = e^B, where e is the base of the natural log.

Question 6

What characterizes an unadjusted odds ratio?

Answer 6

Involves only 1 independent variable and does not consider confounding variables.

Question 7

What characterizes an adjusted odds ratio?

Answer 7

Involves more than 1 independent variable and accounts for confounding variables.

Question 8

How to report logistic regression?

Answer 8