10 - Multiple Regression with Polynomials

Question 1

Theoretically, when should you consider using polynomial regression?

Answer 1

When you have theory or reason to expect the relationship between the predictor and outcome to be curvilinear

Question 2

Empirically, when should you consider using polynomial regression?

Answer 2

When the assumptions for linear regression are violated

Question 3

True or False: Polynomial regression must include all lower-order terms

Answer 3

True

Question 4

What do the coefficients in a polynomial regression tell us?

Answer 4

The strength and direction of the curves

Question 5

What should you do if polynomial terms are too highly correlated with each other?

Answer 5

Use orthogonal polynomials (what they are, idk)

Question 6

How many points of inflection does a polynomial model have?

Answer 6

The highest degree of the polynomial - 1

Question 7

How do you use orthogonal polynomials in R?

Answer 7

Use he poly() function in the lm() formula