1. Multiple Regression

Question 1

Preconditions for a multiple regression?

Answer 1

A continuous outcome variable (predictors can be continuous and categorical)

Question 2

Procedure used where a categorical predictor variable has more than two levels?

Answer 2

Dummy coding

Question 3

What is the point of hierarchical regression?

Answer 3

To control for certain variables

Question 4

What statistic determines the significance of R-squared?

Answer 4

F-test

Question 5

What is the difference between R-squared and adjusted R-squared?

Answer 5

The adjusted measurement control for sample size and the amount of predictors.

Question 6

What is the significance test used for individual predictors?

Answer 6

t-test

Question 7

What is the measurement provided by sr “part” (semi-partial correlation)? Range from?

Answer 7

Unique overlap of an individual IV with the DV. -1 - +1

Question 8

What does Durbin-Watson tell us?

Answer 8

The significance of an additional predictor on its own

Question 9

Residuals should lie between ?? and ?? so as not to be regarded as outliers?

Answer 9

-3.29 and 3.29

Question 10

A Cook’s distance result of less than ?? is a concern?

Answer 10

1

Question 11

What does homoscedasticity signify?

Answer 11

Variance of residual constant at all levels of the predictor (opposite true for heteroscedasticity)

Question 12

What is VIF?

Answer 12

Ratio of variance in a model with multiple terms, divided by the variance of a model with one term alone