Chapter 8 Regression

Question 1

Tells us the ration of explained vs unexplained variance

Answer 1

F

Question 2

The b-value tells us the _______ of the line and the _________ of the ____________.If it is significant then _________ variable significantly predicts the_________ variable.

Answer 2

The b-value tells us the GRADIENT of the line and the STRENGTH of the RELATIONSHIP. IF it is significant then the …PREDICTOR variable significantly predicts the OUTCOME variable.

Question 3

We use the _______________ ________ to check the correlation matrix for ________________.

Answer 3

We use the DESCRIPTIVES STATISTICS to check the correlation matrix for MULTICOLLINEARITY.

Question 4

predictors that correlate TOO HIGHLY with one another are indicated by r values

Answer 4

great than .9

Question 5

How do we assess the assumption of independent errors

Answer 5

Check Durbin Watson

Question 6

Criteria for the Durbin Watson

Answer 6

close to 2 or at least between 1 and 3

Question 7

When doing a hierarchical regression we assess the improvement of the model at each level by looking at the

Answer 7

change in R2 and whether it is significant

Question 8

The individual contribution of variables to the regression model can be found in the…
For each predictor variable, you can see if it has made a significant contribution to predicting the outcome by looking at the ….

Answer 8

• Coefficients table

- column labelled Sig. (values less than .05 are significant).

Question 9

what tells you the importance of each predictor in a regression model? What does a big value mean?

Answer 9

-standardized beta values
- bigger absolute value = more
important.

Question 10

To check for multicollinearity, use the ____ _______ from the table labelled _________.
· If these values are less than ____, then there probably isn’t cause for concern.

Answer 10

• To check for multicollinearity, use the VIF VALUES from the table labelled COEFFICIENTS.
• If these values are less than 10, then there probably isn’t cause for concern.

Question 11

If you have a VIF value greater than 10 but the average of all VIF values is not substantially greater than ___ there is probably no cause for concern.

Answer 11

1

Question 12

A Cook’s distance greater than ___ indicates it could be influencing the model

Answer 12

1

Question 13

List the things you check to determine if you have outliers or influential cases:
-
-
-
-
-

Answer 13

Mahalanobis Distance
Cooks Distance
Average Leverage
DfBeta
Standardized Residuals

Mahalanobis Cooked an average sized beta and made a fairly standard gravy with the residue.

Question 14

Which graph helps determine homogeneighty of variance? What do we want to see?

Answer 14

• ZPRED ZRESID

- random array of dots

Question 15

what does it mean if the ZRESID ZPRED looks like a funnel

Answer 15

this is probably a

violation of the assumption of homogeneity of variance

Question 16

what does it mean if the ZRESID ZPRED dots have a pattern

Answer 16

this is probably a violation of the assumption of linearity

Question 17

We want histograms to look________ and PP plots to look like _____ _____ to assume normality

Answer 17

We want histograms to look NORMAL and PP plots to look like DIAGONAL LINE to assume normality