Deck 17: Hierarchical Regression

Question 1

A hierarchical regression can be used when you...

A) want to identify the strongest predictors.
B) want to test multiple outcome variables.
C) want to test predictor variables after taking into account control variables.
D) want to test the relationship between predictor variables.

Answer 1

C

Question 2

What does model one in a hierarchical regression usually test?

A) The variance explained by the predictor variables
B) The variance explained by the confound variables
C) The variance explained by the predictor and confound variables
D) The variance explained by a baseline model with no variables

Answer 2

B

Question 3

What does model 2 in a hierarchical regression usually test?

A) The variance explained by the control variables
B) The variance explained by the control variables after taking into account the predictor variables
C) The variance explained by the predictor variables after taking into account the control variables
D) The variance explained by the predictor variables before taking into account control variables

Answer 3

C

Question 4

In SPSS, what statistic tells you the amount of variance explained by model 2?

A) Adjusted R
B) R square change
C) R square
D) Sig F change

Question 5

If the SPSS output shows that the R square change for model 2 is 0.11, then how should this be interpreted?

A) The confound variable model does not significantly predict the outcome variable.
B) The predictor variables explain 11% of the variance in the outcome variable.
C) The confound variables explain 11% of the variance in the outcome variable.
D) The predictor variable model does not significantly explain the outcome variable.

Question 6

To interpret the significance of the final model we should…

A) report the change statistics for model 2.
B) report the significance of any predictors.
C) report the ANOVA statistic for model 1.
D) report the ANOVA statistic for model 2.

Question 7

To interpret the significance of any individual confounding variables you should…

A) report the statistics in the coefficients box displayed under model 2.
B) report the statistics in the coefficients box displayed under model 1.
C) report the change statistics in the model summary box displayed under model 1.
D) report the change statistics in the model summary box displayed under model 2.

Question 8

Which of the following is the correct order to write up a hierarchical regression?

A) Assumptions, model 1, model 2, final model, individual predictors
B) Individual predictors, model 1, model 2, final model, assumptions
C) Final model, individual predictors, model 1, model 2, assumptions
D) Assumptions, final model, model 1, model 2, individual predictors

Question 9

If a confound variable significantly predicts the outcome variable, then how should this be graphed?

A) Scatterplot
B) Bar graph
C) You should not graph confounding variables.
D) It depends on what the confounding variable is.

Question 10

If a hierarchical regression shows that the final model explains 37.2% of the variance, and the confounding variables explain 17.9% of the variance, then how much do the predictor variables explain?

A) 62.8%
B) 82.1%
C) 19.3%
D) 37.2%

Question 11

In the worksheet exercise, as a percentage, how much variance is explained by the control variables?

A) 21.3%
B) 9.7%
C) .97%
D) .213%

Question 12

As a percentage, how much variance in explained by the final model?

A) 8.9%
B) 9.7%
C) 21.3%
D) 29.6%

Question 13

What is the F ratio for the unique predictive effect of the predictor variables (all together)?

A) 25.116
B) 13.227
C) 21.911
D) .001

Question 14

What is the β for the connectedness predictor?

A) .042
B) 7.168
C) 14.281
D) .303

Question 15

Which of the control variables explains a significant amount of variance in the model?

A) Age
B) Sex
C) Connectedness to nature
D) Location of home