multiple regression

Question 1

regression

Answer 1

Regression allows us to explore the relationship of multiple variables with the outcome we are interested in, at the same time

Question 2

different forms of regression analysis

Answer 2

standard multiple regression/ hierarchical multiple regression/ stepwise regression/ logistic regression

Question 3

variables in multiple regression (PV and CV)

Answer 3

To investigate the relationship of multiple predictors with one continuous outcome variable

Question 4

PV

Answer 4

predictor variables can be continuous or categorical

Question 5

CV

Answer 5

categorical variable must be continuous

Question 6

why use multiple regression analysis

Answer 6

to figure out % of variance in the CV/ statistical significance/ which factors are making a unique contribution

Question 7

how good is our model at explaining or
predicting the CV?

Answer 7

r square value range from 0 (0% of variance explained) to 1(100% or variance explained)

Question 8

Standardized Beta Values (β)

Answer 8

can compare PVs to determine which is the strongest predictor, weakest predictor.

Question 9

postive beta values

Answer 9

The closer the β value gets to +/-1, the stronger its predictive influence on the CV

Question 10

negative beta values

Answer 10

The closer the β value gets to 0, the weaker its predictive influence on the CV

Question 11

what do Standardised beta (β) values indicate

Answer 11

the number of standard deviations that scores on
the CV would change/ IF there was a one standard deviation change in the PV

Question 12

Unstandardised beta value (B)

Answer 12

indicates the nature of the predictive relationship between the particular PV and the CV, in terms of the units that you have used to measure your data

Question 13

Multicollinearity

Answer 13

Refers to the relationships between the predictor variables

Question 14

violation of the assumption of multicollinearity

Answer 14

Any correlations between PVs that are greater than .9 indicate a violation of the assumption of multicollinearity

Question 15

Tests for multicollinearity

Answer 15

1. Tolerance – values less than .10 indicate possible multicollinearity
2. VIF – values above 10 indicate possible multicollinearity

Question 16

singularity

Answer 16

Occurs when one PV is actually a combination of 2 or more other PVs

Question 17

Hierarchical Multiple Regression

Answer 17

Variables are added sequentially rather than simultaneously – specified order

Question 18

How does a Multiple Regression analysis differ from a Bivariate Correlation analysis?

Answer 18

bivariate correlation analysis focuses on the relationship between two variables, while multiple regression analysis explores how multiple independent variables collectively influence a dependent variable.