Standard Multiple Regression Flashcards
What is multiple regression?
Variation as a function of multiple predictors
What is a strength of multiple regression?
Better prediction by accounting for more variance
What is multiple correlation?
R: Relationship between criterion and a set of predictors
What are the tests in multiple regression?
- Strength of overall relationship between the criterion and a set of predictors
- The importance of individual predictors
How is the strength of the overall relationship between the criterion and set of predictors measured?
R ^ 2
How is the importance of individual predictors measured?
β or sr
How is predictor importance indicated?
ry1^2 and ry2^2
What is R^2 with uncorrelated predictors?
ry1^2 + ry2^2
How is the correlation between predictor 1 and predictor 2 indicated?
r12^2
How is the variance in the DV accounted for by the shared variance between the predictors indicated?
ry12^2
Is R^2 smaller or larger than the sum of ry1^2 and ry2^2, and why?
Smaller, because it counts the shared variance twice
How is R^2 calculated with correlated predictors?
a + b + c
a: unique contribution of predictor 1
b: unique contribution of predictor 2
c: shared variance
What is partial correlation (pr2)?
The proportion of residual variance in the criterion uniquely accounted for by one predictor
How do you calculate partial correlation (pr2)?
pr2 = a / (a + d)
a: Unique contribution of predictor
d: Unexplained variance in the DV
What is semi-partial correlation (sr2)?
The proportion of total variance in the criterion uniquely accounted for by one predictor
