Multiple Linear Regressions L7

Question 1

What are the assumptions of multiple linear regressions only?

Answer 1

X’s should not be highly correlated

Question 2

What is the equation of multiple linear regression?

Answer 2

Y = a + b1x1+ b2x2 + b3x3 + ……… e

Question 3

Discuss the difference between unstandardised and standardised slopes?

Answer 3

Unstandardised (b’s) = raw scores (metric value)

Standardised (beta’s) = can be directly compared with each other, usually range from -1 to +1’

Question 4

How is the t ratio calculated (MLR)?

Answer 4

T1 = b1/SE of b1

Question 5

What does an F test show?

Answer 5

F test shows whether the set of X predictors significantly predicts Y

Question 6

What is the Enter method?

Answer 6

In the enter method, the predictors (X’s) are entered at the same time in the regression.
One r-squared value is obtained for the set of the predictors

Question 7

What is the Hierarchical method?

Answer 7

The predictors are entered into the regression in sequential steps.
An R-squared value is obtained for each of the combined set of predictors.

Question 8

What does an F change test aim to do?

Answer 8

It looks at the significance of r-squared change from one step to another in a hierarchical method of MLR.
A significant F change value indicates the amount of explained variance increased significantly from one step to another