Lecture 15 - Correlation and Multiple Regression Practice

Question 1

Can you analyse the same dataset using simultaneous, stepwise and hierarchical regression?

Answer 1

Yes!

Question 2

How are variables added in hierarchical regression?

Answer 2

Increasing, eg 2 variables then 7 then 9

Question 3

How are variables added in stepwise regression

Answer 3

Decreasing, eg 6 then 5 then 4

Keeps the one with highest effect and removes it, keeps doing so until removing a variable has very little/no effect

Question 4

How do we calculate the f-ratio?

Answer 4

MEAN SQUARES!!!

bgMS

Question 5

What is R in multiple regression?

Answer 5

The multiple correlation coefficient. It is one measure among a few that tells us about how well the IVs predict the DV

Question 6

What is R squared in multiple regression?

Answer 6

The coefficient of determination

Tells us what proportion of the variance of the DV can be explained by the IVs

There’s also an adjusted r squared which we use in order to report our data more accurately

Question 7

What is the purpose of unstandardised coefficients in linear regression

Answer 7

They indicate how much of the DV varies with one IV when all other IV’s are held constant

Question 8

In regression what do the individual significance of the individual IVs tell us

Answer 8

This tells us wether each IV contributes significantly to predicting the DV