week 5: Hierarchical multiple regression

Question 1

what is hierarchical regression?

Answer 1

Predictor variables are added to a regression model in steps or blocks. This approach allows researchers to examine how the addition of certain variables changes the explained variance in a dependent variable, after accounting for other variable

Question 2

in the coefficients table, what does the t statistic tell us?

Answer 2

which predictors significantly predict the outcome

Question 3

describe how to report the beta values

Answer 3

unstandardised beta (b), standardized beta (B), t-test stat (t), significance (p)

e.g.
Year 1 mark is significantly associated with exam mark; b=-3.09, β=-.10, t=2.27, p=.025

Question 4

What does the casewise diagnostics table in SPSS tell us?

Answer 4

whether we have any extreme scores (looking for values greater than 2)

Question 5

to meet the assumptions of homoscedasticity, the residuals at each level of the predictor value should have…?

Answer 5

the same variance

Question 6

What does the F Change statistic tell us?

Answer 6

whether the new additional variables significantly increase the proportion of variance accounted for

Question 7

to compare the strength of associations between different predictor variables and the outcome variable, what statistic would you read?

Answer 7

standardised beta