Multiple Regression

Question 1

multiple regression

Answer 1

• use adjusted R2- adjusts for the number of predictors in a model
• multicollinearity: occurs when independent variables in a regression model are highly correlated

Question 2

multicollinearity

Answer 2

• if 2 or more predictor variables in your model are highly correlated with each other
• they do not provide unique/ independent information to the model
• can adversely effect regression estimates
• large amount of variance explained but no significant predictors

Question 3

identifying multicollinearity

Answer 3

• look for high correlations between variables in a correlation matrix (rule of thumb r>.80)
• tolerance statistic: percentage of variance in the IV accounted for by other IVs- high multicollinearity= low tolerance (value of .20 or less)
• variance inflation factor- indicates how much of the standard error will be inflated, VIF over 4 suggests multicollinearity

Question 4

what to do if you have multicollinearity issues

Answer 4

• increase sample size: stabilise regression coefficients
• remove redundant variables
• if 2 or more variables are important, create a variable that takes both of them into account

Question 5

what to do in R

Answer 5

• label categorical variables
• run the multiple regression
• check for multicollinearity

Question 6

summary

Answer 6

• use regression to predict outcomes
• MR tells us whether the amount of variance in the outcome predicted by the predictors is significant
• B statistics tells us how much weight or importance to assign to each predictor; you need to report the value, direction or significance of- with multiple predictors we need to check for multicollinearity