Lecture 16- Multiple Regression

Question 1

How is multiple regression different from simple regression?

Answer 1

In simple regression form an equation to show the relationship between one predictor variable (x) and one to be predicted variable (y). In multiple regression we still have one to be predicted variable (y) but we can take information from multiple predictor variables (x’s)

Question 2

What tells you if the multiple regression model generated in jamovi is a good predictor? What further can be done?

Answer 2

• The P value in the top table which is labelled model fit measures (same as always less than significance level= good predictor)
• Can then look at the P values for each of the predictor variables (x’s) to see which is having an impact on the overall equation i.e. which is giving useful information. Same deal if less than P value then having a statistically significant effect. You find this information in what is known as the model coefficients table.

Question 3

What is a ANCOVA and how is it used in statistical modelling?

Answer 3

• Mixture of multiple regression and analysis of variants
• Can put numerical variables (covariants) or categorical/ nominal variables
• Again look at P values

Question 4

What is the important thing during set up of a regression model?

Answer 4

What variables should be used in the regression model.