Unit 4 · Relationships
How to run a multiple regression in SPSS
Interpret multiple regression output in SPSS: Adjusted R Square, Beta weights, significance and VIF.
Quick answer: How do you run a multiple regression in SPSS?
- Click Analyze › Regression › Linear.
- Move the outcome into Dependent and all predictors into Independent(s).
- Under Statistics… tick Collinearity diagnostics, then click OK.
- Read Adjusted R Square, the ANOVA F and each predictor’s Beta and Sig.
Report it in APA style: The predictors explained 32% of the variance in scores, F(3, 116) = 18.12, p < .001.
What is multiple regression?
Multiple regression predicts a Scale outcome from two or more predictors at once. Each predictor’s coefficient shows its effect while holding the other predictors constant, so you can see which variables make a unique contribution.
When to use it
- Several possible predictors of one Scale outcome.
- You want to control for some variables while testing others.
- A common rule of thumb is at least 10–15 participants per predictor; a power analysis is better.
How it works
The equation adds a slope for each predictor: Ŷ = b0 + b1X1 + b2X2 + …. R² shows the variance explained by all predictors together; Adjusted R² corrects for the fact that R² always rises when you add predictors. Each b is the change in Y for a one-unit rise in that predictor with the others held constant. Standardised β values let you compare predictors measured in different units.
Assumptions
- All the assumptions of simple regression (linearity, independent, normal and homoscedastic residuals).
- No multicollinearity: predictors shouldn’t be too highly correlated. VIF above 10 (or 5) is a warning.
- No highly influential cases (Cook’s distance below 1).
Worked example
Revision hours (β = .41) and anxiety (β = −.34) significantly predicted exam score, but attendance didn’t (β = .09, p = .253). Together they explained 32% of the variance, F(3, 116) = 18.12, p < .001.
Key terms
- Unique contribution
- What a predictor adds beyond the others.
- Adjusted R²
- R² penalised for the number of predictors.
- Multicollinearity
- Predictors overlapping so much that their effects can’t be separated.
- VIF
- Variance inflation factor, a multicollinearity check.
Common mistakes
- Adding many predictors to a small sample.
- Using stepwise methods to let SPSS pick predictors without theory.
- Comparing raw b values for predictors in different units: use β.
How to do it in SPSS
The SPSS syntax
Prefer typing to clicking? Open File › New › Syntax, paste this in, select it and choose Run › Selection. Every dialog's Paste button writes the same kind of code.
REGRESSION
/STATISTICS COEFF OUTS R ANOVA COLLIN TOL
/DEPENDENT score
/METHOD=ENTER hours attendance anxiety.Practice data
Download the file used in this lesson and run it yourself in SPSS. Your results will closely match the screens above, though a few numbers may differ slightly.
- exam_multi.savRevision hours, attendance, anxiety and exam score (N = 120)
Now practise it interactively →
Reading is only half of it. Tap through the real SPSS menus, read the output and pick the right write-up in 7 quick questions, with instant feedback on every answer.
Practice for this lesson is part of Pro, with every unit, unlimited hearts and saved progress.