Unit 12 · Moderation and mediation

How to run a moderation analysis in SPSS with PROCESS

Test moderation with Hayes PROCESS Model 1 in SPSS: install PROCESS, read the interaction term and conditional effects, and report it.

Free tutorialPro practiceAbout 5 minutes7 practice questions

Quick answer: How do you run a moderation analysis in SPSS with PROCESS?

  1. Install the free PROCESS macro from processmacro.org.
  2. Click Analyze › Regression › PROCESS, then enter Y, X and the moderator W.
  3. Choose Model number 1, set 5000 bootstrap samples and mean-centre the product terms under Options.
  4. Read the interaction term (Int_1) and the conditional effects.

Report it in APA style: Social support moderated the effect of stress on wellbeing, b = 0.11, 95% CI [0.03, 0.19], p = .007.

Last updated November 2026 · Not sure this is the right test? Try the Which statistical test? tool.

What is moderation?

Moderation happens when the strength or direction of a relationship between X and Y depends on a third variable, the moderator (W). It answers questions about when or for whom an effect occurs, such as whether social support buffers the effect of stress on wellbeing.

When to use it

  • You expect an effect to be stronger, weaker or reversed for some people.
  • Testing buffering or amplifying factors.

How it works

Moderation is tested with an interaction term: X × W added to a regression with X and W. If the interaction is significant, the effect of X changes across values of W. PROCESS Model 1 does this automatically, mean-centres the variables if you ask it to, and gives conditional effects (simple slopes): the effect of X at low (−1 SD), average and high (+1 SD) values of the moderator.

Assumptions

  • The same assumptions as multiple regression.
  • Mean-centring X and W makes the lower-order coefficients easier to interpret.

Worked example

Social support moderated the effect of stress on wellbeing, b = 0.11, 95% CI [0.03, 0.19], p = .007. Stress was linked to much lower wellbeing when support was low (b = −0.95) than when it was high (b = −0.29).

Key terms

Moderator
A variable that changes the X → Y relationship.
Interaction term
The product of X and W.
Simple slopes
The X → Y effect at specific values of W.
Mean-centring
Subtracting the mean so zero represents an average score.

Common mistakes

  • Confusing moderation with mediation.
  • Interpreting the main effects of X and W as if there were no interaction.
  • Not probing a significant interaction with simple slopes or a plot.

How to do it in SPSS

The PROCESS dialog

Moderation asks when or for whom an effect happens: does social support soften the effect of stress on wellbeing? Most psychology courses use Andrew Hayes’ free PROCESS macro for this.

PROCESS v4 by Andrew F. Hayes
Variables
Participant ID [id]
Age in years [age]
▶
Y variable:1
wellbeing
▶
X variable:2
stress
▶
Moderator variable W:3
support
Model number: 14
Number of bootstrap samples: 5000
Options5MulticategoricalLong variable names
OKPasteResetCancelHelp
  1. 1
    Y variableThe outcome.
  2. 2
    X variableThe predictor whose effect might change.
  3. 3
    Moderator WThe variable that might change the X → Y effect.
  4. 4
    Model numberModel 1 is simple moderation. Hayes’ book numbers every model; Model 4 is mediation.
  5. 5
    OptionsChoose to mean-centre the variables that define products, and to generate data for plotting the interaction.

PROCESS isn’t built into SPSS. Download it free from processmacro.org, then install the custom dialog (.spd) via Extensions › Utilities › Install Custom Dialog. Your version number may differ.

Reading moderation output

The key line is the interaction term. If it’s significant, look at the conditional effects to see how the effect changes.

Output1 [Document1] – SPSS Statistics Viewer
FileEditViewDataTransformAnalyzeGraphsUtilitiesExtensionsWindowHelp
Output
Log
PROCESS
Model Summary
Model
Interaction test
Conditional effects
Model
coeffsetpLLCIULCI
constant60.12.33182.18.00059.4760.77
stress−.62.11−5.64.000−.84−.40
support.48.095.33.000.30.66
Int_11.11.042.75.007.03.19

Product terms key: Int_1 : stress x support

Test(s) of highest order unconditional interaction(s)
R2-chngFdf1df2p
X*W.02827.561.00196.00.007
Conditional effects of the focal predictor at values of the moderator(s)
supportEffectsetpLLCIULCI
−3.00−.953.14−6.79.000−1.23−.67
.00−.62.11−5.64.000−.84−.40
3.00−.294.13−2.16.032−.55−.03

Moderator values are the mean and ±1 SD (support was mean-centred).

  1. 1
    Int_1The stress × support product. b = .11, p = .007: support significantly moderates the effect of stress.
  2. 2
    R2-chngThe interaction adds 2.8% to the variance explained, F(1, 196) = 7.56.
  3. 3
    Low supportAt 1 SD below the mean, each point of stress lowers wellbeing by .95.
  4. 4
    High supportAt 1 SD above the mean the effect shrinks to −.29. Support buffers the effect of stress.

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.

* Run process.sps (from processmacro.org) once per session first, then:
process y=wellbeing/x=stress/w=support/model=1/center=1/boot=5000.

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.

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.

More in Moderation and mediation