PROCESS macro (Hayes) for moderation and mediation

Description

Using Hayes's PROCESS models to test whether a UX effect operates through a mechanism (mediation) or depends on user characteristics (moderation), with bootstrapped inference.

PROCESS is a regression-based macro for SPSS, SAS and R that implements over 90 numbered templates for mediation, moderation and their combinations (conditional process analysis). Model 4 tests simple mediation (X to M to Y), Model 1 simple moderation (X by W predicting Y), and Models 7 and 14 moderated mediation. It estimates indirect effects with percentile bootstrap confidence intervals, probes interactions with simple slopes and Johnson-Neyman regions, and reports an index of moderated mediation. Rigour comes from specifying the model from theory before analysis, mean-centring where appropriate, using at least 5,000 bootstrap samples, and recognising that cross-sectional mediation cannot establish temporal or causal ordering; the X to M to Y order must be justified by design.

Attributes

Loss type
Siloed
Stage
Evaluation
Effort
Days

Steps

  1. Specify the conceptual model (which variable is X, M, W, Y and covariates) and choose the matching PROCESS model number.
  2. Install PROCESS v4 and run it from syntax, e.g. process y=sat/x=design/m=ease/model=4/boot=5000/seed=2026.
  3. For moderation, request simple slopes at the 16th, 50th and 84th percentiles of W and the Johnson-Neyman region.
  4. Interpret indirect effects as significant only if the 95% bootstrap CI excludes zero; for moderated mediation check the index of moderated mediation CI.
  5. Check standard regression assumptions and use heteroscedasticity-consistent standard errors (hc3) where needed.
  6. Report path coefficients, the indirect effect with CI and a diagram, and state the causal assumptions.

Sources

  • SPSS REGRESSION
  • PLUM
  • LOGISTIC REGRESSION procedures; Hayes PROCESS macro v4
  • Hayes, A. F. (2022) Introduction to Mediation, Moderation, and Conditional Process Analysis, 3rd ed.
  • Baron, R. M. & Kenny, D. A. (1986) The moderator-mediator variable distinction. JPSP
  • Preacher, K. J. & Hayes, A. F. (2008) Asymptotic and resampling strategies for assessing and comparing indirect effects. Behavior Research Methods
  • The Inquiry Series, C8 — SPSS for Regression, Moderation & Mediation

Also known as

  • Conditional process analysis
  • Hayes PROCESS

Categories

  • Analytics & Behaviour
  • Statistics & Modelling