Parallel Mediation (k mediators)

Verified

Mediation (legacy hub)

Independently verified. Every statistic this test reports has been re-derived against an independent reference — never the library the pipeline itself calls — the rendered output was read back in a browser, and the result is locked with a committed regression suite.

Run this test straight away on a free built-in teaching dataset — no data of your own needed — or bring your own. Either opens the guided workspace: variable setup, assumption diagnostics, results with effect sizes and confidence intervals, figures, and APA-ready reporting.

Loading teaching datasets…

Or use your own dataset

Loading your datasets…

Tests several mediators operating side-by-side between X and Y, and compares their indirect effects.

Parallel multiple mediation estimates each mediator's specific indirect effect (controlling for the others) plus the total indirect effect, each with a bootstrap CI.

Worked example

Does a programme improve outcomes via BOTH motivation and skills?

Programme → {motivation, skills} → performance was tested with parallel mediation (5,000 bootstraps).

Result

Motivation carried a significant indirect effect (0.13, 95% CI [0.07, 0.20]); skills did not (−0.01, CI included 0).

How you'd report it (APA)

In a parallel mediation, motivation significantly mediated the effect (0.13, 95% CI [0.07, 0.20]) while skills did not.

Try it yourself: Load this ready-made sample and follow the run above.

When to use it

  • Several candidate mechanisms at once
    You want to know which mediator carries the effect. e.g. a programme → {motivation, skills} → performance.
  • The mediators are contemporaneous
    They sit side-by-side between X and Y and are not assumed to cause one another.

When NOT to — use instead

Hypotheses

H₀: each specific indirect effect a_j·b_j (through mediator M_j, adjusting for the others) is zero.
Hₐ: at least one specific indirect effect is non-zero. Each is tested by its own bootstrap CI; a mediator mediates when its CI excludes 0.

Parameter tested: each mediator's specific indirect effect a_j·b_j, plus the total indirect effect, all with percentile bootstrap CIs

Assumptions (and what to do if they fail)

Mediators do not cause each otherhigh

Check: If M1 plausibly drives M2, the parallel model mis-specifies the system.

If violated: The specific indirect effects are biased; a serial model is the correct one.

No unmeasured M–Y confoundinghigh

Check: Each mediator–outcome link needs no omitted common cause.

If violated: The corresponding specific indirect effect is biased.

Mediators not near-collinearmedium

Check: Strongly correlated mediators inflate the SEs of their specific effects.

If violated: A mediator can look non-significant only because it shares variance with another. Report the correlations.

Ready to run a Parallel Mediation (k mediators) on your own data?

Guided setup, automatic assumption checks, effect sizes, figures and an APA write-up.

Run this test →