Mediation & Moderated Mediation Visualizer

Causal Inference Post-Estimation

Visualize how effects propagate through mediation and moderated mediation models. Enter your calibrated coefficients from prior analysis and explore direct, indirect, and total effects interactively.

👨‍🏫 Professor Mode: Guided Learning Experience

New to mediation analysis? Enable Professor Mode for step-by-step guidance through understanding direct, indirect, and total effects in mediation and moderated mediation models!

OVERVIEW

Mediation analysis examines how an effect occurs by testing whether variable M (mediator) transmits the effect of X on Y. Moderated mediation adds complexity by allowing these pathways to vary across levels of a moderator W.

💡 This is a visualization tool, not an estimator. Enter coefficients and standard errors from your statistical software (SPSS PROCESS, Stata, R mediation, etc.) to explore the structural relationships visually.

📖 What this tool does
  • Accepts calibrated path coefficients and SEs from your analysis
  • Dynamically propagates structural equations as you adjust X and W
  • Visualizes direct, indirect, and total effect decomposition
  • Approximates uncertainty using delta-method formulas
  • Generates conditional effect plots for moderated mediation
📘 Supported Model Types

Model A: Single Mediator

$$M = aX$$

$$Y = c'X + bM$$

Optional moderation at a-path, b-path, or direct path (c′)

Model B: Parallel Mediators

$$M_1 = a_1 X, \quad M_2 = a_2 X$$

$$Y = c'X + b_1 M_1 + b_2 M_2$$

Two independent mediators (no moderation in v1.0)

MARKETING SCENARIOS

Load calibrated parameters from realistic marketing research contexts. Each scenario includes coefficients and uncertainty estimates.

INPUTS & SETTINGS

Model Configuration

When enabled, path coefficients become conditional on moderator W.

📥 Import from PROCESS / Mplus / lavaan output

Paste output from your statistical software to auto-populate coefficients.

Download JSON template:

Path Coefficients

Enter the unstandardized coefficients and standard errors from your analysis output.

a-path (X → M)

a(W) = a₁ = 0.50

b-path (M → Y)

b(W) = b₁ = 0.60

c′-path (X → Y, direct)

c′(W) = c′₁ = 0.30

INTERACTIVE CONTROLS

VISUALIZATIONS

X → M Relationship

Slope = a(W). Vertical error bar = SE(M̂) = |X| × SE(a). ℹ️ See formulas

Predicted M̂ → Y Relationship

Slope = b(W). Horizontal error = SE(M̂). Vertical = propagated SE via delta method. ℹ️ See formulas

Structural Diagram

Arrows show path coefficients. Blue = direct, Orange = indirect pathway.

X → Y Effect Decomposition

Blue = Direct, Orange = Indirect, Black = Total. Error bars = SE × |X|. ℹ️ Formulas

Effect Decomposition Summary

Error bars show 95% confidence intervals. Effects computed at current X and W values.

Effect Decomposition at Current Values

Direct Effect (c′) 0.30 [0.06, 0.54]
Indirect Effect (ab) 0.30 [0.16, 0.44]
Total Effect (c) 0.60 [0.35, 0.85]

INTERPRETATION

Adjust the coefficients and sliders above to see an interpretation here.

TECHNICAL DOCUMENTATION

📐 Model Equations

Single Mediator Model

$$M = a(W) \cdot X$$ $$Y = c'(W) \cdot X + b(W) \cdot M$$

Conditional Path Coefficients (when moderated)

$$a(W) = a_1 + a_3 W$$ $$b(W) = b_1 + b_3 W$$ $$c'(W) = c'_1 + c'_3 W$$

Parallel Mediator Model

$$M_1 = a_1 X, \quad M_2 = a_2 X$$ $$Y = c'X + b_1 M_1 + b_2 M_2$$

📊 Effect Definitions
Effect Formula Interpretation
Direct Effect $c'(W)$ Effect of X on Y not through M
Indirect Effect $a(W) \times b(W)$ Effect of X on Y transmitted through M
Total Effect $c'(W) + a(W) \times b(W)$ Combined direct + indirect effects

For parallel mediators, total indirect effect = $a_1 b_1 + a_2 b_2$

📈 Uncertainty Approximation

Converting CI to SE

$$SE = \frac{\text{Upper} - \text{Lower}}{2 \times 1.96}$$

Variance of Conditional Path Coefficients

$$\text{Var}(a(W)) = \text{Var}(a_1) + W^2 \cdot \text{Var}(a_3)$$

Similarly for b(W) and c′(W). Independence of coefficients is assumed.

Indirect Effect Variance (Delta Method)

$$SE_{IE(W)} \approx \sqrt{b(W)^2 \cdot \text{Var}(a(W)) + a(W)^2 \cdot \text{Var}(b(W))}$$

Total Effect Variance

$$SE_{TE(W)} \approx \sqrt{\text{Var}(c'(W)) + \text{Var}(IE(W))}$$

Propagated Uncertainty in Visualizations

The charts show error bars that propagate uncertainty through the mediation path:

X → M Plot (vertical error): Uncertainty in predicted mediator

$$SE(\hat{M}) = |X| \cdot SE(a)$$

M̂ → Y Plot: Uncertainty propagates through both the a-path and b-path

  • Horizontal error: $SE(\hat{M}) = |X| \cdot SE(a)$
  • Vertical error (delta method): $$SE(\hat{Y}_{via \hat{M}}) = \sqrt{(b \cdot SE(\hat{M}))^2 + (\hat{M} \cdot SE(b))^2}$$

This combines: (1) uncertainty in where M̂ lands, and (2) uncertainty in the b slope.

X → Y Decomposition (vertical errors): Effect uncertainty scaled by X

$$SE(\hat{Y}_{direct}) = |X| \cdot SE(c')$$ $$SE(\hat{Y}_{indirect}) = |X| \cdot SE_{IE}$$ $$SE(\hat{Y}_{total}) = |X| \cdot SE_{TE}$$

⚠️ Important Limitation: These are approximate standard errors using the delta method with assumed independence. For formal inference, use bootstrap CIs from your estimation software.

🔧 Version & Boundaries

Version 1.0 Capabilities

  • ✔ Single mediator model
  • ✔ Optional moderation at any path
  • ✔ Parallel mediation (two mediators)
  • ✔ Approximate uncertainty via delta method
  • ✔ Conditional effect plots

Not Included in v1.0

  • ✘ Multiple moderators
  • ✘ Serial mediation (chained mediators)
  • ✘ Covariance-aware inference
  • ✘ Bootstrap resampling engine