Correlation and regression

Serial Mediation Analysis

Test a causal chain such as X → M1 → M2 → Y. Every specific indirect effect, the total indirect effect and pairwise contrasts are tested with bootstrap confidence intervals.

Free · No sign-upReplaces SPSS PROCESS macroWorks with your data

Updated 13 October 2026 · By Bibin Xavier

Data 250 × 9

Paste from Excel, or drop a CSV, TSV or Excel file here.Upload file

Example data 250 rows, 9 variables. Data stays in this browser tab.

Employeetext · 250 levels	Leadershipnumeric	Trustnumeric	Engagementnumeric	Autonomynumeric	Tenure_Yearsnumeric	Performancenumeric	Agenumeric	Femalenumeric · coded
E001	4.56	3.94	4.12	3.21	4.9	5.36	28	1
E002	4.42	3.02	4.35	3.12	10	4.26	43	1
E003	4.82	4.94	4.78	4.23	19.4	4.06	21	0
E004	4.65	4.08	3.53	3.37	5.4	4.22	21	0
E005	6.43	4.58	5.35	3.31	10.9	6.12	44	0
E006	4.86	5.49	5.41	4.61	15.5	4.72	38	0
E007	2.9	3.79	5.55	2.84	12.2	4.86	35	0
E008	4.6	4.11	3.5	4.37	2.3	5.09	39	0

Variables and options

PROCESS model

X (predictor)

Y (outcome)

Mediators (M) 2 selected

For serial models, select mediators in causal order.

Covariates (optional) 0 selected

Covariates enter every equation, as in PROCESS.

Bootstrap samples

Confidence level

Standard errors

Results update automatically when you change an option.

Results

250 rows · example data

PROCESS model6Serial mediation

Sample size250

Bootstrap samples5,000percentile 95% CI

Standard errorsOLS

Conceptual diagram: Model 6

Table 1

Outcome: Trust

Predictor	coeff	se	t	p	LLCI	ULCI
constant	2.6007	0.2372	10.963	< .001	2.1335	3.0679
Leadership	0.3941	0.0517	7.627	< .001	0.2923	0.4959

Note. R = .4359, R² = .1900, MSE = 0.4988, F(1, 248) = 58.173, p < .001.

Table 2

Outcome: Engagement

Predictor	coeff	se	t	p	LLCI	ULCI
constant	1.7884	0.3781	4.730	< .001	1.0437	2.5331
Leadership	0.3277	0.0751	4.363	< .001	0.1798	0.4756
Trust	0.3293	0.0831	3.964	< .001	0.1657	0.4929

Note. R = .4465, R² = .1994, MSE = 0.8535, F(2, 247) = 30.759, p < .001.

Table 3

Outcome: Performance

Predictor	coeff	se	t	p	LLCI	ULCI
constant	1.7918	0.3279	5.465	< .001	1.1461	2.4376
Leadership	0.0991	0.0647	1.531	.127	-0.0284	0.2266
Trust	0.3281	0.0711	4.613	< .001	0.1880	0.4682
Engagement	0.2447	0.0528	4.631	< .001	0.1406	0.3487

Note. R = .5245, R² = .2751, MSE = 0.5884, F(3, 246) = 31.113, p < .001.

Table 4

Total effect of Leadership on Performance

Effect	se	t	p	LLCI	ULCI
0.3404	0.0620	5.490	< .001	0.2183	0.4625

Table 5

Direct effect of Leadership on Performance

Effect	se	t	p	LLCI	ULCI
0.0991	0.0647	1.531	.127	-0.0284	0.2266

Table 6

Indirect effect: Leadership → Trust → Performance

Effect	BootSE	BootLLCI	BootULCI	Completely standardized
0.1293	0.0311	0.0699	0.1923	0.1251

Table 7

Indirect effect: Leadership → Engagement → Performance

Effect	BootSE	BootLLCI	BootULCI	Completely standardized
0.0802	0.0212	0.0431	0.1271	0.0775

Table 8

Indirect effect: Leadership → Trust → Engagement → Performance

Effect	BootSE	BootLLCI	BootULCI	Completely standardized
0.0318	0.0111	0.0138	0.0567	0.0307

Table 9

Total indirect effect

Effect	BootSE	BootLLCI	BootULCI
0.2413	0.0411	0.1640	0.3257

Table 10

Pairwise contrasts between indirect effects

Contrast	Effect	BootSE	BootLLCI	BootULCI
(Leadership → Trust → Performance) − (Leadership → Engagement → Performance)	0.0491	0.0382	-0.0251	0.1238
(Leadership → Trust → Performance) − (Leadership → Trust → Engagement → Performance)	0.0976	0.0322	0.0353	0.1603
(Leadership → Engagement → Performance) − (Leadership → Trust → Engagement → Performance)	0.0484	0.0210	0.0100	0.0933

Note. A CI excluding zero means the two indirect effects differ.

Interpretation

All of 3 indirect effects are significant: Leadership → Trust → Performance (0.129); Leadership → Engagement → Performance (0.080); Leadership → Trust → Engagement → Performance (0.032). The direct effect of Leadership is not statistically significant, which indicates indirect-only (full) mediation.

The indirect effect Leadership → Trust → Performance was significant, b = 0.129, 95% CI [0.070, 0.192]; The indirect effect Leadership → Engagement → Performance was significant, b = 0.080, 95% CI [0.043, 0.127]; The indirect effect Leadership → Trust → Engagement → Performance was significant, b = 0.032, 95% CI [0.014, 0.057], based on 5,000 bootstrap samples (PROCESS Model 6).

How to cite this tool

Using these results in a thesis, journal article, report or class assignment? Please cite this free tool so others can find it. It keeps the site free.

APA 7

Xavier, B. (2026). Serial Mediation Analysis [Web application]. FreeStatisticalTools.com. https://freestatisticaltools.com/serial-mediation-analysis/

BibTeX

@software{xavier_serialmediation_2026,
  author = {Xavier, Bibin},
  title = {Serial Mediation Analysis},
  publisher = {FreeStatisticalTools.com},
  year = {2026},
  url = {https://freestatisticaltools.com/serial-mediation-analysis/},
  note = {Web application}
}

Suggested sentence for your methods section: “The analysis was carried out with the free Serial Mediation Analysis tool on FreeStatisticalTools.com (Xavier, 2026).”

What is serial mediation?

Serial mediation tests a causal chain of mediators, for example X → M1 → M2 → Y. Each specific indirect effect, including the full chain, is estimated with bootstrap confidence intervals, together with the total indirect effect and pairwise contrasts.

When to use serial mediation

Use serial mediation when mediators influence each other in a fixed order, for example leadership builds trust, trust builds engagement, and engagement raises performance.

Assumptions

Linear relationships and correctly specified causal ordering (X before M before Y), justified by theory or design.
Independent observations and, for OLS standard errors, constant residual variance (choose HC3 if this is doubtful).
No unmeasured confounding of the paths between mediators and the outcome.

How to set up your data

One row per case and one numeric column per variable. Code dichotomous variables 0/1. Moderators with two values are probed at those two values; continuous moderators at the 16th, 50th and 84th percentiles, or the mean ± 1 SD.

How to use this tool

Enter your data. Paste columns from Excel, Google Sheets or SPSS with variable names in the first row, or upload a CSV or Excel file. The example data shows the expected layout.
Choose the variables. Select the x (predictor), y (outcome), mediators (m) and covariates (optional).
Set the options. Fill in process model, bootstrap samples, confidence level and standard errors.
Read the results. Results update automatically. Read the plain-English interpretation, then use Copy table to paste APA-style tables into Word and Copy for the results sentence.

How to report serial mediation in APA style

Report each specific indirect effect, especially the full chain X → M1 → M2 → Y, with its bootstrap CI.

Frequently asked questions

How many indirect effects does serial mediation have?

With two mediators there are three: X → M1 → Y, X → M2 → Y and X → M1 → M2 → Y. With k mediators there are 2^k − 1.

Does the order of mediators matter?

Yes. Select the mediators in their hypothesized causal order; the first can affect the second, but not the other way round.

Can I add covariates?

Yes. Covariates enter every equation, as in PROCESS.

Related tools

PROCESS is a macro for SPSS, SAS and R written by Andrew F. Hayes. This tool is an independent implementation of the published regression methods and model numbering (Hayes, 2022) and is not affiliated with, endorsed by or derived from the PROCESS software.

Browse all categories

Data cleaning and preparation · Describe and explore · Compare means · Nonparametric tests · Correlation and regression · Bayesian statistics · Categorical data · Reliability and agreement · Multivariate analysis · Structural equation modeling · Econometrics and time series · Panel data and causal inference · Marketing and decision research · Survival, clinical and meta-analysis · Study planning and calculators · PLS-SEM