Determining Power and Sample Size for Simple and Complex Mediation Models
The DOI matches its Crossref record; 30 of the 31 references it cites were found in a registry or an index.
Abstract
Mediation analyses abound in social and personality psychology. Current recommendations for assessing power and sample size in mediation models include using a Monte Carlo power analysis simulation and testing the indirect effect with a bootstrapped confidence interval. Unfortunately, these methods have rarely been adopted by researchers due to limited software options and the computational time needed. We propose a new method and convenient tools for determining sample size and power in mediation models. We demonstrate our new method through an easy-to-use application that implements the method. These developments will allow researchers to quickly and easily determine power and sample size for simple and complex mediation models.
Overview
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
In the citing paper
“We recommend that researchers estimate sample size requirements for mediational models prior to data collection (Schoemann et al., 2017), which addresses many but not all problems with mediation (see Götz et al., 2021).”
From Acceptance and Commitment Therapy Processes and Mediation: Challenges and How to Address Them (2023) · cites this paper as Schoemann et al., 2017
In the citing paper
“A Monte Carlo power analysis was conducted using Schoemann et al’s (2017) web-based tool MCpowrMed which suggested that the sample size provided sufficient power (.93) for simple mediation analyses.”
From The protective effects of social support and family functioning on parenting stress among Hispanic/Latino/a American immigrant parents with traumatic life experiences: A mediation analysis (2023) · cites this paper as Schoemann et al’s (2017)
In the citing paper
“We used Monte Carlo power analysis for indirect effects [61] (https://schoemanna.shinyapps.io/mc_power_med/, accessed on 5 August 2021) to determine the sample size for our proposed mediational model. We needed at least 163 participants to reach a power of 0.80, assuming correlations of r = 0.30 (SD = 1.00) among the predictor (independent variable), the mediator, and the dependent variable.”
From Power Distance Belief and Workplace Communication: The Mediating Role of Fear of Authority (2022) · cites this paper as [61]
Citations
Cites
Baron, R. M., & Kenny, D. A. (1986) · Journal of Personality and Social Psychology · doi:10.1037/0022-3514.51.6.1173 · Cited by 91,532 · cited 1×
“Historically, mediation was examined with the “causal steps” approach by satisfying four criteria described by Baron and Kenny (1986) through a series of regression models.”
Method(2013) · Cited by 44,563
Statistical power analyses using G∗Power 3.1: tests for correlation and regression analyses
Faul, F., Erdfelder, E., Buchner, A., & Lang, A.-G. (2009) · Behavior Research Methods · doi:10.3758/brm.41.4.1149 · Cited by 37,615 · cited 3×
“In addition, many of these methods have been implemented in user-friendly software such as G*Power (Faul et al., 2009) and Optimal Design (Raudenbush et al., 2011).”
MethodAsymptotic Confidence Intervals for Indirect Effects in Structural Equation Models
Sobel, M. E. (1982) · Sociological Methodology · doi:10.2307/270723 · Cited by 12,810 · cited 1×
“Sobel (1982) proposed a formula for calculating the standard error of ab, SEab, which permits calculation of a z-score statistic (ab/SEab) as well as a confidence interval for ab.”
MethodBeyond Baron and Kenny: Statistical Mediation Analysis in the New Millennium
Hayes, A. F. (2009) · Communication Monographs · doi:10.1080/03637750903310360 · Cited by 9,757 · cited 2×
“Although the causal steps approach has been widely popular, research suggests that it is low in power to detect mediation (MacKinnon, Lockwood, Hoffman, West, & Sheets, 2002) and is no longer considered best practice (Hayes, 2009).”
BackgroundA comparison of methods to test mediation and other intervening variable effects
MacKinnon, D. P., Lockwood, C. M., Hoffman, J. M., West, S. G., & Sheets, V. (2002) · Psychological Methods · doi:10.1037/1082-989x.7.1.83 · Cited by 9,406
Estimating the reproducibility of psychological science
Open Science Collaboration (2015) · Science · doi:10.1126/science.aac4716 · Cited by 8,515
(2010) · Cited by 7,034
Introduction to Statistical Mediation Analysis
MacKinnon, D. P. (2012) · doi:10.4324/9780203809556 · Cited by 5,618 · cited 3×
“Several authors have provided detailed reviews of mediation analysis (Gunzler, Chen, Wu, & Zhang, 2013; Hayes, 2009, 2013; MacKinnon, 2008; Preacher, 2015; Rucker, Preacher, Tormala, & Petty, 2011), but discussions on power analysis and sample size calculations for these models are relatively sparse.”
BackgroundRequired sample size to detect the mediated effect
Fritz, M. S., & MacKinnon, D. P. (2007) · Psychological Science · doi:10.1111/j.1467-9280.2007.01882.x · Cited by 4,810
Cole, D. A., & Maxwell, S. E. (2003) · Journal of Abnormal Psychology · doi:10.1037/0021-843x.112.4.558 · Cited by 3,605 · cited 1×
“Longitudinal data, or application of additional model assumptions, can lead to stronger causal claims about the relationships between variables (Cole & Maxwell, 2003; Preacher, 2015).”
BackgroundMediation Analysis in Social Psychology: Current Practices and New Recommendations
Rucker, D. D., Preacher, K. J., Tormala, Z. L., & Petty, R. E. (2011) · Social and Personality Psychology Compass · doi:10.1111/j.1751-9004.2011.00355.x · Cited by 2,430
Hayes, A. F., & Scharkow, M. (2013) · Psychological Science · doi:10.1177/0956797613480187 · Cited by 2,387 · cited 2×
“Thus, researchers interested in balancing power and Type I error rate to test the indirect effect should consider the Monte Carlo confidence interval (Hayes & Scharkow, 2013).”
BackgroundMethodHow to Use a Monte Carlo Study to Decide on Sample Size and Determine Power
Muthén, L. K., & Muthén, B. O. (2002) · Structural Equation Modeling: A Multidisciplinary Journal · doi:10.1207/s15328007sem0904_8 · Cited by 2,362
Measuring the Prevalence of Questionable Research Practices With Incentives for Truth Telling
John, L. K., Loewenstein, G., & Prelec, D. (2012) · Psychological Science · doi:10.1177/0956797611430953 · Cited by 2,239
Advantages of Monte Carlo confidence intervals for indirect effects
Preacher, K. J., & Selig, J. P. (2012) · Communication Methods and Measures · doi:10.1080/19312458.2012.679848 · Cited by 1,908 · cited 2×
“Monte Carlo confidence intervals, also known as parametric bootstrap confidence intervals, assume normality of the regression coefficients a and b but not the product of the two terms2 (Preacher & Selig, 2012).”
BackgroundThe unicorn, the normal curve, and other improbable creatures.
Micceri, T. (1989) · Psychological Bulletin · doi:10.1037//0033-2909.105.1.156 · Cited by 1,485 · cited 1×
“Extending the app to allow for the specification of nonnormal variables, which tend to be normative in psychology (Micceri, 1989), would also enhance the accuracy of the app.”
BackgroundDirect and Indirect Effects: Classical and Bootstrap Estimates of Variability
Bollen, K. A., & Stine, R. (1990) · Sociological Methodology · doi:10.2307/271084 · Cited by 1,305 · cited 1×
“However, Bollen and Stine (1990) have noted the distribution of ab can deviate from normality, particularly in smaller samples, and methods that do not assume normality of the indirect effect are preferred.”
BackgroundMediation Models for Longitudinal Data in Developmental Research
Selig, J. P., & Preacher, K. J. (2009) · Research in Human Development · doi:10.1080/15427600902911247 · Cited by 804 · cited 1×
“Implementing models that are the most common in and relevant to social psychology research, such as dyadic mediation models (Ledermann, Macho, & Kenny, 2011), longitudinal mediation models (Selig & Preacher, 2009), or models combining mediation and moderation (Hayes, 2013), is our priority.”
BackgroundAdvances in Mediation Analysis: A Survey and Synthesis of New Developments
Preacher, K. J. (2015) · Annual Review of Psychology · doi:10.1146/annurev-psych-010814-015258 · Cited by 793 · cited 3×
“Several authors have provided detailed reviews of mediation analysis (Gunzler, Chen, Wu, & Zhang, 2013; Hayes, 2009, 2013; MacKinnon, 2008; Preacher, 2015; Rucker, Preacher, Tormala, & Petty, 2011), but discussions on power analysis and sample size calculations for these models are relatively sparse.”
BackgroundAssessing Mediation in Dyadic Data Using the Actor-Partner Interdependence Model
Ledermann, T., Macho, S., & Kenny, D. A. (2011) · Structural Equation Modeling: A Multidisciplinary Journal · doi:10.1080/10705511.2011.607099 · Cited by 709
Introduction to mediation analysis with structural equation modeling.
(2013) · PubMed Central · doi:10.3969/j.issn.1002-0829.2013.06.009 · Cited by 661
Power of the Likelihood Ratio Test in Covariance Structure Analysis
Satorra, A., & Saris, W. E. (1985) · Psychometrika · doi:10.1007/bf02294150 · Cited by 485 · cited 1×
“…been applied to many types of statistical models and research designs including linear regression, the generalized linear model (Faul, Erdfelder, Buchner, & Lang, 2009), randomized trials and cluster randomized trails (Spybrook et al., 2011), and structural equation models (Satorra & Saris, 1985).”
MethodPower Analysis for Complex Mediational Designs Using Monte Carlo Methods
Thoemmes, F., MacKinnon, D., & Reiser, M. (2010) · Structural Equation Modeling: A Multidisciplinary Journal · doi:10.1080/10705511.2010.489379 · Cited by 298 · cited 1×
“There are a few different quantities that will meet this criterion, including model parameter estimates (Selig & Preacher, 2008; Zhang, 2014), measures of variance explained (Thoemmes et al., 2010), and correlation or covariance matrices.”
BackgroundMonte Carlo Confidence Intervals for Complex Functions of Indirect Effects
Tofighi, D., & MacKinnon, D. P. (2015) · Structural Equation Modeling A Multidisciplinary Journal · doi:10.1080/10705511.2015.1057284 · Cited by 174 · cited 1×
“Tofighi and MacKinnon (2016) examined power and Type I error in a more complex mediation model (i.e., X!”
BackgroundMonte Carlo based statistical power analysis for mediation models: methods and software
Zhang, Z. J. (2013) · Behavior Research Methods · doi:10.3758/s13428-013-0424-0 · Cited by 159 · cited 4×
“Implementing a Monte Carlo Power analysis for mediation models requires knowledge of specific software (e.g., R, or Mplus), is computationally intensive (Zhang, 2014), and can prove difficult as users must specify all population parameters for a specific model of interest.”
BackgroundMethodInfluentialVazire, S. (2015) · Social Psychological and Personality Science · doi:10.1177/1948550615603955 · Cited by 105
Number of Replications Required in Monte Carlo Simulation Studies: A Synthesis of Four Studies
Mundform, D. J., Schaffer, J. R., Kim, M., Shaw, D. J., Thongteeraparp, A., & Supawan, P. (2011) · Journal of Modern Applied Statistical Methods · doi:10.22237/jmasm/1304222580 · Cited by 100
Using Monte Carlo simulations to determine power and sample size for planned missing designs
Schoemann, A. M., Miller, P., Pornprasertmanit, S., & Wu, W. (2014) · International Journal of Behavioral Development · doi:10.1177/0165025413515169 · Cited by 51
Self-citationMcConnell, A. R. (2012) · Social Psychological and Personality Science · doi:10.1177/1948550612469596 · Cited by 3
Research works printed but not linked: 1
- Gunzler D., 2013, Shanghai Archives of Psychiatry, 25, 390
Cited by
Lakens, D. (2022) · Collabra Psychology · doi:10.1525/collabra.33267 · Cited by 1,962
Power Analysis for Parameter Estimation in Structural Equation Modeling: A Discussion and Tutorial
Wang, Y. A., & Rhemtulla, M. (2021) · Advances in Methods and Practices in Psychological Science · doi:10.1177/2515245920918253 · Cited by 544
Stein, J.-P., Linda Breves, P., & Anders, N. (2024) · New Media & Society · doi:10.1177/14614448221102900 · Cited by 393
“Based on our planned statistical analyses, we used the Monte Carlo Power Analysis for Indirect Effects method provided by Schoemann et al. (2017) to calculate the minimum sample size for the current study.”
MethodA Practical Primer To Power Analysis for Simple Experimental Designs
Perugini, M., Gallucci, M., & Costantini, G. (2018) · International Review of Social Psychology · doi:10.5334/irsp.181 · Cited by 408
“Said otherwise, the study Figure 10: Shiny app for Monte Carlo power analysis (Schoemann et al., 2017).”
MethodSample Size Requirements for Simple and Complex Mediation Models
Sim, M., Kim, S.-Y., & Suh, Y. (2022) · Educational and Psychological Measurement · doi:10.1177/00131644211003261 · Cited by 269
“As discussed, several studies have investigated sample sizes in mediation analysis (e.g., Fritz & MacKinnon, 2007; Liu & Wang, 2019; Schoemann et al., 2017).”
BackgroundMethod
Metadata
- Authors
- Schoemann, A. M., Boulton, A. J., & Short, S. D.
- Affiliations
- East Carolina University (US)
- University of Delaware (US)
- College of Charleston (US)
- Corresponding author
- Schoemann, Alexander M.
- Venue
- Social Psychological and Personality Science
- Publisher
- SAGE Publications
- Published
- 2017-05
- Volume
- 8
- Issue
- 4
- Pages
- 379-386
- ISSN
- 1948-5506 (ISSN-L)
- 1948-5514
- OpenAlex
- W2625117584
- Cited by
- 1,837
- References
- 31
- Citations by year
- 2026: 384 · 2025: 374 · 2024: 318 · 2023: 220 · 2022: 221 · 2021: 156 · 2020: 97 · 2019: 40 · 2018: 25 · 2017: 1
- FWCI
- 31.00
- Type
- journal article
- Language
- en
- Open access
- closed
- Licence
- journals.sagepub.com/page/policies/text-and-data-mining-license (tdm)
- Locations
- Social Psychological and Personality Science (publishedVersion)
- Topics
- Physical Sciences › Mathematics › Statistics and Probability › Advanced Causal Inference Techniques
- Social Sciences › Decision Sciences › Statistics, Probability and Uncertainty › Meta-analysis and systematic reviews
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Bayesian Inference
- Keywords
- mediation analysis, statistical power, sample size determination, indirect effects, Monte Carlo simulation
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.