Sample Size Requirements for Structural Equation Models: An Evaluation of Power, Bias, and Solution Propriety

By E. Wolf, K. Harrington, Shaunna L. Clark, Mark W. Miller. Educational and Psychological Measurement, 2013

3,247 citations recorded by the source corpus.

Determining sample size requirements for structural equation modeling (SEM) is a challenge often faced by investigators, peer reviewers, and grant writers. Recent years have seen a large increase in SEMs in the behavioral science literature, but consideration of sample size requirements for applied SEMs often relies on outdated rules-of-thumb. This study used Monte Carlo data simulation techniques to evaluate sample size requirements for common applied SEMs. Across a series of simulations, we systematically varied key model properties, including number of indicators and factors, magnitude of factor loadings and path coefficients, and amount of missing data. We investigated how changes in these parameters affected sample size requirements with respect to statistical power, bias in the parameter estimates, and overall solution propriety. Results revealed a range of sample size requirements (i.e., from 30 to 460 cases), meaningful patterns of association between parameters and sample size, and highlight the limitations of commonly cited rules-of-thumb. The broad “lessons learned” for determining SEM sample size requirements are discussed.

Subfields: Computer Networks and Communications, Experimental and Cognitive Psychology, Management Science and Operations Research

View Paper: https://www.semanticscholar.org/paper/f4381a0e0f025d2959e990287e0502cb955763e6

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

In Experimental and Cognitive Psychology

In Management Science and Operations Research

Citations in the corpus

Emotional intelligence, emotional labor, coach burnout, job satisfaction, and turnover intention in sport leadership (2018, methodology): “Emotional intelligence Emotional intelligence was assessed by Wong and Law’s (2002) questionnaire which asks respondents to assess their level of emotional intelligence (Walter et al.”
Teaching Who You Are: Connecting Teachers’ Civic Education Ideology to Instructional Strategies (2018, methodology): “Instead, Wolf et al. (2013) recommended conducting proactive Monte Carlo analysis to determine bias and statistical F ig u re 1.”
Cognitive-Social Sources of Wellbeing: Differentiating the Roles of Coping Style, Social Support and Emotional Intelligence (2016, background): “Given uncertainty over minimum Ns for modeling (Wolf et al. 2013), replication of findings with a larger sample size would be desirable.”
Examining Cultural Intelligence and Cross-Cultural Negotiation Effectiveness (2015, background): “Furthermore, Wolf et al. (2013) assert that mediation models with smaller direct effects require larger sample sizes to achieve acceptable statistical power.”
Practitioner’s Guide to Latent Class Analysis: Methodological Considerations and Common Pitfalls. (2020, background): “Similarly, Wolf et al (40) in simulation studies found that models with 20% missing data required an approximate increase of 50% in their sample size to achieve sufficient power.”
Examining the effects of environmentally-specific servant leadership on green work outcomes among hotel employees: the mediating role of climate for green creativity (2021, positive): “This sample size also met the minimum required to run a SEM modeling for a statistical power of 80% at α = 0.05 (Wolf et al., 2013).”
Understanding the Blockchain technology adoption in supply chains-Indian context (2018, methodology): “Further, Sideridis et al. (2014) and Wolf et al. (2013) have recommended small samples as sufﬁcient for SEM analysis.”
An introduction to latent variable mixture modeling (part 1): overview and cross-sectional latent class and latent profile analyses. (2014, positive): “In recent years, the Monte Carlo simulation method has emerged as a promising approach for estimating sample size in the context of structural equation modeling in general and LVMM in particular (Muthén & Muthén, 1998–2012, 2002; Wolf et al., 2013).”
Response to Leslie Hayduk’s Review of Principles and Practice of Structural Equation Modeling, 4th Edition (2018, neutral): “Ignoring a significant model χ 2 is bad practice especially if the power of the test is low, which is typically true in most published SEM studies (Wolf et al. 2013).”