SEM Sample Size Calculator
A-priori minimum sample size for structural equation modeling — computed three ways, with a justification paragraph you can paste straight into a methods chapter, references included.
Recommended minimum: —
Citation-ready justification
References (APA 7)
- Bentler, P. M., & Chou, C.-P. (1987). Practical issues in structural modeling. Sociological Methods & Research, 16(1), 78–117.
- Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). Lawrence Erlbaum Associates.
- Jackson, D. L. (2003). Revisiting sample size and number of parameter estimates: Some support for the N:q hypothesis. Structural Equation Modeling, 10(1), 128–141.
- Kline, R. B. (2016). Principles and practice of structural equation modeling (4th ed.). Guilford Press.
- Westland, J. C. (2010). Lower bounds on sample size in structural equation modeling. Electronic Commerce Research and Applications, 9(6), 476–487.
How the three estimates work
The power-based minimum applies Cohen's (1988) framework through the Fisher-z transformation: the sample required to detect your anticipated effect size at the chosen power and significance level. Westland's (2010) lower bound reflects model structure — the ratio of observed indicators to latent variables — because complex measurement models need more information regardless of effect size. The N:q heuristic ties sample size to the approximate number of freely estimated parameters, following Jackson (2003) and Kline (2016); this calculator estimates q from your model dimensions. The recommended minimum is the largest of the three — the defensible choice under reviewer scrutiny.
This tool provides established a-priori conventions for planning, not a substitute for study-specific power simulation (e.g., Monte Carlo) where model particulars demand it. Need the full measurement model reviewed? That is what we do — start a conversation.