Is EFA Required Before PLS-SEM?
EFA is not required before every PLS-SEM analysis. The decision depends on theoretical maturity, scale quality, and whether the objective is exploration or confirmation.
Use EFA when structure is uncertain or scales are new or extensively translated; avoid using the same sample for exploration and claims of confirmation without cross-validation.
EFA: decisions need several layers of evidence
- Data
Check items and the correlation matrix
- Factor retention
Parallel analysis, scree and theory
- Rotation
Choose for the factor relationships
- Interpret loadings
Review cross-loadings and item meaning
Conceptual diagram. Do not delete items solely to achieve a loading cutoff or better index.
Source: Lê Minh · Conceptual illustration, not software output.1. Why Is There No Mandatory Rule?
PLS-SEM has its own measurement model for assessing loadings, reliability, and validity. A well-established reflective scale used in an appropriate context does not necessarily require another EFA (Fabrigar et al., 1999).
Conversely, new scales, substantive revisions, extensive translation, or suspected cross-loadings may require structural exploration before model specification.
2. A Practical Decision Tree
| Situation | Approach |
|---|---|
| New scale or uncertain dimensionality | EFA in an exploratory sample followed by validation in another sample |
| Established scale in a similar context | May proceed directly to a theoretically specified measurement model |
| Translation or revision of many items | Piloting, EFA/ESEM, or confirmation with cross-validation |
| Composite/formative | Do not impose reflective EFA; assess content, weights, and collinearity |
| Only one available sample | State limitations and avoid strong simultaneous exploration and confirmation claims |
3. PLS-SEM Is Not Automatically CFA
Estimating loadings in PLS-SEM does not make every analysis confirmatory. Confirmation comes from a prespecified measurement model and assessment of its agreement with the data.
When the objective is testing a common-factor model, consider CFA/CCA or appropriate confirmatory methods rather than relying only on reliability–validity criteria (Henseler & Schuberth, 2020).
Item removal requires considering loadings, content, conceptual coverage, and effects on comparability with prior research.
4. Reporting Template
Example report
The study [conducted/did not conduct] EFA before PLS-SEM because of [scale maturity/context/revisions]. The measurement model was specified based on [source] and assessed using [criteria/methods]. If the same sample was used, this was acknowledged as a limitation of confirmation.
- State whether the objective is exploration or confirmation.
- Distinguish reflective from composite/formative measurement.
- Do not treat EFA as a mandatory ritual.
- Prefer an independent sample or validation split.
- Protect content validity when removing items.
- Describe every modification made after examining the data.
Conclusion
The right question concerns which evidence is needed to defend the scale structure in its specific context, rather than simply whether EFA should be run.
References
- Henseler, J., & Schuberth, F. (2020). Using confirmatory composite analysis to assess emergent variables in business research. Journal of Business Research, 120, 147–156.https://doi.org/10.1016/j.jbusres.2020.07.026
- Fabrigar, L. R., Wegener, D. T., MacCallum, R. C., & Strahan, E. J. (1999). Evaluating the use of exploratory factor analysis in psychological research. Psychological Methods, 4(3), 272–299.https://doi.org/10.1037/1082-989X.4.3.272
Academic note: EFA and PLS-SEM address different questions; choices must reflect construct type and research objectives.