How to Write the EFA Section

A comprehensive guide to reporting Exploratory Factor Analysis in your academic papers.

Why We Report EFA

Exploratory Factor Analysis (EFA) helps researchers check whether survey items group together into meaningful factors (constructs). When writing, you need to show that your data is suitable, explain how many factors were extracted, and confirm that items loaded well.

1

Sampling Adequacy

Before doing EFA, prove your dataset is suitable using these two tests:

Kaiser-Meyer-Olkin (KMO)

> .90 Excellent
.80 – .89 Good
.70 – .79 Acceptable
< .60 Not Suitable

Bartlett’s Test of Sphericity

Tests if correlation matrix ≠ identity matrix.

Significant result (p < .05) = PASS

2

Factor Extraction

Using methods like Principal Component Analysis (PCA).

Eigenvalue > 1 rule Only keep factors that explain meaningful variance (value greater than 1).
Scree Plot Visual check: keep factors before the "elbow" drop in the graph.
3

Rotation

Rotation simplifies the structure for interpretation.

Varimax
(Orthogonal)
Assumes factors are
NOT correlated
Oblimin
(Oblique)
Allows correlation
among factors
4

Factor Loadings

The strength of the relationship between an item and its factor.

  • Loadings ≥ .40 are generally acceptable.
  • Loadings > .60 are strong evidence of alignment.
  • Avoid Cross-loadings: One item loading strongly on two different factors.

Model Example (Academic Style)

report_final.docx

Exploratory Factor Analysis (EFA). To examine construct validity, EFA was conducted in SPSS using Principal Component Analysis with Varimax rotation. The Kaiser-Meyer-Olkin (KMO) value was 0.884 and Bartlett’s Test of Sphericity was significant (χ² = 4147.503, p < .001), confirming sampling adequacy. EFA supported a three-factor structure aligned with the theoretical model, with items loading cleanly on their respective constructs.

Copy-Ready Templates

Choose the template that matches your results.

EFA – Good Fit

Exploratory Factor Analysis (EFA). To examine construct validity, EFA was conducted in SPSS using [extraction method] with [rotation method]. The Kaiser-Meyer-Olkin (KMO) value was [x.xxx] and Bartlett’s Test of Sphericity was significant (χ² = [xxxx.xx], p < .001), confirming sampling adequacy. EFA supported a [k]-factor structure aligned with the theoretical model, with items loading cleanly on their respective constructs.

Copied!

EFA – Acceptable Fit

Exploratory Factor Analysis (EFA). EFA was conducted in SPSS using [extraction method] with [rotation method]. The KMO value was [x.xxx] and Bartlett’s Test of Sphericity was significant (χ² = [xxxx.xx], p < .001), confirming sampling adequacy. A [k]-factor solution was extracted, though some items showed [cross-loadings/moderate loadings]. Overall, the results provided reasonable evidence of construct validity, but further refinement may be necessary.

Copied!

EFA – Poor Fit

Exploratory Factor Analysis (EFA). EFA was conducted in SPSS using [extraction method] with [rotation method]. The KMO value was [x.xxx] and Bartlett’s Test of Sphericity was [non-significant/weak], suggesting limited sampling adequacy. Factor extraction did not reveal a clear structure, with many items showing [low loadings or cross-loadings]. These results indicate that the measurement model lacks construct validity and requires substantial revision before further analysis.

Copied!