Factor rotation (Varimax vs. Oblimin)

Description

Choosing between orthogonal (Varimax) and oblique (Oblimin, Promax) rotation so that the factor structure of UX constructs reflects whether they are correlated.

Rotation redistributes variance among retained factors to approach simple structure, where each item loads strongly on one factor. Varimax keeps factors uncorrelated; Oblimin and Promax allow them to correlate. Most UX constructs (usefulness, ease, trust, satisfaction) are correlated, so forcing orthogonality distorts loadings and inflates cross-loadings. Methodologists recommend starting with an oblique rotation: if factor correlations are all below about 0.32 (roughly 10% shared variance; Tabachnick and Fidell), an orthogonal solution is defensible. With oblique rotation, the pattern matrix (unique contributions) is interpreted for item assignment and the structure matrix (correlations) as a check. Rigour comes from justifying the choice and reporting the factor correlation matrix.

Attributes

Loss type
Siloed
Stage
Definition
Effort
Hours

Steps

  1. Fit the EFA with an oblique rotation (Oblimin with delta 0, or Promax with kappa 4) as the default.
  2. Inspect the factor correlation matrix.
  3. If all factor correlations are below about 0.32, consider re-running with Varimax for simplicity and compare.
  4. With oblique solutions, assign items using the pattern matrix and check the structure matrix for consistency.
  5. Report the rotation method, its parameters and the factor correlation matrix.

Sources

  • SPSS FACTOR
  • RELIABILITY procedures; syntax automation; parallel analysis via Python plugin
  • Kaiser, H. F. (1958) The varimax criterion for analytic rotation in factor analysis. Psychometrika
  • Tabachnick, B. G. & Fidell, L. S. (2019) Using Multivariate Statistics, 7th ed.
  • Field, A. (2024) Discovering Statistics Using IBM SPSS Statistics, 6th ed.
  • The Inquiry Series, C7 — SPSS for Survey Analysis — EFA, Reliability & Scale Validation
  • Discovering Statistics Using R/SPSS (Andy Field)
  • Quantitative User Experience Research (Chris Chapman & Kerry Rodden)
  • Movie Booking Playbook (Vivek Gopinath)
  • Thurstone, Multiple-Factor Analysis (1947)
  • Discovering Statistics (Field)
  • Quantitative UX Research (Chapman)
  • Movie Booking Playbook (Gopinath)
  • The Inquiry Series, Ch 50 — Factor Analysis — Discovering the Latent Structure of Experience

Also known as

  • Orthogonal vs. oblique rotation
  • Varimax rotation
  • Oblimin rotation
  • Promax rotation
  • Factor rotation (oblique/orthogonal)

Categories

  • Analytics & Behaviour
  • Survey & Psychometrics