Polychoric Matrix

Verified

Psychometrics (legacy hub)

Independently verified. Every statistic this test reports has been re-derived against an independent reference — never the library the pipeline itself calls — the rendered output was read back in a browser, and the result is locked with a committed regression suite.

Run this test straight away on a free built-in teaching dataset — no data of your own needed — or bring your own. Either opens the guided workspace: variable setup, assumption diagnostics, results with effect sizes and confidence intervals, figures, and APA-ready reporting.

Loading teaching datasets…

Or use your own dataset

Loading your datasets…

Estimates correlations between ORDINAL items (Likert scales, binary items) under the assumption of an underlying continuous bivariate normal distribution.

Pearson correlations on ordinal items are ATTENUATED — the discrete nature of the response truncates true correlations toward zero. Polychoric (for ordinal × ordinal) and tetrachoric (for binary × binary) correlations recover the underlying continuous correlation. Reports the polychoric matrix, convergence diagnostics, and matrix positivity (positive-definite required for downstream EFA / CFA).

Worked example

How do we correlate ordinal (Likert) items properly?

Polychoric correlations estimate the latent continuous correlation behind ordinal items — more appropriate than Pearson for Likert data before factoring.

Result

Polychoric correlations among items ranged .35–.68, higher than the Pearson values, yielding a cleaner factor solution.

How you'd report it (APA)

Polychoric correlations were used for the ordinal items (range .35–.68) prior to factor analysis.

When to use it

  • EFA / CFA input prep for ordinal Likert items
    30-item depression scale with 4-point Likert responses.
  • Tetrachoric matrix for binary-item analysis
    60-item multiple-choice math test (0/1 scoring) on n=400.

When NOT to — use instead

  • Continuous (interval/ratio) items
    Pearson is appropriate for continuous data — polychoric is unnecessary and adds estimator noise. → KMO + Bartlett
  • Likert with 7+ response options
    5+ point Likert can be treated as approximately continuous; Pearson is acceptable. → KMO + Bartlett
  • Single item or pair (not a matrix)
    Polychoric MATRIX needs ≥ 3 items. → AVE + CR + HTMT
  • Sample too small (n < 200)
    Polychoric estimates are noisy on small n. → KMO + Bartlett

Assumptions (and what to do if they fail)

Observed responses arise from latent continuous variablesmedium

Check: See the assumption diagnostics in the workspace.

If violated: The workspace flags this and suggests a robust or nonparametric alternative.

Items use the same underlying scale (or an ordered Likert)medium

Check: See the assumption diagnostics in the workspace.

If violated: The workspace flags this and suggests a robust or nonparametric alternative.

Listwise exclusion of incomplete pairsmedium

Check: See the assumption diagnostics in the workspace.

If violated: The workspace flags this and suggests a robust or nonparametric alternative.

Ready to run a Polychoric Matrix on your own data?

Guided setup, automatic assumption checks, effect sizes, figures and an APA write-up.

Run this test →