Aiken's V

Verified

Reliability (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…

Content-validity agreement — Aiken's V summarizes how strongly an expert panel endorses each item's relevance.

Aiken's V rescales expert ratings to 0–1 with a significance test; higher V means stronger consensus that an item measures the intended content.

Worked example

Do experts agree the items are relevant?

Eight experts rated each item's relevance on a 1–5 scale; Aiken's V quantifies content-validity agreement per item.

Result

12 of 15 items reached Aiken's V ≥ .80, supporting content validity (the three weakest items fell below and would be revised).

How you'd report it (APA)

Content validity was supported, with 12 of 15 items at Aiken's V ≥ .80.

Try it yourself: Load this ready-made sample and follow the run above.

When to use it

  • Quantifying expert content-validity ratings
    A panel rates each item's relevance on an ordered scale (e.g. 1–5) and you want a per-item index with a significance test. e.g. 8 experts rating 15 items.
  • You need item-level decisions
    V is computed per item, so it tells you which specific items to keep, revise, or drop.

When NOT to — use instead

  • Experts rate only relevant / not relevant
    A binary relevance judgement is the CVI's territory, not a rescaled rating. → Content Validity Index
  • You want the 'essential' proportion
    Lawshe's essential/useful/not-necessary format gives the CVR. → Content Validity Ratio

Assumptions (and what to do if they fail)

Ratings are on a genuine ordered scalehigh

Check: The rating categories (1–5, say) are equally spaced and the number of scale points is stated — V's formula depends on the scale minimum and range.

If violated: A wrong scale range rescales V incorrectly; a value can even exceed its own 0–1 bound (a tell that the scale was mis-set).

Enough expertsmedium

Check: Content-validity panels are typically 5–10 experts; too few makes V and its significance unstable.

If violated: A single expert swings the index; the significance test has almost no power.

Experts rate independentlymedium

Check: Ratings are made without consultation between experts.

If violated: Correlated ratings overstate consensus — the agreement is partly shared opinion, not independent endorsement.

Ready to run a Aiken's V on your own data?

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

Run this test →