Reliability & Convergent Validity Calculator
Calculate composite reliability, AVE and Cronbach's alpha in your browser. Results are descriptive evidence under stated assumptions—not a universal pass/fail decision about an instrument.
Reliability and validity calculators
1 · Composite Reliability & AVE from loadings
Paste positive, standardised factor loadings for one construct, one per line (e.g. 0.78). Reverse-code or align indicators before calculating; this simple form does not estimate a measurement model.
2 · Cronbach's alpha from raw data
Paste raw responses: one respondent per line, items separated by commas, tabs or spaces (same item count per line).
Method, reference markers and limits
Method version 1.1 · reviewed 20 July 2026. The displayed .70 CR/alpha and .50 AVE values are commonly used reference markers, not universal decision rules. Interpret them with the construct, item count, purpose, sample, uncertainty and model fit. Cronbach's alpha alone does not establish reliability; see McNeish (2018), Thanks Coefficient Alpha, We'll Take It From Here ↗.
Not calculated: confidence intervals, omega, ordinal alpha, item-total diagnostics, discriminant validity, HTMT, model fit or invariance. Raw responses remain in your browser. Declare missing-data handling and reverse-coded items before interpreting the result. Full measurement-model assessment requires more evidence — that is what we do.
How to read this tool
Purpose: Calculate composite reliability, average variance extracted and Cronbach's alpha with explicit data checks.
- Maturity
- Production
- Method version
- 1.1
- Reviewed
- Processing
- Local browser only
Evidence basis
Privacy
Raw responses and loadings remain in the user's browser.
Limitations
Displayed thresholds are common reference markers, not universal pass/fail rules; alpha alone does not establish unidimensionality or reliability.
Decision boundary
The output does not establish that an instrument is valid or suitable for a particular study.
Use the result. Check the reasoning.
This interface is a decision aid—not an automatic verdict. The 1 linked Library entry explain the source workflow, assumptions and boundary conditions that should travel with the result.