---
title: "Test-retest reliability (temporal stability)"
method_id: SM.RLB.STB.TIME.TESTRETEST
family: Reliability
version: 1.1.0
date_modified: 2026-06-18
canonical: https://clarus.ofrencber.com/library/test_retest_reliability.md
source: Clarus method library
---

# Test-retest reliability (temporal stability)

> A classical statistical method in the Clarus library (Reliability family).
> Clarus selects, assumption-checks and runs this method on your data with a
> deterministic rule engine and real numerical libraries (scipy / statsmodels);
> it never generates numbers, citations or results.

**Also known as:** temporal stability; test-retest correlation; stability coefficient; retest reliability; coefficient of stability.

## Hypotheses

- **Null (H0):** not_applicable
- **Alternative (H1):** not_applicable

## When to use it

Use this method when your goal is to:
- assess stability of a measure over time
- quantify test retest reliability
- report temporal consistency of scores on two occasions

## Data it expects

- **dependent:** 1 numeric measurement recorded TWICE on the same cases (same instrument, two occasions)
- **independent:** not_applicable (the two columns are time points, not groups)
- **pairing:** fully paired — every case has a time-1 and a time-2 score (wide subject x occasion matrix)

## Assumptions Clarus checks

- **construct stability over interval** (severity: warning; on violation: warn_only)
- **no carryover practice or memory effect** (severity: warning; on violation: warn_only)
- **bivariate linearity and no outliers** (severity: warning; on violation: warn_only)

## Effect size reported

- **correlation r** (Pearson (1896); used as the classical stability coefficient)

## Honest limitations

- Always report the retest interval — a stability coefficient only means something relative to the time gap it covers.
- Pearson r measures rank/relative consistency and is blind to a systematic shift between occasions; report an absolute-agreement ICC(2,1) when practice, learning, or regression effects are plausible.
- Real change in the construct between occasions is confounded with unreliability; a low coefficient may reflect genuine change, not a bad instrument.
- A small sample yields a wide confidence interval; a single discrepant pair can dominate the coefficient.
- High test-retest reliability indicates stability, NOT validity — a measure can be reliably stable yet measure the wrong construct.
- Results are sample- and interval-specific and generalize only to similar populations, instruments, and time gaps.

## Primary sources

- Shrout, P. E., & Fleiss, J. L. (1979). Intraclass correlations: Uses in assessing rater reliability. Psychological Bulletin, 86(2), 420-428. https://doi.org/10.1037/0033-2909.86.2.420
- Koo, T. K., & Li, M. Y. (2016). A guideline of selecting and reporting intraclass correlation coefficients for reliability research. Journal of Chiropractic Medicine, 15(2), 155-163. https://doi.org/10.1016/j.jcm.2016.02.012
- Fisher, R. A. (1921). On the probable error of a coefficient of correlation deduced from a small sample. Metron, 1, 3-32.
- Nunnally, J. C., & Bernstein, I. H. (1994). Psychometric Theory (3rd ed.). McGraw-Hill, Ch.7 — temporal stability / coefficient of stability.
- Field, A. (2013). Discovering Statistics Using IBM SPSS Statistics (4th ed.), Ch.7 (correlation) & Ch.17 (reliability).

## How to cite this

To cite this Clarus method page (the page itself — for the method's own primary sources, see above):

**Plain text**

Clarus, "Test-retest reliability (temporal stability)", version 1.1.0, Clarus method library, 2026. https://clarus.ofrencber.com/library/test_retest_reliability.md

**APA**

Clarus. (2026). Test-retest reliability (temporal stability) (Version 1.1.0) [Statistical method, Clarus method library]. Retrieved from https://clarus.ofrencber.com/library/test_retest_reliability.md

**BibTeX**

```bibtex
@misc{clarus-test-retest-reliability,
  author       = {Clarus},
  title        = {Test-retest reliability (temporal stability)},
  howpublished = {Clarus method library},
  version      = {1.1.0},
  year         = {2026},
  url          = {https://clarus.ofrencber.com/library/test_retest_reliability.md}
}
```

_No DOI is minted for Clarus method pages yet; this citation uses the canonical URL, version and date. Cite the primary sources above for the method's scientific provenance._

---

METHOD SM.RLB.STB.TIME.TESTRETEST · VERSION 1.1.0 · UPDATED 2026-06-18 · SOURCE Clarus method library