Journal article · 2017 · Open access
Reliability Statistics
PM&R, 9(6), 622–628 · Published 1 June 2017 · DOI 10.1016/j.pmrj.2017.05.001
In brief
“Reliability Statistics” is a 2017 journal article by Kristin Lynn Sainani, published in PM&R. OpenAlex counts 84 citations as of 16 November 2026. It is open access: the full text is free to read at PM&R.
- Citations
- 84
- Most cited in
- 2024 (17)
- Field-weighted impact
- 1.50
- References
- 6
- Authors
- 1
Abstract
Measurement tools used in clinical studies need to be sufficiently reliable. Reliability means that the tool gives consistent results when administered by different people (interrater reliability) or at different time points (test-retest reliability). Statistical tests of association are not appropriate for assessing reliability—reliability statistics assess agreement rather than association. Reliability statistics include the Kappa statistic for categorical scales and the intraclass correlation coefficient (ICC) for continuous scales. Multiple versions of both the Kappa statistic and the ICC exist, and researchers need to know how to choose the right type for their problem. Additional statistics are needed to gauge how useful an instrument is for measuring changes over time. This article reviews statistics for reliability and also gives tips for designing reliability studies. The statistics discussed work the same whether applied to raters or time points. Categorical scales include binary variables such as whether an athlete has a concussion; ordered categories such as whether it's probable, uncertain, or unlikely that an athlete has a concussion; and unordered categories, such as race. Researchers commonly will report absolute agreement as a measure of interrater (or test-retest) reliability for categorical scales. For example, if 2 physicians evaluate 100 athletes and agree on the presence or absence of a concussion for 70 athletes, the absolute agreement is 70%. However, this statement ignores the fact that agreement can occur due to chance. If 2 physicians each just flipped a coin to decide, they would agree 50% of the time. The Kappa statistic corrects for this chance agreement. There are multiple versions of Kappa—choosing the right one depends on how many raters (or time points) are involved, whether the categories are ordered or unordered, and other factors. Kappa is always less than or equal to 1, where 1 implies perfect agreement and 0 implies no better than chance. Kappa can be negative if raters agree less often than expected by chance. Examples of how to calculate Cohen's Kappa. In (A), the raters do no better than chance. In (B), the raters beat chance. (C) and (D) illustrate how Kappa decreases as the sample becomes more homogenous. Shaded cells indicate rater agreement. Cohen's Kappa can underestimate agreement if the sample is too homogenous. If most subjects are concussed or most are not concussed, the expected agreement due to chance is high. For example, if raters rate 5% of subjects as concussed and 95% as not concussed, the expected agreement due to chance is a whopping 90.5% (see the Sidebar for more details on how to calculate the expected…
Cite this paper
Formatted from the catalogue record. Check names, capitals and page numbers against the paper before you submit.
APA 7
Sainani, K. L. (2017). Reliability Statistics. PM&R, 9(6), 622–628. https://doi.org/10.1016/j.pmrj.2017.05.001
MLA 9
Sainani, Kristin Lynn. “Reliability Statistics.” PM&R, vol. 9, no. 6, 2017, pp. 622–628. https://doi.org/10.1016/j.pmrj.2017.05.001.
Chicago
Sainani, Kristin Lynn. 2017. “Reliability Statistics.” PM&R 9 (6): 622–628. https://doi.org/10.1016/j.pmrj.2017.05.001.
Harvard
Sainani, K. L. (2017) ‘Reliability Statistics’, PM&R, 9(6), pp. 622–628. doi:10.1016/j.pmrj.2017.05.001.
IEEE
K. L. Sainani, “Reliability Statistics,” PM&R, vol. 9, no. 6, pp. 622–628, 2017, doi: 10.1016/j.pmrj.2017.05.001.
BibTeX
@article{sainani2017reliability,
title = {{Reliability Statistics}},
author = {Sainani, Kristin Lynn},
journal = {PM\&R},
year = {2017},
volume = {9},
number = {6},
pages = {622--628},
doi = {10.1016/j.pmrj.2017.05.001},
url = {https://doi.org/10.1016/j.pmrj.2017.05.001}
}
Download for a reference manager: BibTeX (.bib) · RIS for Zotero, Mendeley and EndNote (.ris)
Citations by year
| Year | Citations | Share of the best year |
|---|---|---|
| 2017 | 3 | |
| 2018 | 2 | |
| 2019 | 2 | |
| 2020 | 3 | |
| 2021 | 6 | |
| 2022 | 8 | |
| 2023 | 7 | |
| 2024 | 17 | |
| 2025 | 14 | |
| 2026 | 13 |
Its field-weighted citation impact is 1.50: it has been cited 1.50 times as often as the average work of the same field, type and year (1.00 is the world average).
Topic and keywords
Social Sciences › Decision Sciences › Statistics, Probability and Uncertainty › Reliability and Agreement in Measurement
- inter-rater reliability
- test-retest reliability
- reliability statistics
- Cohen's kappa
- intraclass correlation coefficient
- continuous scale
- chance-corrected agreement
- sample homogeneity
Work with this paper
Most-cited papers in Decision Sciences
- The PRISMA 2020 statement: an updated guideline for reporting systematic reviewsMatthew James Page, Joanne E. McKenzie, Patrick M. Bossuyt and 23 more · BMJ · 2021 · 144,919 citations
- Evaluating Structural Equation Models with Unobservable Variables and Measurement ErrorClaes Fornell, David F. Larcker · Journal of Marketing Research · 1981 · 118,336 citations
- Cutoff criteria for fit indexes in covariance structure analysis: Conventional criteria versus new alternativesLi‐tze Hu, Peter M. Bentler · Structural Equation Modeling A Multidisciplinary Journal · 1999 · 117,461 citations
- Common method biases in behavioral research: A critical review of the literature and recommended remedies.Philip M. Podsakoff, SCOTT B. MACKENZIE, Jeong Yeon Lee and 1 more · Journal of Applied Psychology · 2003 · 89,248 citations
- Preferred reporting items for systematic reviews and meta-analyses: the PRISMA statementDavid Moher, Alessandro Liberati, Jennifer Tetzlaff and 2 more · BMJ · 2009 · 85,524 citations
Questions about this paper
How do I cite this paper in APA style?
Sainani, K. L. (2017). Reliability Statistics. PM&R, 9(6), 622–628. https://doi.org/10.1016/j.pmrj.2017.05.001
How many times has this paper been cited?
84 times, according to OpenAlex on 16 November 2026. Google Scholar and Scopus count different sets of sources, so their numbers differ.
Is this paper free to read?
Yes. It is open access: read the free full text at PM&R.
Who wrote this paper?
Kristin Lynn Sainani.
What is the DOI of this paper?
10.1016/j.pmrj.2017.05.001. The permanent link is https://doi.org/10.1016/j.pmrj.2017.05.001.
Bibliographic data from OpenAlex (CC0), last checked 16 November 2026. ResearchGuru does not host this paper; the links go to the publisher or the repository that does.