A simple method for removing bias from a popular measure of standardized effect size: Adjusted partial eta squared
Abstract
Accurate estimates of population effect size are critical to empirical science, for both reporting experimental results and conducting a priori power analyses. Unfortunately, the current most-popular measure of standardized effect size, partial eta squared ([Formula: see text]), is known to have positive bias. Two less-biased alternatives, partial epsilon squared ([Formula: see text]) and partial omega squared ([Formula: see text]), have both existed for decades, but neither is often employed. Given that researchers appear reluctant to abandon [Formula: see text], this article provides a simple method for removing bias from this measure, to produce a value referred to as adjusted partial eta squared (adj [Formula: see text]). Some of the many benefits of adopting this measure are briefly discussed.
Integrity signals
Registry
Cites
Statistics
Claims the paper makes
Claims other papers make about this one
Each is checked against this paper’s own text: does it say what the citing paper says it does?
Effect sizes were interpreted using partial η2 thresholds: values less than .06 were considered small effects, values between .06 and .14 as moderate effects, and values greater than .14 as large effects.
“Effect sizes were reported using partial η 2 , where values less than .06 were considered small effects, values between .06 and .14 as moderate effects, and values greater than .14 as large effects (Mordkoff, 2019).”
From Analysis of the Efficacy of Mindfulness Meditation in a Mutual Aid Group for Bipolar Disorder (2024) · cites this paper as Mordkoff, 2019
Cites
Statistical Power Analysis for the Behavioral Sciences
Muller, K. E., & Cohen, J. (1989) · Technometrics · doi:10.2307/1270020 · Cited by 99,588 · see citing works
Why Most Published Research Findings Are False
Ioannidis, J. P. A. (2005) · PLoS Medicine · doi:10.1371/journal.pmed.0020124 · Cited by 10,505 · see citing works
Eta squared and partial eta squared as measures of effect size in educational research
Richardson, J. T. E. (2011) · Educational Research Review · doi:10.1016/j.edurev.2010.12.001 · Cited by 3,498 · see citing works
Designing Experiments and Analyzing Data: A Model Comparison Perspective
Maxwell, S., & Delaney, H. D. (1990) · Journal of the American Statistical Association · doi:10.2307/2289744 · Cited by 2,593 · see citing works
Measuring the Prevalence of Questionable Research Practices With Incentives for Truth Telling
John, L. K., Loewenstein, G., & Prelec, D. (2012) · Psychological Science · doi:10.1177/0956797611430953 · Cited by 2,239 · see citing works
Book Review: Design and Analysis: A Researcher's Handbook
Aaker, D. A., & Keppel, G. (1976) · Journal of Marketing Research · doi:10.2307/3150757 · Cited by 2,033 · see citing works
Recommended effect size statistics for repeated measures designs
Bakeman, R. A. (2005) · Behavior Research Methods · doi:10.3758/bf03192707 · Cited by 1,813 · see citing works
Olejnik, S., & Algina, J. (2003) · Psychological Methods · doi:10.1037/1082-989x.8.4.434 · Cited by 1,622 · see citing works
Eta-squared and partial eta-squared in fixed factor anova designs
Cohen, J. (1973) · Educational and Psychological Measurement · doi:10.1177/001316447303300111 · Cited by 1,333 · see citing works
Everitt, B. J. (2001) · Psychology Press eBooks · doi:10.4324/9781410605542 · Cited by 1,055 · see citing works
Eta Squared, Partial Eta Squared, and Misreporting of Effect Size in Communication Research
Levine, T. R., & Hullett, C. R. (2002) · Human Communication Research · doi:10.1111/j.1468-2958.2002.tb00828.x · Cited by 998 · see citing works
Steiger, J. H. (2004) · Psychological Methods · doi:10.1037/1082-989x.9.2.164 · Cited by 567 · see citing works
Cumming, G., & Finch, S. (2001) · Educational and Psychological Measurement · doi:10.1177/0013164401614002 · Cited by 528 · see citing works
Grissom, R. J., & Kim, J. J. (2012) · doi:10.4324/9780203803233 · Cited by 470 · cited 1× · see citing works
“…2013) and at least one of these alternatives has been strongly recommended in most detailed, statistical discussions (e.g., Carroll & Nordholm, 1975; Grissom & Kim, 2012; Hays, 1963; Keppel, 1991; Maxwell & Delaney, 2004; see also Bakeman, 2005; Olejnik & Algina, 2003), neither of these…”
BackgroundAlbers, C. J., & Lakens, D. (2017) · Journal of Experimental Social Psychology · doi:10.1016/j.jesp.2017.09.004 · Cited by 366 · see citing works
Smithson, M. J. (2001) · Educational and Psychological Measurement · doi:10.1177/00131640121971392 · Cited by 275 · see citing works
An Unbiased Correlation Ratio Measure
Kelley, T. L. (1935) · Proceedings of the National Academy of Sciences · doi:10.1073/pnas.21.9.554 · Cited by 152 · see citing works
Is Omega Squared Less Biased? a Comparison of Three Major Effect Size Indices in One-Way Anova
Okada, K. (2013) · Behaviormetrika · doi:10.2333/bhmk.40.129 · Cited by 128 · see citing works
Measures of strength of association: A comparative examination.
Maxwell, S. E., Camp, C. J., & Arvey, R. D. (1981) · Journal of Applied Psychology · doi:10.1037/0021-9010.66.5.525 · Cited by 125 · see citing works
Sampling Characteristics of Kelley's ε and Hays' ω
Carroll, R. M., & Nordholm, L. A. (1975) · Educational and Psychological Measurement · doi:10.1177/001316447503500304 · Cited by 83 · see citing works
Keselman, H. J. (1975) · Canadian Psychological Review/Psychologie canadienne · doi:10.1037/h0081789 · Cited by 46 · see citing works
Okada, K. (2016) · Behavior Research Methods · doi:10.3758/s13428-016-0760-y · Cited by 34 · see citing works
Flagged first, then by citations
Cited by
Sorted by citations
Dominguez-Rodriguez, A., Herdoiza-Arroyo, P. E., González-Ramírez, L. P., Martínez-Arriaga, R. J., Villarreal-Zegarra, D., Santos da Silva, A. C., González-Cantero, J. O., Vargas Salinas, V. S., S. Mensorio, M., Cisneros Hernández, A. A., et al. (2024) · Frontiers in Psychology · doi:10.3389/fpsyg.2024.1279847 · Cited by 3 · see citing works
“The adjusted partial eta squared (adj η 2) will be used to evaluate the effect size, knowing that the values 0.02 indicate a small effect size, 0.15 medium, and 0.30 large (Mordkoff, 2019).”
MethodInfluentialAnalysis of the Efficacy of Mindfulness Meditation in a Mutual Aid Group for Bipolar Disorder
Burgos-Julián, F. A., Díaz-Silveira, C., Ruiz-Íñiguez, R., & Santed-Germán, M. A. (2024) · Clinica y Salud · doi:10.5093/clysa2024a4
- Authors
- Mordkoff, J. T.
- Affiliations
- University of Iowa (US)
- Corresponding author
- Mordkoff, J. Toby
- Venue
- Advances in Methods and Practices in Psychological Science
- Publisher
- SAGE Publications
- Published
- 2019-07-30
- Volume
- 2
- Issue
- 3
- Pages
- 228-232
- ISSN
- 2515-2459 (ISSN-L)
- 2515-2467
- OpenAlex
- W2965249208
- Cited by
- 122
- References
- 22
- Citations by year
- 2026: 9 · 2025: 18 · 2024: 17 · 2023: 22 · 2022: 20 · 2021: 14 · 2020: 17 · 2019: 4
- FWCI
- 9.39
- Type
- journal article
- Language
- en
- Open access
- closed
- Locations
- Advances in Methods and Practices in Psychological Science (publishedVersion)
- Topics
- Social Sciences › Decision Sciences › Statistics, Probability and Uncertainty › Meta-analysis and systematic reviews
- Social Sciences › Psychology › Experimental and Cognitive Psychology › Psychology Research and Bibliometrics
- Physical Sciences › Mathematics › Statistics and Probability › Statistical Methods and Bayesian Inference
- Keywords
- partial eta squared, bias correction, standardized effect size, a priori power analysis
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
