The DOI matches its Crossref record; 2 of the 2 references it cites were found in a registry or an index.
Confidence intervals are a more useful complement to nonsignificant tests than are power calculations
Abstract
Many leading journals, including Behavioral Ecology, emphasize the importance of considering the power of statistical tests in the light of nonsignificant results. However, there is considerable scope for misinterpretation of what this advice actually implies. The common conception among biologists is that a nonsignificant result with low power is not to be relied on, but a nonsignificant result with high power is strong support for the null hypothesis. Here we will draw on a recent paper by Hoenig and Heisey (2001) to explain why use of (post hoc) power analysis actually provides no more information than does the p value itself. Imagine we are interested in knowing whether the type of bean on which a beetle larva develops affects the size of the beetle as an adult. We raise 10 beetle larvae on black-eyed beans and 10 on mung beans, and we measure their size as adults. The mean size of beetle raised on black-eyed beans is 5.32 mg (± 0.33 SD); on mung beans, 4.95 mg (± 0.56 SD). A t test to compare the mean size of the two groups of beetles yields a nonsignificant p value ( \(\mathit{p}\ {=}\ 0.09\) ); what can we conclude?
Overview
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Run the check — free, no account needed. It reads the open-access text and checks the sources it cites.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Schuirmann, D. J. (1987) · Journal of Pharmacokinetics and Biopharmaceutics · doi:10.1007/bf01068419 · Cited by 2,696
The Abuse of Power: The Pervasive Fallacy of Power Calculations for Data Analysis
Hoenig, J. M., & Heisey, D. M. (2001) · The American Statistician · doi:10.1198/000313001300339897 · Cited by 2,046
Cited by
Ellis, P. D. (2010) · doi:10.1017/cbo9780511761676 · Cited by 2,461
A farewell to Bonferroni: the problems of low statistical power and publication bias
Nakagawa, S. (2004) · Behavioral Ecology · doi:10.1093/beheco/arh107 · Cited by 2,097
Time for some a priori thinking about post hoc testing
Ruxton, G. D., & Beauchamp, G. (2008) · Behavioral Ecology · doi:10.1093/beheco/arn020 · Cited by 599
Winfree, R., Williams, N. M., Gaines, H., Ascher, J. S., & Kremen, C. (2008) · Journal of Applied Ecology · doi:10.1111/j.1365-2664.2007.01418.x · Cited by 515
Sun, S., Pan, W., & Wang, L. L. (2010) · Journal of Educational Psychology · doi:10.1037/a0019507 · Cited by 342
Metadata
- Authors
- Colegrave, N.
- Venue
- Behavioral Ecology
- Publisher
- Oxford University Press (OUP)
- Published
- 2003-05-01
- Volume
- 14
- Issue
- 3
- Pages
- 446-447
- ISSN
- 1045-2249 (ISSN-L)
- 1465-7279
- OpenAlex
- W2151278120
- Cited by
- 250
- References
- 2
- FWCI
- 15.65
- Type
- journal article
- Language
- en
- Open access
- closed
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
How to cite
Cite CoChat for the checks on this page, and the paper for its own findings.
- This page
- CoChat. (2026). Citation check: Confidence intervals are a more useful complement to nonsignificant tests than are power calculations. https://app.cochat.ai/papers/10.1093/beheco/14.3.446
- The paper
- Colegrave, N. (2003). Confidence intervals are a more useful complement to nonsignificant tests than are power calculations. Behavioral Ecology. https://doi.org/10.1093/beheco/14.3.446
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.