Effect size, confidence interval and statistical significance: a practical guide for biologists

By Shinichi Nakagawa, Innes C. Cuthill. Biological Reviews of The Cambridge Philosophical Society, 2007

3,619 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Statistics and Probability

View Paper: https://www.semanticscholar.org/paper/f2ba29f3119cdedc147ac703e8d34bd8eed3d00b

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

In Statistics and Probability

Citations in the corpus

Growth and yield stimulation under elevated CO2 and drought : A meta-analysis on crops (2016, methodology): “A meta-analysis is a powerful tool to quantitatively evaluate the overall pattern across many independent studies because it combines the results into one measure: the effect size (Glass, 1976; Nakagawa and Cuthill, 2007).”
Identifying biologically relevant differences between metagenomic communities (2010, background): “CIs are more informative than P-values, and many proponents of CIs have suggested they make P-values unnecessary (Nakagawa and Cuthill, 2007).”
A primer on investigating the after effects of acute bouts of physical activity on cognition (2019, background): “1 1s 1 2 1 2 Whereas the confidence interval for the repeated measures Cohen’s d comparison (Kelley, 2007; Nakagawa & Cuthill, 2007) is calculated as: ⎜ ⎟ ⎛ ⎝ − ≤ ≤ − ⎞ ⎠ ncp Lower Limit r n d ncp Upper Limit r n $ .”
Repeatability for Gaussian and non‐Gaussian data: a practical guide for biologists (2010, background): “Notably, 95% confidence intervals can function as an indicator of statistical significance as well as an indicator of uncertainty and the presentation of p values should not replace the more informative confidence intervals (Nakagawa & Cuthill, 2007).”
Guidelines for estimating repeatability (2012, background): “In some contexts, this may be sufficient, but to convey the precision of the estimate, it is more informative and appropriate to report the magnitude of the repeatability with an associated confidence interval (Nakagawa & Cuthill 2007).”
A practical guide for using statistical tests to assess randomized algorithms in software engineering (2011, methodology): “In contrast to other guidelines in the literature for other scientiﬁc ﬁelds (e.g., [47] and [30]), the guidelines in this paper are tailored to the speciﬁc properties of randomized algorithms when applied to software engineering problems.”
How to measure and test phylogenetic signal (2012, background): “Many studies not only aim at attesting the presence of phylogenetic signal (i.e. a significant test result) but also at estimating the strength of phylogenic signal (i.e. the effect size, Nakagawa & Cuthill 2007).”
Neglected biological patterns in the residuals (2011, background): “Of course, this also highlights why using p values as a basis for statistical inference is questionable at best (Cohen 1990; Nakagawa and Cuthill 2007).”
A Comprehensive Review of Effect Size Reporting and Interpreting Practices in Academic Journals in Education and Psychology. (2010, background): “Nakagawa and Cuthill (2007): How to calculate effect size and construct confidence intervals on the basis of statistical models is addressed; how to deal with bias of effect size and the violation of assumptions are explicitly discussed.”
Building characteristics, indoor environmental quality, and mathematics achievement in Finnish elementary schools (2016, methodology): “The above methods have been used in sever al environmental health studies [13, 14, 32].”
Model selection and model averaging in behavioural ecology: the utility of the IT-AIC framework (2010, methodology): “For ecology in general, the case for caution in the use of NHST and the potential of IT as an alternative has been made elsewhere (Cohen 1994; Johnson 1999; Anderson et al. 2000; Johnson and Omland 2004; Nakagawa and Cuthill 2007), in some cases, quite forcibly.”
Rural-Urban Differences in Escape Behavior of European Birds across a Latitudinal Gradient (2017, methodology): “We assessed the importance of each predictor based on their effect sizes calculated as partial correlation coefficients (Nakagawa and Cuthill, 2007).”
Violating the normality assumption may be the lesser of two evils (2018, positive): “…coefficient) already work by violating the normality assumption (e.g. fitting a Pearson correlation through the binary data of a 2×2 table; Nakagawa & Cuthill 2007), so why not always report the Gaussian model to begin with, if the primary purpose of the test is to obtain a reliable…”
Computation of measures of effect size for neuroscience data sets (2011, methodology): “The MES toolbox Nakagawa & Cuthill (2007) argue that three types of effect statistics sufﬁce for most situations: r statistics (correlation coefﬁcients such Pearson’s product-moment coefﬁcient, Spearman’s rank correlation, point-biserial and phi), d statistics (Cohen’s d or Hedges’ g ) and the odds…”