• Aucun résultat trouvé

Corrigendum to “A review of effect sizes and their confidence intervals, Part I: The Cohen’s d family”

N/A
N/A
Protected

Academic year: 2022

Partager "Corrigendum to “A review of effect sizes and their confidence intervals, Part I: The Cohen’s d family”"

Copied!
2
0
0

Texte intégral

(1)

¦ 2019 Vol. 15 no. 1

Corrigendum to “A review of effect sizes and their confidence intervals, Part I: The Cohen’s d family”

Jean-Christophe Goulet-Pelletier

a,

B and Denis Cousineau

a

aUniversit´e d’Ottawa

Abstract Two errors have been found in the published article that are corrected here. All the simulations reported were error-free.

Keywords Effect size, standard error, confidence intervals, Cohen’sd.

B jgoul014@uottawa.ca

JCGP

: 0000-0002-2016-549X;

DC

: 0000-0001-5908-0402 10.20982/tqmp.15.1.p054

Acting Editor Roland Pfister (Uni- versit¨at W¨urzburg)

Two errors have been identified in the published article Goulet-Pelletier and Cousineau (2018). First, thedDc for- mula in Table 1, page 244, is erroneous. The correct for- mula is

dDc =dD×p

2 (1−r)

(the division sign is replaced by a multiplication sign). This error is found within the text, on page 252, right column, first paragraph, with the incorrect sentence

the result ofdDisp

2 (1−r)times larger than Cohen’sdp

(the symboldDanddpreplacesSDandSp). The mistake comes from the fact that the factor of conversion is incor- rectly applied to the effect size estimate instead of its stan- dard deviation. Therefore, when applied to the Cohen’sd effect size, the factor of conversion is1/p

2 (1−r). Hence, the equation (12a), at page 252, should be

dD=dp/p

2 (1−r) (12a)

and equation (12b) on page 252 should be dp=dD×p

2 (1−r) (12b)

(the multiplication sign is replaced by a division sign and vice versa). Note that the simulations have been conducted using the correct formulas (12a) and (12b). Therefore, the results are not affected by this change. The mistake was introduced at the writing phase of the manuscript.

Second, regarding the standard error of Cohen’sd1on page 256, left column, last paragraph, we omitted to men- tion that the standard error ofd1is obtained by changing all “2/n” in Table 3 by “1/n”. Again, the simulations have been conducted using the correct standard error formula (using1/n) ford1. Therefore, the results are not affected by this change.

Authors’ note

We are thankful to Louis Laurencelle and Bertram Walter who brought those mistakes to our attention.

References

Goulet-Pelletier, J.-C., & Cousineau, D. (2018). A review of effect sizes and their confidence intervals, Part I: The Cohen’s d family.The Quantitative Methods for Psy- chology,14(4), 242–265. doi:10.20982/tqmp.14.4.p242

Citation

Goulet-Pelletier, J.-C., & Cousineau, D. (2019). Corrigendum to “A review of effect sizes and their confidence intervals, part I: The Cohen’sdfamily”.The Quantitative Methods for Psychology,15(1), 54–54. doi:10.20982/tqmp.15.1.p054

Copyright © 2019,Goulet-Pelletier and Cousineau.This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) or licensor are credited and that the original

TheQuantitativeMethods forPsychology

2

54

(2)

¦ 2019 Vol. 15 no. 1

publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.

Received: 17/03/2019∼Accepted: 17/03/2019

TheQuantitativeMethods forPsychology

2

55

Références

Documents relatifs

Our research showed that an estimated quarter of a million people in Glasgow (84 000 homes) and a very conservative estimate of 10 million people nation- wide, shared our

It would be most interesting and informative to see carcase traits and body composition of different breeds and crosses compared at optimal slaughter weight. For milk

Once this constraint is lifted by the deployment of IPv6, and in the absence of a scalable routing strategy, the rapid DFZ RIB size growth problem today can potentially

While the IPv6 protocols are well-known for years, not every host uses IPv6 (at least in March 2009), and most network users are not aware of what IPv6 is or are even afraid

(In the UCSB-OLS experiment, the user had to manually connect to Mathlab, login, use MACSYMA, type the input in a form suitable for MACSYMA, save the results in a file

Potential action items for the W3C included investigating the formation of a privacy interest group and formulating guidance about fingerprinting, referrer headers,

The first was that the Internet Architecture Board should review and consider that document in the context of evaluating Birds of a Feather (BoF) session proposals at the

The objective is the development of a predictive model represented by an equation for the power conversion efficiency of dye sensitized solar cells, this model