Home > Research > Publications & Outputs > Improving Inferences about Null Effects with Ba...

Electronic data

  • 2018.05.18_Lakens_McLatchie_Isager_Scheel_Dienes_Improving_Inferences_About_Null_Effects_With_Bayes_Factors_And_Equivalence_Tests_R1

    Rights statement: his is a pre-copy-editing, author-produced PDF of an article accepted for publication in The Journals of Gerontology: Series B following peer review. The definitive publisher-authenticated version Daniël Lakens, Neil McLatchie, Peder M Isager, Anne M Scheel, Zoltan Dienes, Improving Inferences About Null Effects With Bayes Factors and Equivalence Tests, The Journals of Gerontology: Series B 2020 75: 45-57 is available online at: https://academic.oup.com/psychsocgerontology/article/75/1/45/5033832

    Accepted author manuscript, 575 KB, PDF document

    Available under license: CC BY-NC: Creative Commons Attribution-NonCommercial 4.0 International License

Links

Text available via DOI:

View graph of relations

Improving Inferences about Null Effects with Bayes Factors and Equivalence Tests

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Published

Standard

Improving Inferences about Null Effects with Bayes Factors and Equivalence Tests. / Lakens, Daniël; McLatchie, Neil Marvin; Isager, Peder M. et al.
In: Journals of Gerontology Series B: Psychological Sciences and Social Sciences, Vol. 75, No. 1, 01.01.2020, p. 45-57.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

Lakens, D, McLatchie, NM, Isager, PM, Scheel, AM & Dienes, Z 2020, 'Improving Inferences about Null Effects with Bayes Factors and Equivalence Tests', Journals of Gerontology Series B: Psychological Sciences and Social Sciences, vol. 75, no. 1, pp. 45-57. https://doi.org/10.1093/geronb/gby065

APA

Lakens, D., McLatchie, N. M., Isager, P. M., Scheel, A. M., & Dienes, Z. (2020). Improving Inferences about Null Effects with Bayes Factors and Equivalence Tests. Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 75(1), 45-57. https://doi.org/10.1093/geronb/gby065

Vancouver

Lakens D, McLatchie NM, Isager PM, Scheel AM, Dienes Z. Improving Inferences about Null Effects with Bayes Factors and Equivalence Tests. Journals of Gerontology Series B: Psychological Sciences and Social Sciences. 2020 Jan 1;75(1):45-57. Epub 2018 Jun 6. doi: 10.1093/geronb/gby065

Author

Lakens, Daniël ; McLatchie, Neil Marvin ; Isager, Peder M. et al. / Improving Inferences about Null Effects with Bayes Factors and Equivalence Tests. In: Journals of Gerontology Series B: Psychological Sciences and Social Sciences. 2020 ; Vol. 75, No. 1. pp. 45-57.

Bibtex

@article{b261e2f1939d4ddcbee0ca224f8d9f0e,
title = "Improving Inferences about Null Effects with Bayes Factors and Equivalence Tests",
abstract = "Researchers often conclude an effect is absent when a null-hypothesis significance test yields a non-significant p-value. However, it is neither logically nor statistically correct to conclude an effect is absent when a hypothesis test is not significant. We present two methods to evaluate the presence or absence of effects: Equivalence testing (based on frequentist statistics) and Bayes factors (based on Bayesian statistics). In four examples from the gerontology literature we illustrate different ways to specify alternative models that can be used to reject the presence of a meaningful or predicted effect in hypothesis tests. We provide detailed explanations of how to calculate, report, and interpret Bayes factors and equivalence tests. We also discuss how to design informative studies that can provide support for a null model or for the absence of a meaningful effect. The conceptual differences between Bayes factors and equivalence tests are discussed, and we also note when and why they might lead to similar or different inferences in practice. It is important that researchers are able to falsify predictions or can quantify the support for predicted null-effects. Bayes factors and equivalence tests provide useful statistical tools to improve inferences about null effects.",
keywords = "Bayesian statistics, Frequentist statistics, Hypothesis testing, Falsification, Bayes factors, Equivalence tests",
author = "Dani{\"e}l Lakens and McLatchie, {Neil Marvin} and Isager, {Peder M.} and Scheel, {Anne M.} and Zoltan Dienes",
note = "his is a pre-copy-editing, author-produced PDF of an article accepted for publication in The Journals of Gerontology: Series B following peer review. The definitive publisher-authenticated version Dani{\"e}l Lakens, Neil McLatchie, Peder M Isager, Anne M Scheel, Zoltan Dienes, Improving Inferences About Null Effects With Bayes Factors and Equivalence Tests, The Journals of Gerontology: Series B 2020 75: 45-57 is available online at: https://academic.oup.com/psychsocgerontology/article/75/1/45/5033832",
year = "2020",
month = jan,
day = "1",
doi = "10.1093/geronb/gby065",
language = "English",
volume = "75",
pages = "45--57",
journal = "Journals of Gerontology Series B: Psychological Sciences and Social Sciences",
publisher = "Gerontological Society of America",
number = "1",

}

RIS

TY - JOUR

T1 - Improving Inferences about Null Effects with Bayes Factors and Equivalence Tests

AU - Lakens, Daniël

AU - McLatchie, Neil Marvin

AU - Isager, Peder M.

AU - Scheel, Anne M.

AU - Dienes, Zoltan

N1 - his is a pre-copy-editing, author-produced PDF of an article accepted for publication in The Journals of Gerontology: Series B following peer review. The definitive publisher-authenticated version Daniël Lakens, Neil McLatchie, Peder M Isager, Anne M Scheel, Zoltan Dienes, Improving Inferences About Null Effects With Bayes Factors and Equivalence Tests, The Journals of Gerontology: Series B 2020 75: 45-57 is available online at: https://academic.oup.com/psychsocgerontology/article/75/1/45/5033832

PY - 2020/1/1

Y1 - 2020/1/1

N2 - Researchers often conclude an effect is absent when a null-hypothesis significance test yields a non-significant p-value. However, it is neither logically nor statistically correct to conclude an effect is absent when a hypothesis test is not significant. We present two methods to evaluate the presence or absence of effects: Equivalence testing (based on frequentist statistics) and Bayes factors (based on Bayesian statistics). In four examples from the gerontology literature we illustrate different ways to specify alternative models that can be used to reject the presence of a meaningful or predicted effect in hypothesis tests. We provide detailed explanations of how to calculate, report, and interpret Bayes factors and equivalence tests. We also discuss how to design informative studies that can provide support for a null model or for the absence of a meaningful effect. The conceptual differences between Bayes factors and equivalence tests are discussed, and we also note when and why they might lead to similar or different inferences in practice. It is important that researchers are able to falsify predictions or can quantify the support for predicted null-effects. Bayes factors and equivalence tests provide useful statistical tools to improve inferences about null effects.

AB - Researchers often conclude an effect is absent when a null-hypothesis significance test yields a non-significant p-value. However, it is neither logically nor statistically correct to conclude an effect is absent when a hypothesis test is not significant. We present two methods to evaluate the presence or absence of effects: Equivalence testing (based on frequentist statistics) and Bayes factors (based on Bayesian statistics). In four examples from the gerontology literature we illustrate different ways to specify alternative models that can be used to reject the presence of a meaningful or predicted effect in hypothesis tests. We provide detailed explanations of how to calculate, report, and interpret Bayes factors and equivalence tests. We also discuss how to design informative studies that can provide support for a null model or for the absence of a meaningful effect. The conceptual differences between Bayes factors and equivalence tests are discussed, and we also note when and why they might lead to similar or different inferences in practice. It is important that researchers are able to falsify predictions or can quantify the support for predicted null-effects. Bayes factors and equivalence tests provide useful statistical tools to improve inferences about null effects.

KW - Bayesian statistics

KW - Frequentist statistics

KW - Hypothesis testing

KW - Falsification

KW - Bayes factors

KW - Equivalence tests

U2 - 10.1093/geronb/gby065

DO - 10.1093/geronb/gby065

M3 - Journal article

VL - 75

SP - 45

EP - 57

JO - Journals of Gerontology Series B: Psychological Sciences and Social Sciences

JF - Journals of Gerontology Series B: Psychological Sciences and Social Sciences

IS - 1

ER -