Manipulating the Alpha Level Cannot Cure Significance Testing

Trafimow, David and Amrhein, Valentin and Areshenkoff, Corson N. and Barrera-Causil, Carlos J. and Beh, Eric J. and Bilgiç, Yusuf K. and Bono, Roser and Bradley, Michael T. and Briggs, William M. and Cepeda-Freyre, Héctor A. and Chaigneau, Sergio E. and Ciocca, Daniel R. and Correa, Juan C. and Cousineau, Denis and de Boer, Michiel R. and Dhar, Subhra S. and Dolgov, Igor and Gómez-Benito, Juana and Grendar, Marian and Grice, James W. and Guerrero-Gimenez, Martin E. and Gutiérrez, Andrés and Huedo-Medina, Tania B. and Jaffe, Klaus and Janyan, Armina and Karimnezhad, Ali and Korner-Nievergelt, Fränzi and Kosugi, Koji and Lachmair, Martin and Ledesma, Rubén D. and Limongi, Roberto and Liuzza, Marco T. and Lombardo, Rosaria and Marks, Michael J. and Meinlschmidt, Gunther and Nalborczyk, Ladislas and Nguyen, Hung T. and Ospina, Raydonal and Perezgonzalez, Jose D. and Pfister, Roland and Rahona, Juan J. and Rodríguez-Medina, David A. and Romão, Xavier and Ruiz-Fernández, Susana and Suarez, Isabel and Tegethoff, Marion and Tejo, Mauricio and van de Schoot, Rens and Vankov, Ivan I. and Velasco-Forero, Santiago and Wang, Tonghui and Yamada, Yuki and Zoppino, Felipe C. M. and Marmolejo-Ramos, Fernando. (2018) Manipulating the Alpha Level Cannot Cure Significance Testing. Frontiers in Psychology, 9 (699). pp. 1-7.

Full text not available from this repository.

Official URL: https://edoc.unibas.ch/69390/

Downloads: Statistics Overview

Abstract

We argue that making accept/reject decisions on scientific hypotheses, including a recent call for changing the canonical alpha level from; p; = 0.05 to; p; = 0.005, is deleterious for the finding of new discoveries and the progress of science. Given that blanket and variable alpha levels both are problematic, it is sensible to dispense with significance testing altogether. There are alternatives that address study design and sample size much more directly than significance testing does; but none of the statistical tools should be taken as the new magic method giving clear-cut mechanical answers. Inference should not be based on single studies at all, but on cumulative evidence from multiple independent studies. When evaluating the strength of the evidence, we should consider, for example, auxiliary assumptions, the strength of the experimental design, and implications for applications. To boil all this down to a binary decision based on a; p; -value threshold of 0.05, 0.01, 0.005, or anything else, is not acceptable.

Faculties and Departments:	03 Faculty of Medicine > Bereich Medizinische Fächer (Klinik) > Psychosomatik > Psychosomatik (Schäfert) 03 Faculty of Medicine > Departement Klinische Forschung > Bereich Medizinische Fächer (Klinik) > Psychosomatik > Psychosomatik (Schäfert) 07 Faculty of Psychology > Departement Psychologie > Health & Intervention > Klinische Psychologie und Epidemiologie (Lieb)
UniBasel Contributors:	Meinlschmidt, Gunther and Amrhein, Valentin and Tegethoff, Marion
Item Type:	Article, refereed
Article Subtype:	Research Article
Publisher:	Frontiers Media
e-ISSN:	1664-1078
Note:	Publication type according to Uni Basel Research Database: Journal article
Related URLs:	Document
Identification Number:	doi: 10.3389/fpsyg.2018.00699 pmid: 29867666 isi: 000432576400002
Last Modified:	07 Aug 2019 13:09
Deposited On:	18 Feb 2019 15:57

Repository Staff Only: item control page