Dansk Center for Forskningsanalyse

NHST is still logically flawed

Publikation: Bidrag til tidsskrift/Konferencebidrag i tidsskrift /Bidrag til avisLetterpeer review

Standard

NHST is still logically flawed. / Schneider, Jesper W.

I: Scientometrics, Bind 115, Nr. 1, 2018, s. 627-635.

Publikation: Bidrag til tidsskrift/Konferencebidrag i tidsskrift /Bidrag til avisLetterpeer review

Harvard

Schneider, JW 2018, 'NHST is still logically flawed', Scientometrics, bind 115, nr. 1, s. 627-635. https://doi.org/10.1007/s11192-018-2655-4

APA

CBE

MLA

Vancouver

Schneider JW. NHST is still logically flawed. Scientometrics. 2018;115(1):627-635. doi: 10.1007/s11192-018-2655-4

Author

Schneider, Jesper W. / NHST is still logically flawed. I: Scientometrics. 2018 ; Bind 115, Nr. 1. s. 627-635.

Bibtex

@article{4891cb12530f42bfa5c513bf3bd8c15d,
title = "NHST is still logically flawed",
abstract = "In this elaborate response to Wu (in Scientometrics, 2018), I maintain that null hypothesis significance testing (NHST) is logically flawed. Wu (2018) disagrees with this claim presented in Schneider (in Scientometrics 102(1):411–432, 2015). In this response, I examine the claim in more depth and demonstrate that since NHST is based on one conditional probability alone and framed in a probabilistic modus tollens framework of reasoning, it is by definition logically invalid. I also argue that disregarding this logically fallacy, as most researchers do, and treating the p value as a heuristic value for dichotomous decisions against the null hypothesis, is a risky business that often leads to false-positive claims.",
keywords = "Conditional probability, Inference logic, Modus tollens, Null hypothesis significance test, p values, Statistical inference",
author = "Schneider, {Jesper W.}",
year = "2018",
doi = "10.1007/s11192-018-2655-4",
language = "English",
volume = "115",
pages = "627--635",
journal = "Scientometrics",
issn = "0138-9130",
publisher = "Akademiai Kiado Rt.",
number = "1",

}

RIS

TY - JOUR

T1 - NHST is still logically flawed

AU - Schneider, Jesper W.

PY - 2018

Y1 - 2018

N2 - In this elaborate response to Wu (in Scientometrics, 2018), I maintain that null hypothesis significance testing (NHST) is logically flawed. Wu (2018) disagrees with this claim presented in Schneider (in Scientometrics 102(1):411–432, 2015). In this response, I examine the claim in more depth and demonstrate that since NHST is based on one conditional probability alone and framed in a probabilistic modus tollens framework of reasoning, it is by definition logically invalid. I also argue that disregarding this logically fallacy, as most researchers do, and treating the p value as a heuristic value for dichotomous decisions against the null hypothesis, is a risky business that often leads to false-positive claims.

AB - In this elaborate response to Wu (in Scientometrics, 2018), I maintain that null hypothesis significance testing (NHST) is logically flawed. Wu (2018) disagrees with this claim presented in Schneider (in Scientometrics 102(1):411–432, 2015). In this response, I examine the claim in more depth and demonstrate that since NHST is based on one conditional probability alone and framed in a probabilistic modus tollens framework of reasoning, it is by definition logically invalid. I also argue that disregarding this logically fallacy, as most researchers do, and treating the p value as a heuristic value for dichotomous decisions against the null hypothesis, is a risky business that often leads to false-positive claims.

KW - Conditional probability

KW - Inference logic

KW - Modus tollens

KW - Null hypothesis significance test

KW - p values

KW - Statistical inference

U2 - 10.1007/s11192-018-2655-4

DO - 10.1007/s11192-018-2655-4

M3 - Letter

AN - SCOPUS:85041095998

VL - 115

SP - 627

EP - 635

JO - Scientometrics

JF - Scientometrics

SN - 0138-9130

IS - 1

ER -