Baseline comparisons and covariate fishing: bad statistical habits we should have broken yesterday

S.L.K. Gruijters

Baseline comparisons and covariate fishing: bad statistical habits we should have broken yesterday

Research output: Contribution to journal › Article › Academic › peer-review

Abstract

Checks on baseline differences in randomized controlled trials (RCTs) are often done using null-hypothesis significance tests (NHSTs). However, the use of NHSTs to establish the degree of baseline similarity is inappropriate, potentially misleading, and, simply, logically incoherent.

Original language	English
Pages (from-to)	205-209
Number of pages	5
Journal	The European Health Psychologist
Volume	18
Issue number	5
Publication status	Published - 8 Nov 2016
Externally published	Yes

Access to Document

https://www.ehps.net/ehp/index.php/contents/article/view/1661/pdf_161Licence: CC BY-SA

Cite this

@article{b63c880069cf4100b593b79b3f0e4017,

title = "Baseline comparisons and covariate fishing: bad statistical habits we should have broken yesterday",

abstract = "Checks on baseline differences in randomized controlled trials (RCTs) are often done using null-hypothesis significance tests (NHSTs). However, the use of NHSTs to establish the degree of baseline similarity is inappropriate, potentially misleading, and, simply, logically incoherent.",

author = "S.L.K. Gruijters",

year = "2016",

month = nov,

day = "8",

language = "English",

volume = "18",

pages = "205--209",

journal = "The European Health Psychologist",

issn = "2225-6962",

publisher = "European Health Psychology Society",

number = "5",

}

TY - JOUR

T1 - Baseline comparisons and covariate fishing

T2 - bad statistical habits we should have broken yesterday

AU - Gruijters, S.L.K.

PY - 2016/11/8

Y1 - 2016/11/8

N2 - Checks on baseline differences in randomized controlled trials (RCTs) are often done using null-hypothesis significance tests (NHSTs). However, the use of NHSTs to establish the degree of baseline similarity is inappropriate, potentially misleading, and, simply, logically incoherent.

AB - Checks on baseline differences in randomized controlled trials (RCTs) are often done using null-hypothesis significance tests (NHSTs). However, the use of NHSTs to establish the degree of baseline similarity is inappropriate, potentially misleading, and, simply, logically incoherent.

M3 - Article

SN - 2225-6962

VL - 18

SP - 205

EP - 209

JO - The European Health Psychologist

JF - The European Health Psychologist

IS - 5

ER -

Baseline comparisons and covariate fishing: bad statistical habits we should have broken yesterday

Abstract

Access to Document

Fingerprint

Cite this