2017
DOI: 10.31219/osf.io/sbp6k
|View full text |Cite
Preprint
|
Sign up to set email alerts
|

Why Psychologists Should by Default Use Welch's t-test Instead of Student's t-test (in press for the International Review of Social Psychology).

Abstract: When comparing two independent groups, researchers in Psychology commonly use Student’s t-test. Assumptions of normality and of homogeneity of variance underlie this test. More often than not, when these conditions are not met, Student’s t-test can be severely biased, and leads to invalid statistical inferences. Moreover, we argue that the assumption of equal variances will seldom hold in psychological research and that choosing between Student’s t-test or Welch’s t-test based on the outcomes of a test of the … Show more

Help me understand this report

Search citation statements

Order By: Relevance

Paper Sections

Select...
2
2
1

Citation Types

1
49
0
1

Year Published

2018
2018
2022
2022

Publication Types

Select...
7
1

Relationship

0
8

Authors

Journals

citations
Cited by 60 publications
(51 citation statements)
references
References 0 publications
1
49
0
1
Order By: Relevance
“…Based on the simulation studies of Koh and Cribbie (2013), we can recommend that the non‐inferiority test based on the Welch's F statistic (i.e., the test with p ‐value calculated from equation (15)) is almost always preferable (with regard to statistical power and Type I error rate) to the test which requires an assumption of homogeneous variance (i.e., the test with p ‐value calculated from equation (12)). This agrees with similar recommendations for using Welch's t test (e.g., Delacre, Lakens, & Leys, 2017; Ruxton, 2006). We also point interested readers to the related work of Jan and Shieh (2019).…”
Section: A Non‐inferiority Test For the Anovaboldη2 Parametersupporting
confidence: 89%
“…Based on the simulation studies of Koh and Cribbie (2013), we can recommend that the non‐inferiority test based on the Welch's F statistic (i.e., the test with p ‐value calculated from equation (15)) is almost always preferable (with regard to statistical power and Type I error rate) to the test which requires an assumption of homogeneous variance (i.e., the test with p ‐value calculated from equation (12)). This agrees with similar recommendations for using Welch's t test (e.g., Delacre, Lakens, & Leys, 2017; Ruxton, 2006). We also point interested readers to the related work of Jan and Shieh (2019).…”
Section: A Non‐inferiority Test For the Anovaboldη2 Parametersupporting
confidence: 89%
“…To demonstrate the magnitude of the observed effects for F tests, partial eta‐squared (η p 2 ) values are shown along with their 90% CIs (Steiger, 2004). For t tests, Welch‐corrected statistics are reported for parametric tests (Delacre, Lakens, & Leys, 2017), Wilcoxon statistics for nonparametric tests, and, to demonstrate the magnitude of the observed effects, Cohen's d values as standardized mean differences and their 95% CIs (Lakens, 2013). All analyses were conducted in R (R Core Team, 2019; via: Kelley, 2018; Lawrence, 2016).…”
Section: Methodsmentioning
confidence: 99%
“…Eigenvector alignment is applied herein to detect differences between pairs of ROIs in functional connectivity networks. The significance of these differences in EA are determined using Welch's t test [32]. To mitigate against false detections of significance, EA is only assessed for ROI pairs that are significant when compared with connectivity matrices generated with uniformly distributed random numbers between 0 and 1.…”
Section: Filtering Resultsmentioning
confidence: 99%