Over half of bilingual advantage articles have a statistical reporting error

To err is human, and some research articles contain statistical reporting errors (for example, a p -value of 0.5 being reported as 0.05). But such errors can lead to incorrect interpretations of data and thus can impede scientific progress, so they should be kept to a minimum. The current study quan...

Full description

Saved in:
Bibliographic Details
Published in:Journal of Cultural Cognitive Science 2024-08, Vol.8 (2), p.167-176
Main Author: Schroeder, Scott R.
Format: Article
Language:English
Subjects:
Behavioral Science and Psychology
Cognitive Linguistics
Cognitive Psychology
Philosophy of Mind
Psychology
Research Paper
Citations: Items that this one cites
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:To err is human, and some research articles contain statistical reporting errors (for example, a p -value of 0.5 being reported as 0.05). But such errors can lead to incorrect interpretations of data and thus can impede scientific progress, so they should be kept to a minimum. The current study quantified statistical reporting errors in articles testing the bilingual advantage hypothesis. In total, 380 articles and over 6,000 statistical tests were assessed. Results indicated that more than 50% of articles had at least 1 error, and that on average about 10% of statistical tests in an article had an error. Fortunately, most of these errors were minor errors. However, there was some evidence for more minor errors in articles that reported a bilingual advantage (compared to articles that reported no bilingual advantage), though this difference was small and inconsistent. Researchers, reviewers, and editors involved in bilingual advantage research are encouraged to verify the accuracy of statistical reporting prior to publication.
ISSN:2520-100X
2520-1018
DOI:10.1007/s41809-024-00147-5