Return to search

Investigating the Misrepresentation of Statistical Significance in Empirical Articles

In an attempt to preserve research integrity, the aim of this study is to examine how often statistical results are being misrepresented in empirical studies by using terms such as “marginally significant,” “approached significance,” or “trend toward significance” when interpreting findings. The use of these terms gives ambiguous significance to results that are in fact nonsignificant, which threatens future research by contributing to issues such as the replication crisis. For this study, data were coded from 437 empirical articles published online in The Journal of Personality and Social Psychology (JPSP) over a 4-year period between 2017 and 2020. According to our findings, although misrepresentation of statistical results are prevalent within JPSP articles, rates decreased significantly over the four-year time period examined. Additionally, as the number of studies published in JPSP increased each year during the four-year period examined, there may be a potential rise in representatively sound studies and decrease of misrepresentation within this discipline.

Identiferoai:union.ndltd.org:ETSU/oai:dc.etsu.edu:honors-1824
Date01 May 2021
CreatorsLybrand, Blythe, Blackhart, Ginette, Parish, Amanda, Lowe, Hannah
PublisherDigital Commons @ East Tennessee State University
Source SetsEast Tennessee State University
Detected LanguageEnglish
Typetext
Formatapplication/pdf
SourceUndergraduate Honors Theses
RightsCopyright by the authors., http://creativecommons.org/licenses/by-nc-nd/3.0/

Page generated in 0.0019 seconds