P-curving x-phi: Does experimental philosophy have evidential value?. (31st October 2019)

• Record Type: Journal Article
• Authors: Stuart, Michael T; Colaço, David; Machery, Edouard
• Is Part Of: Analysis. Volume 79:Number 4(2019:Oct.)
• Page Start: 669
• Page End: 684
• Publication Date: 2019-10-31
• Subjects: Philosophy -- Periodicals; 105
• DOI: 10.1093/analys/anz007
• Languages: English
• ISSNs: 0003-2638

Abstract

Abstract: In this article, we analyse the evidential value of the corpus of experimental philosophy (x-phi). While experimental philosophers claim that their studies provide insight into philosophical problems, some philosophers and psychologists have expressed concerns that the findings from these studies lack evidential value. Barriers to evidential value include selection bias (i.e., the selective publication of significant results) and p-hacking (practices that increase the odds of obtaining a p-value below the significance level). To find out whether the significant findings in x-phi papers result from selection bias or p-hacking, we applied a p-curve analysis to a corpus of 365 x-phi chapters and articles. Our results suggest that this corpus has evidential value, although there are hints of p-hacking in a few parts of the x-phi corpus.