Questionable research practices when using confirmatory factor analysis. Issue 1 (13th February 2019)

• Record Type: Journal Article
• Authors: Crede, Marcus; Harms, Peter
• Is Part Of: Journal of managerial psychology. Volume 34:Issue 1(2019)
• Page Start: 18
• Page End: 30
• Publication Date: 2019-02-13
• Subjects: Structural equation modelling -- Scale development -- Research methods -- Psychometrics; Organizational behavior -- Periodicals; Management -- Psychological aspects -- Periodicals; Psychology, Industrial -- Periodicals; Industrial management -- Psychological aspects -- Periodicals; 158.7
• DOI: 10.1108/JMP-06-2018-0272
• Languages: English
• ISSNs: 0268-3946

Abstract

Abstract : Purpose: The purpose of this paper is to describe common questionable research practices (QRPs) engaged in by management researchers who use confirmatory factor analysis (CFA) as part of their analysis. Design/methodology/approach: The authors describe seven questionable analytic practices and then review one year of journal articles published in three top-tier management journals to estimate the base rate of these practices. Findings: The authors find that CFA analyses are characterized by a high base rate of QRPs with one practice occurring for over 90 percent of all assessed articles. Research limitations/implications: The findings of this paper call into question the validity and trustworthiness of results reported in much of the management literature. Practical implications: The authors provide tentative guidelines of how editors and reviewers might reduce the degree to which the management literature is characterized by these QRPs. Originality/value: This is the first paper to estimate the base rate of six QRPs relating to the widely used analytic tool referred to as CFA in the management literature.