A perspective on using partial least squares structural equation modelling in data articles

• Published in: Data in brief Vol. 48; p. 109074
• Main Authors: Ringle, Christian M., Sarstedt, Marko, Sinkovics, Noemi, Sinkovics, Rudolf R.
• Format: Journal Article
• Language: English
• Published: Netherlands Elsevier Inc 01.06.2023; Elsevier
• Subjects: Data; Discriminant validity; HTMT; Open science; Partial least squares; PLS-SEM; Structural equation modelling; HTMT; Structural equation modelling; Discriminant validity; Open science; Partial least squares; PLS-SEM
• ISSN: 2352-3409; 2352-3409
• DOI: 10.1016/j.dib.2023.109074

This perspective article on using partial least squares structural equation modelling (PLS-SEM) is intended as a guide for authors who wish to publish datasets that can be analysed with this method as stand-alone data articles. Stand-alone data articles are different from supporting data articles in that they are not linked to a full research article published in another journal. Nevertheless, authors of stand-alone data articles will be required to clearly demonstrate and justify the usefulness of their dataset. This perspective article offers actionable recommendations regarding the conceptualisation phase, the types of data suitable for PLS-SEM and quality criteria to report, which are generally applicable to studies using PLS-SEM. We also present adjusted versions of the HTMT metric for discriminant validity testing that broaden its applicability. Further, we highlight the benefit of linking data articles to already published research papers that employ the PLS-SEM method.