Bootstrapping and PLS-SEM: A step-by-step guide to get more out of your bootstrap results

Abstract

Statistical inference, which relies on bootstrapping in partial least squares structural equation modeling (PLS-SEM), lies at the heart of developing practically relevant and academically rigorous theory. Inspection of PLS-SEM applications in European management research reveals that there is still much to be gained in terms of bootstrapping. This paper suggests several bootstrapping best practices and demonstrates how to conduct them for frequently encountered, yet often ignored, PLS-SEM situations such as the assessment of (non) direct effects, the comparison of effects, and the evaluation of the coefficient of determination.