Content area
Full text
Partial least squares structural equation modeling (PLS-SEM) belongs to the toolbox of methods that researchers in marketing and other disciplines frequently use in their empirical analyses (Hair et al., 2012; Rasoolimanesh and Ali, 2018; Nitzl, 2016). Even though widely appreciated, PLS-SEM has been in the center of a lively scholarly debate for several years (Latan and Noonan, 2017; Richter et al., 2016; Rigdon et al., 2017b), which accompanied and fostered the method’s continuous development and extension. Contributions pointing out both strengths and limitations of the method have appeared at a rapid pace, with some early articles incorrectly concluding “there is no use for PLS whatsoever” (Antonakis et al., 2010, p. 1103), while more recent works conclude that researchers should “feel the love for PLS” (Petter, 2018, p. 12). The underlying discussions certainly have impacted methodological research on PLS-SEM (Khan et al., 2019), and methods scholars were quick to tackle some of the criticisms by advancing the method, while debunking other issues as being unsubstantiated or grossly exaggerated (Henseler et al., 2014; Sarstedt et al., 2016). Unfortunately, a recent article with alleged “new guidelines” was published (Henseler et al., 2016) that was incorrectly positioned as the latest thinking in PLS-SEM[1].
While the concept of “new guidelines” sounds intuitively appealing, a careful consideration of recent research shows that several of these guidelines have led to substantial controversy, and thereby prevented unambiguous agreement on how to deal with emerging topics such as goodness-of-fit testing, the nature of composite models, and several more. At the same time, some users of PLS-SEM seem to be unaware of recent developments which have clearly extended and improved the toolbox for analyzing PLS path models (Latan and Noonan, 2017; Hair et al., 2018a, 2018b; Matthews et al., 2018).
Taking up Henseler et al.’s (2017b) approach, this research note highlights some of these recent developments and offers a counterpoint to emerging calls to change PLS-SEM use, several of which we argue are misguided and add little or no value. Our overarching aim of this research note is extending prior efforts (Henseler, 2017b) to enable researchers, reviewers, and editors to judge recent controversies on their own (Hair et al., 2013). Most importantly, this...