A Note on How to Conduct a Factor-Based PLS-SEM Analysis

A Note on How to Conduct a Factor-Based PLS-SEM Analysis

Ned Kock (Division of International Business and Technology Studies, Texas A&M International University, Laredo, TX, USA)
Copyright: © 2015 |Pages: 9
DOI: 10.4018/ijec.2015070101
OnDemand PDF Download:
$30.00
List Price: $37.50

Abstract

The composite-factor estimation dichotomy has been the epicenter of a long and ongoing debate among proponents and detractors of the use of the partial least squares (PLS) approach for structural equation modeling (SEM). In this brief research note the author discusses the implementation of a new method to conduct factor-based PLS-SEM analyses, which could be a solid step in the resolution of this debate. This method generates estimates of both true composites and factors, in two stages, fully accounting for measurement error. The author's discussion is based on an illustrative model in the field of e-collaboration. A Monte Carlo experiment suggests that model parameters generated by the method are asymptotically unbiased. The method is implemented as part of the software WarpPLS, starting in version 5.0. This note provides enough details for the method's implementation in other venues such as R and GNU Octave.
Article Preview

Illustrative Model

Our discussion is based on the illustrative model depicted in Figure 1, which builds on an actual empirical study in the field of e-collaboration (Kock, 2005; 2008; Kock & Lynn, 2012). This illustrative model incorporates the belief that e-collaboration technology use () by teams of workers tasked with the development of new products in organizations (e.g., a new consulting service, a new car part) increases both team efficiency () and team performance (). Team efficiency () is related to the speed and cost at which teams operate. Team performance () is related to how well the new products developed by teams perform in terms of sales and profits.

In this illustrative model is the path coefficient for the link going from factor to factor ; is the loading for the jth indicator of factor ; is the indicator error for the jth indicator of factor ; is the measurement error associated with ; and and is the structural error associated with , which exists only for endogenous factors. An endogenous factor has at least one other factor pointing at it in the model.

Complete Article List

Search this Journal:
Reset
Open Access Articles: Forthcoming
Volume 13: 4 Issues (2017): 2 Released, 2 Forthcoming
Volume 12: 4 Issues (2016)
Volume 11: 4 Issues (2015)
Volume 10: 4 Issues (2014)
Volume 9: 4 Issues (2013)
Volume 8: 4 Issues (2012)
Volume 7: 4 Issues (2011)
Volume 6: 4 Issues (2010)
Volume 5: 4 Issues (2009)
Volume 4: 4 Issues (2008)
Volume 3: 4 Issues (2007)
Volume 2: 4 Issues (2006)
Volume 1: 4 Issues (2005)
View Complete Journal Contents Listing