Start Date

11-8-2016

Description

Some of the models using partial least squares (PLS) in Information Systems (IS) field may have serious problems because do not properly address endogeneity. This may suppose a problem in IS theory building because it may lead IS scholars to non-correct results. Although the IS community’s awareness is rising, we do not have a clear understanding of the problem nor fine-grained practical guidelines on how to address the endogeneity in IS empirical research using PLS. Further, none of the PLS software packages has test of endogeneity capabilities. This paper explains and illustrates how to address endogeneity in research using PLS path modeling, and contribute to IS research in two ways: (1) we define the problem of endogeneity in empirical research and explain its main causes with IS research examples, (2) we show how to address endogeneity by correcting for omitted variables in PLS path modeling with composite and factor models.

Share

COinS
 
Aug 11th, 12:00 AM

How to Address Endogeneity in Partial Least Squares Path Modeling

Some of the models using partial least squares (PLS) in Information Systems (IS) field may have serious problems because do not properly address endogeneity. This may suppose a problem in IS theory building because it may lead IS scholars to non-correct results. Although the IS community’s awareness is rising, we do not have a clear understanding of the problem nor fine-grained practical guidelines on how to address the endogeneity in IS empirical research using PLS. Further, none of the PLS software packages has test of endogeneity capabilities. This paper explains and illustrates how to address endogeneity in research using PLS path modeling, and contribute to IS research in two ways: (1) we define the problem of endogeneity in empirical research and explain its main causes with IS research examples, (2) we show how to address endogeneity by correcting for omitted variables in PLS path modeling with composite and factor models.