The Negative Consequences of Measurement Model Misspecification: A Response to Aguirre-urreta and Marakas

• Published in: MIS quarterly 2012-03, Vol.36 (1), p.139-146
• Main Authors: Jarvis, Cheryl Burke, MacKenzie, Scott B., Podsakoff, Philip M.
• Format: Article
• Language: English
• Subjects: Coefficients; Construct validity; Constructive empiricism; Consumer research; Estimation bias; Marketing; Measurement techniques; Modeling; Parametric models; Reporting standards; Research methodology; Research methods; Research Notes; Specifications; Studies
• ISSN: 0276-7783; 2162-9730
• DOI: 10.2307/41410410

It has been more than 40 years since Blalock (1964) noted the distinction between what he called "cause" (formative) and "effect" (reflective) indicators of latent variables, and three decades since the academic literature recognized that some SEM measurement models don't fit classical test theory's assumptions about the direction of causality of the relationships between constructs and their indicators. However, recently researchers in a variety of disciplines have raised questions surrounding the correct conceptualization and operationalization of formative indicator measurement models. In these articles the authors attempted not only to illustrate the extent of measurement model misspecification in their literatures and the potential consequences of such misspecification but, more importantly, to provide needed guidance to researchers about how to determine which type of measurement model is conceptually appropriate, develop and purify scales using formative indicators, and specify structural equation models incorporating these composite latent variables.