Stopping "How" from Driving "What": Advice on Avoiding Measurement Item Mis-specification

The recent debate on measurement item mis-specification has renewed interest in the construct explication process; however, the focus of this debate has been the interaction between latent variables and their measures (“how”), instead of between latent variables and the constructs they represent (“what”). This paper highlights the problems created when framing the formativereflective discussion in terms of “how” instead of “what.” We argue it is necessary to precede measurement item construction (or appropriation) with careful construct definition and operationalization by focusing first on what is being measured, rather than on how it will be measured. Two solutions for better understanding measurement item construction are proposed: 1) carefully defining assumptions and boundary conditions of a construct in its theoretical context, and 2) carefully defining assumptions and boundary conditions of a latent variable in its measurement context. Recommendations regarding correct interpretation of the measurement process by authors, reviewers, and editors are offered.