Abstract
Purpose – Despite the increasing use of formative measurement models in literature, little is known about potential consequences for substantive theory testing. Against this background, the aims of this chapter are (1) to highlight some problems that may arise when formative instead of reflective measures are used to test even simple theoretical models with covarianced-based methodologies, (2) to illustrate some approaches that might help overcome these problems, (3) to pinpoint potential interpretation difficulties of the results involving re-specified measurement models, and (4) to stimulate discussion on the implications for theory development when models are tested with formative measures.Methodology/approach – Potential consequences of formative measurement models for theory testing are highlighted using an empirical study on consumer animosity as an illustrative example and applying covarianced-based structural equations modeling procedures for estimation purposes.Findings – The empirical study shows (a) that some scaling options for the (composite) latent variable result in non-convergence problems, (b) that, assuming convergence, parameter estimates, standard errors, and significance levels vary depending on the scaling method used, and (c) that goodness-of-fit statistics cannot be used as diagnostic measures for the appropriateness of divergent results.Originality/value of paper – The contribution of this chapter is two-fold: First, it shows that to enable estimation, it is often necessary to modify (i.e., expand) the original theoretical model in a conceptually reasonable manner and to do so before data collection. Second, it demonstrates that alternative scaling options for composite latent variables may result in inconsistent substantive conclusions. Consequently, the impact of formative measurement on theory testing is a critical topic and needs to receive further attention in future literature.
Keywords
Related Publications
Structural equation modeling in practice: A review and recommended two-step approach.
In this article, we provide guidance for substantive researchers on the use of structural equation modeling in practice for theory testing and development. We present a comprehe...
Interpretational confounding is due to misspecification, not to type of indicator: Comment on Howell, Breivik, and Wilcox (2007).
R. D. Howell, E. Breivik, and J. B. Wilcox (2007) have argued that causal (formative) indicators are inherently subject to interpretational confounding. That is, they have argue...
A Review and Synthesis of the Measurement Invariance Literature: Suggestions, Practices, and Recommendations for Organizational Research
The establishment of measurement invariance across groups is a logical prerequisite to conducting substantive cross-group comparisons (e.g., tests of group mean differences, inv...
Evaluating Structural Equation Models with Unobservable Variables and Measurement Error
The statistical tests used in the analysis of structural equation models with unobservable variables and measurement error are examined. A drawback of the commonly applied chi s...
A Critical Review of Construct Indicators and Measurement Model Misspecification in Marketing and Consumer Research
A review of the literature suggests that few studies use formative indicator measurement models, even though they should. Therefore, the purpose of this research is to (a) discu...
Publication Info
- Year
- 2011
- Type
- book-chapter
- Pages
- 11-30
- Citations
- 39
- Access
- Closed
External Links
Social Impact
Social media, news, blog, policy document mentions
Citation Metrics
Cite This
Identifiers
- DOI
- 10.1108/s1474-7979(2011)0000022004