Recently there has been a renewed interest in formative measurement and its role in properly specified models. Formative measurement models are difficult to identify, and hence to estimate and test. Existing solutions to the identification problem are shown to not adequately represent the formative constructs of interest. We propose a new two-step approach to operationalize a formatively measured construct that allows a closely matched common factor equivalent to be included in any structural equation model. We provide an artificial example and an original empirical study of privacy to illustrate our approach. Detailed proofs are given in an appendix.

No description available.

Ma thèse développe deux parties. La première considère que la psychologie expérimentale doit clarifier son positionnement épistémologique pour clarifier la validation formelle de sa démarche, sans forcément devoir se référencer au cadre de la Science Physique. A partir d’une réflexion critique, je propose de décaler le cadre épistémologique en psychologie et de poser clairement un cadre pragmatique-réaliste. La thèse essentielle défendue dans ce travail est : 1/ les propriétés mentales doivent être comprises comme des phénomènes émergents, ce qui implique que leurs analyses ne peuvent se faire ni au niveau neuronal, ni au niveau de la dynamique interne de processus cognitifs, mais nécessairement au niveau de ces phénomènes émergents ; 2/ pour analyser les propriétés mentales comme formes émergentes, la psychométrie a besoin d’user de concepts qui sont en tension permanente entre une objectivité et une intersubjectivité ; en conséquence, la psychométrie doit affirmer une démarche pragmatiste-réaliste, en rupture avec l’empirisme-réaliste classique ; 3/ une approche pragmatiste-réaliste, basée entre autre sur l’abduction, permet de dépasser les contradictions pointées dans la littérature académique sur les propriétés mentales et leurs mesures ; 4/ un cadre de mesure de propriétés mentales par des variables latentes devient dès lors possible si ce cadre est compris lui aussi comme pragmatiste-réaliste ; 5/ mais ce recours au pragmatisme-réaliste renvoie en conséquence une critique à la fois des modèles avec variables latentes développés dans la littérature académique et les usages sociaux de ces modèles. La seconde partie de ma thèse porte sur un cadre particulier de formalisation des variables latentes : le cadre formatif. Je développe des simulations Monte Carlo pour vérifier le spectre des paramètres permettant une mesure formative efficiente dans le cadre d’un positionnement réaliste-empirique. / My thesis considers that experimental psychology must clarify its epistemological position to clarify the formal validation of its approach, without necessarily having to refer to the framework of Science Physics. From a critical reflection, I propose to shift the epistemological framework in psychology and clearly pose a pragmatic-realistic framework. The main thesis of this work is: 1 / mental properties must be understood as emerging phenomena, which implies that their analysis can not be done nor at the neuronal level, nor at the internal dynamics of cognitive processes, but necessarily at these emerging phenomena; 2 / to analyze the mental properties as emerging forms, psychometrics need to use concepts that are in permanent tension between objectivity and intersubjectivity; accordingly, psychometrics must assert a pragmatic-realist approach, breaking with classical empiricism-realistic; 3 / a pragmatist-realistic approach, based among other things on the abduction, can overcome the contradictions pointed in the academic literature on mental properties and their measurements; 4 / a framework for measuring mental properties by latent variables becomes possible if the framework is also understood as a pragmatic-realist; 5 / but use realistic-pragmatic returns accordingly critical of both models with latent variables developed in the academic literature and the social uses of these models. The second part of my thesis focuses on a specific part of formalization of latent variables: the formative model. I develop Monte Carlo simulations to check the range of parameters for efficient formative measure as part of a realistic-empirical positioning.

In the operations management research, the inconsistent use of the same term for different concepts and the use of the similar concepts for different constructs potentially causes theoretical and statistical problems. This research addresses the importance of construct definitions and specification methodologically within the context of quality and sustainability management. It involves three essays using multiple quantitative methods such as partial least squares structural equation modeling and multiple regression in different consumer decision-making models in the automobile industry. In the first two essays, a comprehensive literature review results in definition and contextualization of the quality and sustainability constructs as applied to operations management and marketing research. The relationships of these constructs with consumer behavior are empirically tested. Building upon the first two essays, the third essay addresses the methodological issues on formative and reflective measurements by summarizing a procedure of validating formative measurements. The quality construct was used to illustrate the methodology. This research contributes to the literature, theory, and practices in the area of quality and sustainability management.