TITLE

SOME NEW THOUGHTS ON FORMATIVE AND REFLECTIVE MEASUREMENT MODELS IN MARKETING RESEARCH

AUTHOR(S)
Martínez, Jose A.; Martínez, Laura
PUB. DATE
June 2011
SOURCE
Portuguese Journal of Marketing / Revista Portuguesa de Marketin;2011, Issue 26, p74
SOURCE TYPE
Academic Journal
DOC. TYPE
Article
ABSTRACT
We discuss the current conceptualization of reflective and formative measurement models in marketing research. We reformulate the definition of formative models as follows: in a formative measurement model in which a construct is defined in terms of its measurements, the variable of interest susceptible to be measured is an algebraic construction that cannot be measured in a reflective way. We also introduce a new type of indicator, the phantom-effect indicator, which is a numerical trick used to correctly identify some types of causal models. We based our thinking on the definition of a latent variable and the interpretation of the error term, breaking from the current thinking that formative variables do not exist independently from their indicators. In addition, we propose that variables can be classified into three disparate categories with regard to the form of measurement: cake, love, and player performance. We use these terms to illustrate our reasoning. Finally, we offer guidelines to model formative variables when they are causes or consequences of other variables.
ACCESSION #
82313291

 

Related Articles

  • INTERPRETING LATENT VARIABLE INTERACTIONS. Ping Jr., Robert A. // AMA Winter Educators' Conference Proceedings;2002, Vol. 13, p213 

    Using examples, the paper proposes a new procedure for interpreting interactions and quadratics in latent variables. The procedure sheds additional light on these variables, and the paper is intended to help bridge what may be a gap in survey researchers' understanding of the nature of...

  • CONSTRUCT VALIDATION IN MARKETING: A COMPARISON OF METHODS IN ASSESSING THE VALIDITY OF THE AFFECTIVE, CONATIVE, AND COGNITIVE COMPONENTS OF ATTITUDES. John, George; Rave, Torger // Advances in Consumer Research;1979, Vol. 6 Issue 1, p288 

    The assessment of the validity of constructs in marketing are examined. An empirical comparison of different methods of establishing convergent and discriminant validity within the framework of multitrait-multimethod data analysis is made in the case of the affective, conative and cognitive...

  • Identification and Estimation of Dichotomous Latent Variables Models Using Panel Data. Hsiao, Cheng // Review of Economic Studies;Jul91, Vol. 58 Issue 4, p717 

    Identification conditions for binary choice errors in variables models are explored. Conditions for the consistency and asymptotic normality of the maximum likelihood estimators of binary choice models with unbounded explanatory variables are given Two- or three-step estimators to simplify...

  • Testing the multivariate normality of Australian stock returns. Gray, Philip; Kalotay, Egon // Australian Journal of Management (University of New South Wales);Dec98, Vol. 23 Issue 2, p135 

    Utilizes a multivariate test procedure, based on the generalized method of moments, to test whether residuals from market model regressions are multivariate normal. Background on multivariate normal assumption; Distinctions between multivariate and univariate assumptions; Tests on asset pricing...

  • Estimating Latent Structure Models with Categorical Variables: One-Step Versus Three-Step Estimators. Bolck, Annabel; Croon, Marcel; Hagenaars, Jacques // Political Analysis;Jan2004, Vol. 12 Issue 1, p3 

    We study the properties of a three-step approach to estimating the parameters of a latent structure model for categorical data and propose a simple correction for a common source of bias. Such models have a measurement part (essentially the latent class model) and a structural (causal) part...

  • Structural equation modeling in medical research: a primer. Beran, Tanya N.; Violato, Claudio // BMC Research Notes;2010, Vol. 3, p267 

    Background: Structural equation modeling (SEM) is a set of statistical techniques used to measure and analyze the relationships of observed and latent variables. Similar but more powerful than regression analyses, it examines linear causal relationships among variables, while simultaneously...

  • A Parsimonious Estimating Technique for Interaction and Quadratic Latent Variables. Ping Jr., Robert A. // Journal of Marketing Research (JMR);Aug1995, Vol. 32 Issue 3, p336 

    The author proposes an alternative estimation technique for quadratic and interaction latent variables in structural equation models using LISREL, EQS. and CALIS. The technique specifies these variables with single indicants. The loading and error terms for the single indicants can be specified...

  • Factor Analysis and Missing Data. Kamakura, Wagner A.; Wedel, Michel // Journal of Marketing Research (JMR);Nov2000, Vol. 37 Issue 4, p490 

    The authors study the estimation of factor models and the imputation of missing data and propose an approach that provides direct estimates of factor weights without the replacement of missing data with imputed values. First, the approach is useful in applications of factor analysis in the...

  • On the Use of Structural Equation Models in Experimental Designs. BAGOZZI, RICHARD P.; YOUJAE YI // Journal of Marketing Research (JMR);Aug1989, Vol. 26 Issue 3, p271 

    New procedures are developed and illustrated for the analysis of experimental data with particular emphasis on MANOVA and MANCOVA designs. The authors begin with one-way designs, including overall tests of significance, step-down analyses, and the use of latent variables. Next they describe a...

Share

Read the Article

Courtesy of THE LIBRARY OF VIRGINIA

Sorry, but this item is not currently available from your library.

Try another library?
Sign out of this library

Other Topics