Identifying Hidden Structures in Marketing's Structural Models through Universal Structure Modeling

Buckler, F. W.; Hennig‐Thurau, Thorsten

doi:10.15358/0344-1369-2008-jrm-2-47

Cited by 19 publications

(25 citation statements)

References 20 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…Rather, the true model could differ markedly from the estimated model in terms of the number of composite variables, the pattern of free and fixed parameters, and the causal flow of the model. Therefore, when PLS-PM practitioners do not have a guiding theory, a more rigorous and sophisticated approach is needed to reveal plausible model structures Presently, we are aware of only one technique for model discovery in the PLS-PM context, namely universal structure modeling (USM; Buckler & Hennig-Thurau, 2008;Turkyilmaz, Oztekin, Zaim, & Demirel, 2013), which is implemented in the Neusrel software package (http://www.neusrel.com/welcome/). Briefly, USM proceeds in two steps: (1) the use of PLS-based exploratory algorithms to assign observed variables to a user-specified number of composites; and (2) application of neural networks to discover the optimal system of linear, nonlinear, and interactive pathways among the composites.…”

Section: Concerning Henseler Et Al's Claim That Pls-pm Is An Exploramentioning

confidence: 99%

Reflections on Partial Least Squares Path Modeling

McIntosh

Edwards

Antonakis

2014

Organizational Research Methods

181

125

View full text Add to dashboard Cite

The purpose of the present article is to take stock of a recent exchange in Organizational Research Methods between critics (Rönkkö & Evermann, 2013) and proponents (Henseler et al., 2014) of partial least squares path modeling (PLS-PM). The two target articles were centered around six principal issues, namely whether PLS-PM: (1) can be truly characterized as a technique for structural equation modeling (SEM); (2) is able to correct for measurement error; (3) can be used to validate measurement models; (4) accommodates small sample sizes; (5) is able to provide null hypothesis tests for path coefficients; and (6) can be employed in an exploratory, model-building fashion. We summarize and elaborate further on the key arguments underlying the exchange, drawing from the broader methodological and statistical literature in order to offer additional thoughts concerning the utility of PLS-PM and ways in which the technique might be improved. We conclude with recommendations as to whether and how PLS-PM serves as a viable contender to SEM approaches for estimating and evaluating theoretical models.

show abstract

Section: Concerning Henseler Et Al's Claim That Pls-pm Is An Exploramentioning

confidence: 99%

Reflections on Partial Least Squares Path Modeling

McIntosh

Edwards

Antonakis

2014

Organizational Research Methods

181

125

View full text Add to dashboard Cite

show abstract

“…No obstante, ninguna propuesta concreta fue explicitada por Cloninger (2008), como han observado también otros autores (Farmer y Goldberg, 2008). Se dispone ya de modelos de análisis no-lineales, que podrían ser de utilidad en el contraste de estas hipótesis (Buckler y Hennig-Thurau, 2008), algo exigible desde el rigor científico.…”

Section: Discussionunclassified

Fiabilidad y validez factorial del TCI-R en una muestra de adictos en tratamiento

Pérez

2013

analesps

View full text Add to dashboard Cite

Resumen: El TCI-R es el más reciente instrumento de medida de la teoría biosocial de Cloninger y se ha mostrado como uno de los cuestionarios más útiles en la predicción de la conducta, especialmente en sus variantes patológicas. A pesar de que la versión española fue publicada hace siete años apenas contamos con estudios que avalen su bondad psicométrica. Se requiere conocer la fiabilidad y validez interna para su aplicación clínica. Se reclutó una muestra de 824 participantes que inician tratamiento por su adicción en un centro específico. Se realizaron pruebas de consistencia interna sobre los ítems, las subescalas y las dimensiones principales del cuestionario y se estudió la validez de constructo mediante análisis factoriales exploratorio multivariado, paralelo y confirmatorio. Los resultados evidenciaron diversas inconsistencias en los diferentes niveles de análisis, si bien la tendencia principal apunta a la confirmación de la estructura teórica. Teniendo en cuenta la validez predictiva mostrada por el TCI-R, se hace necesario un refinamiento que reafirme su valor clínico. Quizá, como sugiere el propio Cloninger, sea necesario utilizar metodologías no-lineales, que capten las relaciones dinámicas entre los sustratos neurológicos que subyacen a los rasgos y un concepto menos estático de los componentes de la personalidad. Palabras clave: Personalidad; temperamento; carácter; adicción; análisis factorial multivariado; validez de constructo; TCI-R; Cloninger.Title: Reliability and factorial validity of the TCI-R in a sample of addicts in treatment Abstract: The TCI-R is the latest instrument to measure Cloninger biosocial theory and it has proven to be one of the most useful questionnaires in predicting behavior, especially in its pathological variants. Although the Spanish version was published seven years ago, there are few studies that support their psychometric goodness. It requires knowledge of the reliability and internal validity for clinical application. A sample of 824 subjects who started treatment for his addiction in a specialized center was recruited. Tests were performed to explore internal consistency of items, facets and main dimensions of the questionnaire, and construct validity studies using multivariate exploratory, parallel and confirmatory factor analysis. The results showed several inconsistencies in the different levels of analysis, although the main trend points to the confirmation of the theoretical structure. Given the predictive validity shown by the TCI-R, a refinement is necessary to reaffirm its clinical value. Perhaps, as suggested by Cloninger own, it will be necessary to use nonlinear methods, which capture the dynamic relationship between the neurological substrates underlying features and a less static concept of the components of personality. Key words: Personality; temperament; character; addiction; multivariate factor analysis; construct validity; TCI-R; Cloninger. Cloninger (1986Cloninger ( , 1987 propuso su modelo biosocial de la personalidad, sustentado en la distinc...

show abstract

“…Buckler and Hennig-Thurau [2] introduce a new innovative tool that can overcome limitations associated with the two traditional types of SEM: covariance-based structural equation modeling (CVSEM) and component-based partial least square (PLS). This tool has been referred to as universal structure modeling (USM) and defined as "a method that enables researchers to apply such an exploratory approach to SEM and thus helps them identify different kinds of "hidden" structures instead of testing a limited set of rival model structures.…”

Section: Universal Structure Modelingmentioning

confidence: 99%

“…USM has addressed the problem of "black-box" inherent to NN. While unlike CVSEM and PLS, USM can provide the following hidden aspects within a structural model [2]: Mathematically speaking, USM specifies the structural model with ŷj as the endogenous latent variable defined by functions of one or more other latent variables y that can be exogenous or endogenous. Formally, ŷj is estimated through y j and defined as the output of a multilayer perceptron (MLP) architecture as the below equation shows:…”

Section: Universal Structure Modelingmentioning

confidence: 99%