PLS path modeling – a confirmatory approach to study tourism technology and tourist behavior

Abstract

Purpose-As technology in tourism and hospitality (TTH) develops technical artifacts according to visitors' demands, it must deal with both behavioral and design constructs in the context of structural equation modeling (SEM). While behavioral constructs are typically modeled as common factors, the study at hand introduces the composite into TTH to model artifacts. To deal with both kinds of constructs, this paper aims to exploit partial least squares path modeling (PLS-PM) as a confirmatory approach to estimate models containing common factors and composites. Design/methodology/approach-The study at hand presents PLS-PM in its current form, i.e. as a fullfledged approach for confirmatory purposes. By introducing the composite to model artifacts, TTH scholars can use PLS-PM to answer research questions of the type "Is artifact xyz useful?", contributing to a further understanding of TTH. To demonstrate the composite model, an empirical example is used. Findings-PLS-PM is a promising approach when the model contains both common factors and composites. By applying the test for overall model fit, empirical evidence can be obtained for latent variables and artifacts. In doing so, researchers can statistically test whether a developed artifact is useful. Originality/value-To the best of the authors' knowledge, this is the first study to discuss the practical application of composite and common factor models in TTH research. Besides introducing the composite to model artifacts, the study at hand also guides scholars in the assessment of PLS-PM results.

The impact of creating shared value on hotels online reputation

“…PLS‐PM is based on an iterative algorithm to obtain weights used for building linear combinations of observed indicators as proxies for all constructs in the model. Thus, PLS‐SEM is an effective method to estimate composite models (Benítez, Henseler, & Castillo, 2017; Müller, Schuberth, & Henseler, 2018) since composites scores are well determined with traditional PLS‐PM (Cepeda‐Carrión, Cegarra‐Navarro, & Cillo, 2019). Further advantages over covariance‐based SEM models include lesser assumptions for the measurement scales, sample size, and data distribution (Chin, 1998).…”

“…Moreover, PLS‐SEM is considered an appropriate software tool for the analysis since it provides fit indexes and also allows for researchers to operationalize models with small or reduced sample size (Henseler, 2018; Henseler, Hubona, & Ray, 2016; Henseler, Ringle, & Sarstedt, 2016). In addition, we specified a “mode A” weighting scheme step to avoid possible collinearity issues due to a high level of correlation between indicators (Becker, Rai, & Rigdon, 2013; Henseler, Hubona, & Ray, 2016; Müller et al, 2018; Rigdon, 2016). We afterwards verified the reliability and validity of the lower‐order‐outer models as specified by Hair, Hult, Ringle, and Sarstedt (2017).…”

El papel de la Cooperación para Desarrollar Innovación Tecnológica en la PYME

“…Siguiendo las recomendaciones de (Müller, Schuberth, & Henseler, 2018) se debe establecer la confiabilidad de los indicadores y constructos, la validez convergente y validez discriminante. La carga (λ) de cada elemento en su construcción debe ser superior a 0,707 para verificar la confiabilidad del indicador (Hair et al, 2017), aún cuando tres elementos se encuentran por debajo de 0,707 estos fueron retenidos según lo sugerido por (Martelo-Landroguez, Cegarra Navarro, & Cepeda-Carrión, 2019).…”

Brand Worlds in Industrial Marketing

“…To assess composite measurement models, researchers should rely on nomological validity, reliability, and the assessment of its composition by investigating the weights of the indicators for possible multicollinearity issues (Henseler, 2017a). Nomological validity can be assessed by means of confirmatory composite analysis (Henseler et al, 2014;Müller, Schuberth, & Henseler, 2018;Schuberth, Henseler, & Dijkstra, 2018). In essence, this statistical technique tests whether the overall model fit of a model that includes the composite in a nomological net as an entity composed of its indicators exhibits a significantly worse fit than a model in which the composite is not included, and where its indicators act within the nomological net without the proportionality constraints imposed on them by the composite (Henseler, 2017a;Henseler et al, 2014;Müller et al, 2018;Schuberth et al, 2018).…”

“…Nomological validity can be assessed by means of confirmatory composite analysis (Henseler et al, 2014;Müller, Schuberth, & Henseler, 2018;Schuberth, Henseler, & Dijkstra, 2018). In essence, this statistical technique tests whether the overall model fit of a model that includes the composite in a nomological net as an entity composed of its indicators exhibits a significantly worse fit than a model in which the composite is not included, and where its indicators act within the nomological net without the proportionality constraints imposed on them by the composite (Henseler, 2017a;Henseler et al, 2014;Müller et al, 2018;Schuberth et al, 2018). If the model with the composite does not have a significantly worse fit, researchers can rely on Ockham's razor and conclude that the composite has nomological validity, that it is the composite that acts within the nomological net, rather than its individual indicators, and that it thus makes sense to create the composite (Henseler, 2017a;Henseler et al, 2016).…”