01 guest intro.indd

5 downloads 0 Views 63KB Size Report
Gudergan, Jörg Henseler, Uta Herbst, Andreas Herrmann, Frank Huber, Geoffrey S. ... Edward E. Rigdon, Gaia Rubera, Matthias Schloderer, Philipp Schmitt, ...
From the Special Issue Guest Editors Dear Colleagues: Business and related social science researchers, particularly marketing researchers, generally agree that structural equation modeling (SEM) has become a dominant analytical tool in empirical research. In the past decade, marketing researchers have embraced SEM techniques to such an extent that it can be justifiably maintained that its use is ubiquitous. Almost every issue of major academic marketing journals reports on research employing SEM. But while such a sweeping statement is in fact accurate, it understates future even more widespread applications of SEM. The latter is because almost all SEM applications in marketing were, until recently, the result of applying the covariance-based (CB) approach to SEM and not the variance-based partial least squares (PLS)-SEM path modeling. Indeed, most empirical marketing researchers have little understanding of the two major types of SEM, or are even aware that there are two. CB-SEM’s statistical objective is to estimate a covariance matrix that matches that of the observed sample data as closely as possible. Hence, the focus is largely on achieving model “fit” assuming valid and reliable constructs. CB-SEM approaches largely ignore the prediction objective. Broadly speaking, empirical marketing research has two objectives: prediction and explanation. One can conclude that previous CB-SEM applications overlooked a major empirical marketing research objective, namely, prediction. The solution to this inherent weakness in previous structural modeling is the far less known PLS path modeling. In contrast to CB-SEM, PLS-SEM’s overriding objective is predicting the dependent (endogenous) variables (constructs). Compared to CB-SEM, PLS-SEM offers other advantages besides emphasizing prediction. Many empirical marketing researchers pay lip service to data characteristics such as heteroskedasticity and lack of normality, noting the robustness of the statistical techniques. In fact, most empirical marketing data is characterized by such inadequacies. Consequently, CB-SEM applications that use the maximum likelihood (ML) algorithm—which most do—overlook the inherent violations of this technique’s required assumptions. Because PLS-SEM does not require these restrictive distributional assumptions, it is often a more viable approach than CB-SEM. The articles included in this special issue address PLS-SEM path modeling’s strengths and limitations. In our introductory article, we provide an overview of the PLS-SEM path modeling approach directed at colleagues unfamiliar with the approach and who need to understand when PLS-SEM is the most appropriate method to address a particular research question. We then explain how to evaluate the results when PLS-SEM is applied. To facilitate future PLS-SEM applications, we provide rules of thumb on choosing PLS-SEM rather than CB-SEM, model specification, data characteristics, and model evaluation throughout the paper. The rules of thumb will be a convenient single source of guidelines for experienced PLS-SEM path modelers and an excellent overview of important considerations for those with limited knowledge of SEM procedures. In the second article, Michael Haenlein and Andreas M. Kaplan analyze the influence of observed heterogeneity on path coefficient significance. The authors employ the technology acceptance model (TAM), which is increasingly used within the marketing area. They show that one of three main relationships—the influence of perceived ease of use on the behavioral intention to adopt a system (PEOU–BI)—has received inconsistent empirical support in prior research. Against this background, their PLS path modeling analysis shows that observed population heterogeneity is a reasonable explanation for the PEOU–BI link’s inconsistent empirical support within the TAM. On

135

136  Journal of Marketing Theory and Practice the basis of these findings, the authors make a methodological contribution by proposing an approach with which to control for changes in a latent variable between two moderating conditions, when the changes are due to respondents perceiving the meaning of the construct as different in each of the moderating conditions (gamma change) as opposed to a true change in the latent variable (alpha change). In addition, their study highlights the importance of conducting a statistical power analysis to determine critical t‑values in the context of a PLS path analysis. The paper has important implications for future PLS path modeling applications, while the results can be used for accurate interpretations and appropriate conclusions. In the third article, Andreas Eggert and Murat Serdaroglu explore the impact of sales technology on salesperson performance. A task-based PLS path modeling analysis reveals that sales technology has a direct effect on salesperson performance. The internal coordination dimension is explained by factors imposed from outside, whereas the customer relationship dimension is driven by factors that trigger voluntary usage. The authors conclude that sales technology must be designed to enable customer relationships, while sales managers must support self-initiating factors that stimulate technology usage for improving customer relationships. The study provides important implications for both researchers and practitioners in the sales management domain. The fourth article by Antonio Navarro, Francisco J. Acedo, Fernando Losada, and Emilio Ruzo applies an integrated model of export activity to analyze heterogeneity in managers’ orientations toward and perceptions of strategic marketing management in foreign markets. The PLS path modeling results confirm that organizational learning acts as an antecedent of both export commitment and adaptation of the marketing mix to foreign markets. Moreover, the results of a finite mixture PLS (FIMIX-PLS) analysis provide evidence of the existence of two latent classes of exporters. The paper not only contributes to the literature by characterizing the two distinct types of exporters (i.e., proactive and conservative exporters), but also by illustrating the importance of addressing the problem of unobserved heterogeneity by means of response-based segmentation techniques such as FIMIX-PLS. Next, Klaus-Peter Wiedmann, Nadine Hennigs, Steffen Schmidt, and Thomas Wuestefeld analyze consumers’ perceptions of heritage brands in the automotive industry. They focus on the functions of the brand as perceived by consumers to reveal drivers and outcomes of brand heritage. Using an extensive model setup, the authors reveal brand heritage’s substantially significant effects on consumer attitudes and behaviors related to the given brand. The findings of their research provide a better understanding of brand heritage from the consumers’ point of view, which is highly valuable for both researchers and marketers. In the sixth article, Rolph E. Anderson and Srinivasan Swaminathan examine the factors that drive customer satisfaction and loyalty in e-markets. The PLS path modeling analysis results substantiate that six key factors (adaptation, commitment, network, assortment, transaction ease, and engagement) significantly affect satisfaction in e-businesses. They further confirm satisfaction’s positive effect on loyalty and that customer trust and inertia moderate this relationship. At lower levels of inertia and trust, customer satisfaction has a greater impact on loyalty. The authors conclude with actionable recommendations for e-business managers to enhance customer satisfaction based on the PLS path modeling results. Finally, Stefan Hoffmann, Robert Mai, and Maria Smirnova focus on the measurement of consumer animosity. They develop a multidimensional scale that distinguishes between three universal drivers of general animosity: perceived threat, antithetical political attitudes, and negative personal experiences. By means of PLS, two studies validate the scale by demonstrating (1) cross-national stability, (2) criterion validity (animosity influences product-specific country-of-origin images, boycotting, and purchase intention), and (3)  construct validity (animosity is embedded in a nomological network of ethnocentrism, patriotism, and cosmopolitanism). The paper provides researchers and practitioners with a clear-cut tool to analyze and address the critical issue of consumer animosity in international marketing.

Spring 2011  137

Conclusions Our objective with this special issue is to offer a set of guidelines to improve applications of both covariance and variance-based SEM in marketing and related business fields. More specifically, a better understanding is needed of the relationship between study objectives and data characteristics regarding SEM capabilities. Such an understanding will enable researchers to determine which SEM method is the best analytical approach to a particular problem. The articles in this special issue contribute significantly to the development of new knowledge and understanding of the emerging technique of PLS-SEM path modeling. They also clarify when and why PLS-SEM or CB-SEM is the most appropriate approach in more detail. Finally, the articles demonstrate the value of PLS-SEM for structural modeling in situations where traditional CB-SEM modeling is questionable. We are grateful to the reviewers who contributed their valuable time and talent to develop this special issue, and ensured the articles’ quality with their constructive comments and suggestions to the authors. Many of the reviewers were not regular members of the JMTP Editorial Review Board and therefore served as ad hoc reviewers. The reviewers for this special issue included Silke Bartsch, Jan Becker, Brad D. Carlson, Felix Eggers, Carsten Erfgen, Vincenzo Esposito Vinzi, Georg Fassott, Arne Floh, Claes Fornell, Karen Gedenk, Sonja Gensler, Oliver Götz, Yani Grégoire, Götz Greve, Siegried P. Gudergan, Jörg Henseler, Uta Herbst, Andreas Herrmann, Frank Huber, Geoffrey S. Hubona, Gary L. Hunter, Sebastian Kaiser, Felix Kessel, Orhan Kocyigit, Kai Kristensen, Graham R. Massey, Arthur H. Money, Christian Nitzl, Dominik Papies, Julie M. Pharr, Arun Rai, Nina L. Reynolds, Petra Riefler, Edward E. Rigdon, Gaia Rubera, Matthias Schloderer, Philipp Schmitt, Jochen Schweitzer, Christine Sichtmann, Rudolf R. Sinkovics, Matthias Söllner, Michael Steiner, Charles R. Taylor, Thorsten Teichert, Birgit Verworn, Stephan M. Wagner, Robert Wilken, and Lorenz Zimmermann. Thank you for your support! Joe F. Hair, Kennesaw State University Christian M. Ringle, Hamburg University of Technology (TUHH), Germany Marko Sarstedt, Ludwig-Maximilians-University, Germany

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.