Search

With the increasing prominence of partial least squares structural equation modeling (PLS-SEM) in business research, the use of latent class analyses for identifying and treating unobserved heterogeneity has also gained momentum. Researchers have introduced various latent class approaches in a PLS-SEM context, of which finite mixture PLS (FIMIX-PLS) plays a central role due to its ability to identify heterogeneity and indicate a suitable number of segments to extract from the data. However, applying FIMIX-PLS requires researchers to make several choices that, if incorrect, could lead to wrong results and false conclusions. Addressing this concern, we present the results of a systematic review of FIMIX-PLS applications published in major business research journals. Our review provides an overview of the interdependencies between researchers’ choices and identifies potential problem areas. Based on our results, we offer concrete guidance on how to prevent common pitfalls when using FIMIX-PLS, and identify future research areas.

Evaluations of convergent and discriminant validity are generally conducted by analyzing constructs in isolation or by comparing pairs of latent variables. These approaches ignore the broader nomological network that is intrinsic to a measure’s construct validity, and fail to test the implications of either perfect correlations (convergence) or imperfect correlations (divergence). This paper proposes congruence assessment as a useful approach to simultaneously examining the relationships between multiple latent variables within nomological networks. Two measures are congruent if they have proportionally equal correlations with other constructs. We present measures for quantifying congruence within nomological networks, discuss statistical tests of significance, and demonstrate their performance in simulation studies. We reanalyze three published studies to contrast findings from congruence assessment versus traditional criteria for convergent and discriminant validity. We also discuss methodological and theoretical implications of congruence assessment, and suggest future research directions for both covariance- and composite-based structural equation modeling.

Purpose This study aims to propose guidelines for the joint use of partial least squares structural equation modeling (PLS-SEM) and fuzzy-set qualitative comparative analysis (fsQCA) to combine symmetric and asymmetric perspectives in model evaluation, in the hospitality and tourism field. Design/methodology/approach This study discusses PLS-SEM as a symmetric approach and fsQCA as an asymmetric approach to analyze structural and configurational models. It presents guidelines to conduct an fsQCA based on latent construct scores drawn from PLS-SEM, to assess how configurations of exogenous constructs produce a specific outcome in an endogenous construct. Findings This research highlights the advantages of combining PLS-SEM and fsQCA to analyze the causal effects of antecedents (i.e., exogenous constructs) on outcomes (i.e., endogenous constructs). The construct scores extracted from the PLS-SEM analysis of a nomological network of constructs provide accurate input for performing fsQCA to identify the sufficient configurations required to predict the outcome(s). Complementing the assessment of the model’s explanatory and predictive power, the fsQCA generates more fine-grained insights into variable relationships, thereby offering the means to reach better managerial conclusions. Originality/value The application of PLS-SEM and fsQCA as separate prediction-oriented methods has increased notably in recent years. However, in the absence of clear guidelines, studies applied the methods inconsistently, giving researchers little direction on how to best apply PLS-SEM and fsQCA in tandem. To address this concern, this study provides guidelines for the joint use of PLS-SEM and fsQCA.

Most project management research focuses almost exclusively on explanatory analyses. Evaluation of the explanatory power of statistical models is generally based on F-type statistics and the R 2 metric, followed by an assessment of the model parameters (e.g., beta coefficients) in terms of their significance, size, and direction. However, these measures are not indicative of a model’s predictive power, which is central for deriving managerial recommendations. We recommend that project management researchers routinely use additional metrics, such as the mean absolute error or the root mean square error, to accurately quantify their statistical models’ predictive power.

PurposeThis research introduces the combined use of partial least squares–structural equation modeling (PLS-SEM) and necessary condition analysis (NCA) that enables researchers to explore and validate hypotheses following a sufficiency logic, as well as hypotheses drawing on a necessity logic. The authors’ objective is to encourage the practice of combining PLS-SEM and NCA as complementary views of causality and data analysis.Design/methodology/approachThe authors present guidelines describing how to combine PLS-SEM and NCA. These relate to the specification of the research objective and the theoretical background, the preparation and evaluation of the data set, running the analyses, the evaluation of measurements, the evaluation of the (structural) model and relationships and the interpretation of findings. In addition, the authors present an empirical illustration in the field of technology acceptance.FindingsThe use of PLS-SEM and NCA enables researchers to identify the must-have factors required for an outcome in accordance with the necessity logic. At the same time, this approach shows the should-have factors following the additive sufficiency logic. The combination of both logics enables researchers to support their theoretical considerations and offers new avenues to test theoretical alternatives for established models.Originality/valueThe authors provide insights into the logic, assessment, challenges and benefits of NCA for researchers familiar with PLS-SEM. This novel approach enables researchers to substantiate and improve their theories and helps practitioners disclose the must-have and should-have factors relevant to their decision-making.

Examining the efficacy of composite-based structural equation modeling (SEM) features prominently in research. However, studies analyzing the efficacy of corresponding estimators usually rely on factor model data. Thereby, they assess and analyze their performance on erroneous grounds (i.e., factor model data instead of composite model data). A potential reason for this malpractice lies in the lack of available composite model-based data generation procedures for prespecified model parameters in the structural model and the measurements models. Addressing this gap in research, we derive model formulations and present a composite model-based data generation approach. The findings will assist researchers in their composite-based SEM simulation studies.

Mediation and conditional process analyses have become popular approaches for examining the mechanisms by which effects operate and the factors that influence them. To estimate mediation models, researchers often augment their structural equation modeling (SEM) analyses with additional regression analyses using the PROCESS macro. This duality is surprising considering that research has long acknowledged the limitations of regression analyses when estimating models with latent variables. In this article, we argue that much of the confusion regarding SEM’s efficacy for mediation analyses results from a singular focus on factor-based methods, and there is no need for a tandem use of SEM and PROCESS. Specifically, we highlight that composite-based SEM methods overcome the limitations of both regression and factor-based SEM analyses when estimating even highly complex mediation models. We further conclude that composite-based SEM methods such as partial least squares (PLS-SEM) are the preferred and superior approach when estimating mediation and conditional process models, and that the PROCESS approach is not needed when mediation is examined with PLS-SEM.

Structural equation modeling (SEM) is a widely applied and useful tool for project management scholars. In this Thoughtlet article, we critically reflect on the measurement philosophy underlying the two streams of SEM and their adequacy for estimating relationships among concepts commonly encountered in the field (e.g., team performance). We also discuss considerations to ponder when making the choice between the two types of SEM as well as between SEM and regression analysis.

Structural equation modelling (SEM) has evolved into two domains, factor-based and component-based, dependent on whether constructs are statistically represented as common factors or components. The two SEM domains are conceptually distinct, each assuming their own population models with either of the statistical construct proxies, and statistical SEM approaches should be used for estimating models whose construct representations correspond to what they assume. However, SEM approaches have often been evaluated and compared only under population factor models, providing misleading conclusions about their relative performance. This is partly because population component models and their relationships have not been clearly formulated. Also, it is of fundamental importance to examine how robust SEM approaches can be to potential misrepresentation of constructs because researchers may often lack clear theories to determine whether a factor or component is more representative of a given construct. Addressing these issues, this study begins by clarifying several population component models and their relationships and then provides a comprehensive evaluation of four SEM approaches - the maximum likelihood approach and factor score regression for factor-based SEM as well as generalized structured component analysis (GSCA) and partial least squares path modelling (PLSPM) for component-based SEM - under various experimental conditions. We confirm that the factor-based SEM approaches should be preferred for estimating factor models, whereas the component-based SEM approaches should be chosen for component models. Importantly, the component-based approaches are generally more robust to construct misrepresentation than the factor-based ones. Of the component-based approaches, GSCA should be chosen over PLSPM, regardless of whether or not constructs are misrepresented.

While nonlinear relationships play an important role in explaining distinct family business behaviors and outcomes, researchers rarely consider them in their theoretical and statistical models. To address this concern, this article introduces partial least squares structural equation modeling (PLS-SEM) as a suitable means for estimating nonlinear effects in latent variable models and describes its advantages vis-a-vis multiple (sum scores) regression. We conceptually compare and empirically illustrate the two methods by means of a family business research model. Based on our discussions, we provide family business researchers with a checklist of best practice recommendations when applying PLS-SEM. The article adds new methodological instruments to the family business researchers’ toolbox that enable them to explain and explore the mutual and often nonlinear interactions between family and business. Thereby, this research contributes to more rigorous and meaningful family business science.

Purpose Partial least squares (PLS) has been introduced as a “causal-predictive” approach to structural equation modeling (SEM), designed to overcome the apparent dichotomy between explanation and prediction. However, while researchers using PLS-SEM routinely stress the predictive nature of their analyses, model evaluation assessment relies exclusively on metrics designed to assess the path model’s explanatory power. Recent research has proposed PLSpredict, a holdout sample-based procedure that generates case-level predictions on an item or a construct level. This paper offers guidelines for applying PLSpredict and explains the key choices researchers need to make using the procedure. Design/methodology/approach The authors discuss the need for prediction-oriented model evaluations in PLS-SEM and conceptually explain and further advance the PLSpredict method. In addition, they illustrate the PLSpredict procedure’s use with a tourism marketing model and provide recommendations on how the results should be interpreted. While the focus of the paper is on the PLSpredict procedure, the overarching aim is to encourage the routine prediction-oriented assessment in PLS-SEM analyses. Findings The paper advances PLSpredict and offers guidance on how to use this prediction-oriented model evaluation approach. Researchers should routinely consider the assessment of the predictive power of their PLS path models. PLSpredict is a useful and straightforward approach to evaluate the out-of-sample predictive capabilities of PLS path models that researchers can apply in their studies. Research limitations/implications Future research should seek to extend PLSpredict’s capabilities, for example, by developing more benchmarks for comparing PLS-SEM results and empirically contrasting the earliest antecedent and the direct antecedent approaches to predictive power assessment. Practical implications This paper offers clear guidelines for using PLSpredict, which researchers and practitioners should routinely apply as part of their PLS-SEM analyses. Originality/value This research substantiates the use of PLSpredict. It provides marketing researchers and practitioners with the knowledge they need to properly assess, report and interpret PLS-SEM results. Thereby, this research contributes to safeguarding the rigor of marketing studies using PLS-SEM.

In their effort to better understand consumer behavior, marketing researchers often analyze relationships between latent variables, measured by sets of observed variables. Partial least squares structural equation modeling (PLS-SEM) has become a popular tool for analyzing such relationships. Particularly the availability of SmartPLS, a comprehensive software program with an intuitive graphical user interface, helped popularize the method. We review the latest version of SmartPLS and discuss its various features. Our aim is to offer researchers with concrete guidance regarding their choice of a PLS-SEM software that fits their analytical needs.

Purpose The purpose of this paper is to explore the knowledge infrastructure of methodological research on partial least squares structural equation modeling (PLS-SEM) from a network point of view. The analysis involves the structures of authors, institutions, countries and co-citation networks, and discloses trending developments in the field. Design/methodology/approach Based on bibliometric data downloaded from the Web of Science, the authors apply various social network analysis (SNA) and visualization tools to examine the structure of knowledge networks of the PLS-SEM domain. Specifically, the authors investigate the PLS-SEM knowledge network by analyzing 84 methodological studies published in 39 journals by 145 authors from 106 institutions. Findings The analysis reveals that specific authors dominate the network, whereas most authors work in isolated groups, loosely connected to the network’s focal authors. Besides presenting the results of a country level analysis, the research also identifies journals that play a key role in disseminating knowledge in the network. Finally, a burst detection analysis indicates that method comparisons and extensions, for example, to estimate common factor model data or to leverage PLS-SEM’s predictive capabilities, feature prominently in recent research. Originality/value Addressing the limitations of prior systematic literature reviews on the PLS-SEM method, this is the first study to apply SNA to reveal the interrelated structures and properties of PLS-SEM’s research domain.

Purpose The purpose of this paper is to review and extend recent simulation studies on discriminant validity measures, contrasting the use of cutoff values (i.e. heuristics) with inferential tests. Design/methodology/approach Based on a simulation study, which considers different construct correlations, sample sizes, numbers of indicators and loading patterns, the authors assess each criterion’s sensitivity to type I and type II errors. Findings The findings of the simulation study provide further evidence for the robustness of the heterotrait–monotrait (HTMT) ratio of correlations criterion as an estimator of disattenuated (perfectly reliable) correlations between constructs, whose performance parallels that of the standard constrained PHI approach. Furthermore, the authors identify situations in which both methods fail and suggest an alternative criterion. Originality/value Addressing the limitations of prior simulation studies, the authors use both directional comparisons (i.e. heuristics) and inferential tests to facilitate the comparison of the HTMT and PHI methods. Furthermore, the simulation considers criteria that have not been assessed in prior research.

Partial least squares path modeling (PLSPM) and generalized structural component analysis (GSCA) constitute composite-based structural equation modeling (SEM) methods, which have attracted considerable interest among methodological and applied researchers alike. Methodological extensions of PLSPM and GSCA have appeared at rapid pace, producing different research streams with different foci and understandings of the methods and their merits. Based on a theoretical comparison of PLSPM and GSCA in terms of model specification, parameter estimation, and results evaluation, we apply a text analytics technique to identify links between dominant topics in methodological research on both methods. We find that researchers have put effort on clearly distinguishing factor and composite models and their implications for the methods’ performance. We also identify an increasing interest in more complex model specifications such as mediating effects and higher-order models. The evidence of converging and diverging PLSPM and GSCA streams of research points out opportunities for advancing the evolution of composite-based SEM.

PurposePartial least squares structural equation modeling (PLS-SEM) is an important statistical technique in the toolbox of methods that researchers in marketing and other social sciences disciplines frequently use in their empirical analyses. The purpose of this paper is to shed light on several misconceptions that have emerged as a result of the proposed “new guidelines” for PLS-SEM. The authors discuss various aspects related to current debates on when or when not to use PLS-SEM, and which model evaluation metrics to apply. In addition, this paper summarizes several important methodological extensions of PLS-SEM researchers can use to improve the quality of their analyses, results and findings.Design/methodology/approachThe paper merges literature from various disciplines, including marketing, strategic management, information systems, accounting and statistics, to present a state-of-the-art review of PLS-SEM. Based on these findings, the paper offers a point of orientation on how to consider and apply these latest developments when executing or assessing PLS-SEM-based research.FindingsThis paper offers guidance regarding situations that favor the use of PLS-SEM and discusses the need to consider certain model evaluation metrics. It also summarizes how to deal with endogeneity in PLS-SEM, and critically comments on the recent proposal to adjust PLS-SEM estimates to mimic common factor models that are the foundation of covariance-based SEM. Finally, this paper opposes characterizing common concepts and practices of PLS-SEM as “out-of-date” without providing well-substantiated alternatives and solutions.Research limitations/implicationsThe paper paves the way for future discussions and suggests a way forward to reach consensus regarding situations that favor PLS-SEM use and its application.Practical implicationsThis paper offers guidance on how to consider the latest methodological developments when executing or assessing PLS-SEM-based research.Originality/valueThis paper complements recently proposed “new guidelines” with the aim of offering a counter perspective on some strong claims made in the latest literature on PLS-SEM. It also clarifies some misconceptions regarding the application of PLS-SEM.

Partial least squares structural equation modeling (PLS-SEM) has become a standard tool for analyzing complex inter-relationships between observed and latent variables in tourism and numerous other fields of scientific inquiry. Along with the recent surge in the method’s use, research has contributed several complementary methods for assessing the robustness of PLS-SEM results. Although these improvements are documented in extant literature, research on tourism has been slow to adopt the relevant complementary methods. This article illustrates the use of recent advances in PLS-SEM, designed to ensure structural model results’ robustness in terms of nonlinear effects, endogeneity, and unobserved heterogeneity in a PLS-SEM framework. Our overarching aim is to encourage the routine use of these complementary methods to increase methodological rigor in the field.

Management researchers often develop theories and policies that are forward-looking. The prospective outlook of predictive modeling, where a model predicts unseen or new data, can complement the retrospective nature of causal-explanatory modeling that dominates the field. Partial least squares (PLS) path modeling is an excellent tool for building theories that offer both explanation and prediction. A limitation of PLS, however, is the lack of a statistical test to assess whether a proposed or alternative theoretical model offers significantly better out-of-sample predictive power than a benchmark or an established model. Such an assessment of predictive power is essential for theory development and validation, and for selecting a model on which to base managerial and policy decisions. We introduce the cross-validated predictive ability test (CVPAT) to conduct a pairwise comparison of predictive power of competing models, and substantiate its performance via multiple Monte Carlo studies. We propose a stepwise predictive model comparison procedure to guide researchers, and demonstrate CVPAT's practical utility using the well-known American Customer Satisfaction Index (ACSI) model.

Structural equation modeling is a multivariate data analysis method for analyzing complex relationships among constructs and indicators. To estimate structural equation models, researchers generally draw on two methods: covariance-based SEM (CB-SEM) and partial least squares SEM (PLS-SEM). Whereas CB-SEM is primarily used to confirm theories, PLS represents a causal–predictive approach to SEM that emphasizes prediction in estimating models, whose structures are designed to provide causal explanations. PLS-SEM is also useful for confirming measurement models. This chapter offers a concise overview of PLS-SEM’s key characteristics and discusses the main differences compared to CB-SEM. The chapter also describes considerations when using PLS-SEM and highlights situations that favor its use compared to CB-SEM.

SEMinR is a software package developed for the R statistical environment (R Core Team, 2021). The package includes a user-friendly syntax for creating and estimating structural equation models using estimators such as partial least squares. In this chapter, we introduce the syntax to create, estimate, and report structural equation models using SEMinR. We demonstrate the four steps to specifying and estimating a structural equation model: (1) loading and cleaning the data, (2) specifying the measurement models, (3) specifying the structural model, and (4) estimating, bootstrapping, and summarizing the model. This chapter also describes how to export results and figures from R for professional, publication-quality reporting.

Generalized structured component analysis (GSCA) is a technically well-established approach to component-based structural equation modeling that allows for specifying and examining the relationships between observed variables and components thereof. GSCA provides overall fit indexes for model evaluation, including the goodness-of-fit index (GFI) and the standardized root mean square residual (SRMR). While these indexes have a solid standing in factor-based structural equation modeling, nothing is known about their performance in GSCA. Addressing this limitation, we present a simulation study’s results, which confirm that both GFI and SRMR indexes distinguish effectively between correct and misspecified models. Based on our findings, we propose rules-of-thumb cutoff criteria for each index in different sample sizes, which researchers could use to assess model fit in practice.

Structural equation modeling (SEM) is a statistical analytic framework that allows researchers to specify and test models with observed and latent (or unobservable) variables and their generally linear relationships. In the past decades, SEM has become a standard statistical analysis technique in behavioral, educational, psychological, and social science researchers’ repertoire. From a technical perspective, SEM was developed as a mixture of two statistical fields - path analysis and data reduction. Path analysis is used to specify and examine directional relationships between observed variables, whereas data reduction is applied to uncover (unobserved) low-dimensional representations of observed variables, which are referred to as latent variables. Since two different data reduction techniques (i.e., factor analysis and principal component analysis) were available to the statistical community, SEM also evolved into two domains - factor-based and component-based (e.g., Jöreskog and Wold 1982). In factor-based SEM, in which the psychometric or psychological measurement tradition has strongly influenced, a (common) factor represents a latent variable under the assumption that each latent variable exists as an entity independent of observed variables, but also serves as the sole source of the associations between the observed variables. Conversely, in component-based SEM, which is more in line with traditional multivariate statistics, a weighted composite or a component of observed variables represents a latent variable under the assumption that the latter is an aggregation (or a direct consequence) of observed variables.

Purpose Researchers often use partial least squares structural equation modeling (PLS-SEM) to estimate path models that include formatively specified constructs. Their validation requires running a redundancy analysis, which tests whether the formatively measured construct is highly correlated with an alternative measure of the same construct. Extending prior knowledge in the field, this paper aims to examine the conditions favoring the use of single vs multiple items to measure the criterion construct in redundancy analyses. Design/methodology/approach Merging the literatures from a variety of fields, such as management, marketing and psychometrics, we first provide a theoretical comparison of single-item and multi-item measurement and offer guidelines for designing and validating suitable single items. An empirical comparison in the context of hospitality management examines whether using a single item to measure the criterion variable yields sufficient degrees of convergent validity compared to using a multi-item measure. Findings The results of an empirical comparison in the context of hospitality management show that, when the sample size is small, a single item yields higher degrees of convergent validity than a reflective construct does. However, larger sample sizes favor the use of reflectively measured multi-item constructs, but the differences are marginal, thus supporting the use of a global single item in PLS-SEM-based redundancy analyses. Originality/value This study is the first to research the efficacy of single-item versus multi-item measures in PLS-SEM-based redundancy analyses. The results illustrate that a convergent validity assessment of formatively measured constructs can be implemented without triggering a pronounced increase in survey length.

Purpose Structural equation modeling (SEM) depicts one of the most salient research methods across a variety of disciplines, including hospitality management. While for many researchers, SEM is equivalent to carrying out covariance-based SEM, recent research advocates the use of partial least squares structural equation modeling (PLS-SEM) as an attractive alternative. We systematically examine how PLS-SEM has been applied in major hospitality research journals with the aim of providing important guidance and, if necessary, opportunities for realignment in future applications. As PLS-SEM in hospitality research is still in an early stage of development, critically examining its use holds considerable promise in order to counteract misapplications which otherwise might reinforce over time. Design/methodology/approach We reviewed all PLS-SEM studies published in six SSCI-indexed hospitality management journals between 2001 and 2015. Tying in with prior studies in the field, our review covers reasons for usin...

Partial least squares structural equation modeling (PLS-SEM) has become a key multivariate analysis technique that human resource management (HRM) researchers frequently use. While most disciplines undertake regular critical reflections on the use of important methods to ensure rigorous research and publication practices, the use of PLS-SEM in HRM has not been analyzed so far. To address this gap in HRM literature, this paper presents a critical review of PLS-SEM use in 77 HRM studies published over a 30-year period in leading journals. By contrasting the review results with state-of-the-art guidelines for use of the method, we identify several areas that offer room of improvement when applying PLS-SEM in HRM studies. Our findings offer important guidance for future use of the PLS-SEM method in HRM and related fields.

The use of partial least squares structural equation modeling (PLS-SEM) has been gaining momentum in family business research. Since the publication of a PLS-SEM guidelines article in the Journal of Family Business Strategy’s special issue on “Innovative and Established Research Methods in Family Business” in 2014, methodological research has developed new model evaluation methods and metrics and sharpened our understanding of the method’s strengths and limitations. In light of these developments, we extend prior guidelines on PLS-SEM applications by discussing new model evaluation procedures (e.g., model selection) and metrics (e.g., PLSpredict). In addition, we highlight the usefulness of methodological extensions for discrete choice modeling and endogeneity assessment that considerably extend the scope of the PLS-SEM method, and emerging opportunities for the application of PLS-SEM with archival (secondary) data. PLS-SEM remains a valuable method in the context of family business research, especially when it comes to gaining a more sophisticated understanding of the drivers of family business behavior. Because of its properties, the PLS-SEM approach proves particularly valuable when the aim is to predict target variables (e.g., family firm performance) in the context of a causal model.

Composite-based structural equation modeling (SEM), and especially partial least squares path modeling (PLS), has gained increasing dissemination in marketing. To fully exploit the potential of these methods, researchers must know about their relative performance and the settings that favor each method’s use. While numerous simulation studies have aimed to evaluate the performance of composite-based SEM methods, practically all of them defined populations using common factor models, thereby assessing the methods on erroneous grounds. This study is the first to offer a comprehensive assessment of composite-based SEM techniques on the basis of composite model data, considering a broad range of model constellations. Results of a large-scale simulation study substantiate that PLS and generalized structured component analysis are consistent estimators when the underlying population is composite model-based. While both methods outperform sum scores regression in terms of parameter recovery, PLS achieves slightly greater statistical power.

Researchers have long been aware of the mathematics of factor indeterminacy. Yet, while occasionally discussed, the phenomenon is mostly ignored. In metrology, the measurement discipline of the physical sciences, uncertainty - distinct from both random error (but encompassing it) and systematic error - is a crucial characteristic of any measurement. This research argues that factor indeterminacy is uncertainty. Factor indeterminacy fundamentally threatens the validity of psychometric measurement, because it blurs the linkage between a common factor and the conceptual variable that the factor represents. Acknowledging and quantifying factor indeterminacy is important for progress in reducing this component of uncertainty in measurement, and thus improving psychological measurement over time. Based on our elaborations, we offer a range of recommendations toward achieving this goal.

This study investigates how firms can achieve high levels of organizational performance under different configurations of absorptive capacity and organizational innovation. The study uses partial least squares structural equation modeling (PLS-SEM) and fuzzy-set qualitative comparative analysis (fsQCA) to test relationships among dimensions of absorptive capacity, organizational innovation, and organizational performance. The results provide support for the absorptive capacity's role for organizational innovation and performance. Furthermore, different configurations of absorptive capacity and organizational innovation conditions lead to better organizational performance.

Purpose The purpose of this paper is to introduce the importance-performance map analysis (IPMA) and explain how to use it in the context of partial least squares structural equation modeling (PLS-SEM). A case study, drawing on the IPMA module implemented in the SmartPLS 3 software, illustrates the results generation and interpretation. Design/methodology/approach The explications first address the principles of the IPMA and introduce a systematic procedure for its use, followed by a detailed discussion of each step. Finally, a case study on the use of technology shows how to apply the IPMA in empirical PLS-SEM studies. Findings The IPMA gives researchers the opportunity to enrich their PLS-SEM analysis and, thereby, gain additional results and findings. More specifically, instead of only analyzing the path coefficients (i.e. the importance dimension), the IPMA also considers the average value of the latent variables and their indicators (i.e. performance dimension). Research limitations/implications An IPMA is tied to certain requirements, which relate to the measurement scales, variable coding, and indicator weights estimates. Moreover, the IPMA presumes linear relationships. This research does not address the computation and interpretation of non-linear dependencies. Practical implications The IPMA is particularly useful for generating additional findings and conclusions by combining the analysis of the importance and performance dimensions in practical PLS-SEM applications. Thereby, the IPMA allows for prioritizing constructs to improve a certain target construct. Expanding the analysis to the indicator level facilitates identifying the most important areas of specific actions. These results are, for example, particularly important in practical studies identifying the differing impacts that certain construct dimensions have on phenomena such as technology acceptance, corporate reputation, or customer satisfaction. Originality/value This paper is the first to offer researchers a tutorial and annotated example of an IPMA. Based on a state-of-the-art review of the technique and a detailed explanation of the method, this paper introduces a systematic procedure for running an IPMA. A case study illustrates the analysis, using the SmartPLS 3 software.

Purpose Part I of this article (European Business Review, Volume 28, Issue 1) offered an overview of unobserved heterogeneity in the context of partial least squares structural equation modeling (PLS-SEM), its prevalence and challenges for social sciences researchers. This paper aims to provide an example that explains how to identify and treat unobserved heterogeneity in PLS-SEM by using the finite mixture PLS (FIMIX-PLS) module in the SmartPLS 3 software (Part II). Design/methodology/approach This case study illustrates the application of FIMIX-PLS using a popular corporate reputation model. Findings The case study demonstrates the capability of FIMIX-PLS to identify whether unobserved heterogeneity significantly affects structural model relationships. Furthermore, it shows that FIMIX-PLS is particularly useful for determining the number of segments to extract from the data. Research limitations/implications Since the introduction of FIMIX-PLS, a range of alternative latent class techniques has appeared. These techniques address some of the limitations of the approach relating to, for example, its failure to handle heterogeneity in measurement models, or its distributional assumptions. This research discusses alternative latent class techniques and calls for the joint use of FIMIX-PLS and PLS prediction-oriented segmentation. Originality/value This article is the first to offer researchers, who have not been exposed to the method, an introduction to FIMIX-PLS. Based on a state-of-the-art review of the technique, the paper offers a step-by-step tutorial on how to use FIMIX-PLS by using the SmartPLS 3 software.

Accounting for heterogeneity in tourism studies remains important to avoid parameter bias (e.g., Mazanec, 2000; Mazanec, Ring, Stangl, & Teichmann, 2010) when employing analysis techniques such as regression (e.g., Ye, Zhang, & Yuen, 2013), partial least squares structural equation modeling (PLS-SEM) (e.g., Song, van der Veen, Li, & Chen, 2012), or covariance structural equation modeling (CB-SEM) (e.g., Jurowski & Gursoy, 2004). Heterogeneity can come in two forms. First, heterogeneity can be observable in that differences between two or more groups of data relate to observable characteristics (e.g., Dolnicar, 2004). Researchers can use these observable characteristics to partition the data into separate groups of observations and compare the group-specific estimates by means of multigroup comparisons. Second, heterogeneity can be unobserved in that it does not depend on one specific observable characteristic or combinations of several characteristics (e.g., Mazanec, 2000, 2001). To identify and treat unobserved heterogeneity, researchers can draw on a variety of latent class techniques. For instance, Assaf, Oh, and Tsionas (2015) employ Bayesian finite mixture modeling within CB-SEM, and Marques and Reis (2015) finite mixture modeling within PLS-SEM. It is the latter approach that this commentary focuses on.

We first provide comprehensive and advanced access to principal component analysis, factor analysis, and reliability analysis. Based on a discussion of the different types of factor analytic procedures (exploratory factor analysis, confirmatory factor analysis, and structural equation modeling), we introduce the steps involved in a principal component analysis and a reliability analysis, offering guidelines for executing them in Stata. Specifically, we cover the requirements for running an analysis, modern options for extracting the factors and deciding on their number, as well as for interpreting and judging the quality of the results. Based on a step-by-step description of Stata’s menu options and code, we present an in-depth discussion of each element of the Stata output. Interpretation of output can be difficult, which we make much easier by means of various illustrations and applications, using a detailed case study to quickly make sense of the results. We conclude with suggestions for further readings on the use, application, and interpretation of factor analytic procedures.

Partial least squares structural equation modeling (PLS-SEM) has become a popular method for estimating path models with latent variables and their relationships. A common goal of PLS-SEM analyses is to identify key success factors and sources of competitive advantage for important target constructs such as customer satisfaction, customer loyalty, behavioral intentions, and user behavior. Building on an introduction of the fundamentals of measurement and structural theory, this chapter explains how to specify and estimate path models using PLS-SEM. Complementing the introduction of the PLS-SEM method and the description of how to evaluate analysis results, the chapter also offers an overview of complementary analytical techniques. A PLS-SEM application of the widely recognized corporate reputation model illustrates the method.

Accounting for unobserved heterogeneity has become a key concern to ensure the validity of results when applying partial least squares structural equation modeling (PLS-SEM). Recent methodological research in the field has brought forward a variety of latent class techniques that allow for identifying and treating unobserved heterogeneity. This chapter raises and discusses key aspects that are fundamental to a full and adequate understanding of how to apply these techniques in PLS-SEM. More precisely, in this chapter, we introduce a systematic procedure for identifying and treating unobserved heterogeneity in PLS path models using a combination of latent class techniques. The procedure builds on the FIMIX-PLS method to decide if unobserved heterogeneity has a critical impact on the results. Based on these outcomes, researchers should use more recently developed latent class methods, which have been shown to perform superior in recovering the segment-specific model estimates. After introducing these techniques, the chapter continues by discussing the means to identify explanatory variables that characterize the latent segments. Our discussion also broaches the issue of measurement invariance testing, which is a fundamental requirement for a subsequent comparison of parameters across groups by means of a multigroup analysis.

This article introduces and motivates an exchange of thoughts on the paper by Edward E. Rigdon in the first of two Long Range Planning special issues on partial least squares structural equation modeling (PLS-SEM) in strategic management published in 2012 and 2013. For 30 years, there has been a heated debate on the benefits and drawbacks of PLS-SEM versus those of its sibling, the covariance-based structural equation modeling (CB-SEM) approach. Edward E. Rigdon's paper is a milestone that proposes a change of thought and encourages the long-required emancipation of the PLS-SEM method from CB-SEM. These developments will have a pronounced impact on the proper application of SEM as a key multivariate analysis method in the strategic management discipline, further enhancing the potential it has as a research tool.

Purpose – The authors aim to present partial least squares (PLS) as an evolving approach to structural equation modeling (SEM), highlight its advantages and limitations and provide an overview of recent research on the method across various fields. Design/methodology/approach – In this review article, the authors merge literatures from the marketing, management, and management information systems fields to present the state-of-the art of PLS-SEM research. Furthermore, the authors meta-analyze recent review studies to shed light on popular reasons for PLS-SEM usage. Findings – PLS-SEM has experienced increasing dissemination in a variety of fields in recent years with nonnormal data, small sample sizes and the use of formative indicators being the most prominent reasons for its application. Recent methodological research has extended PLS-SEM's methodological toolbox to accommodate more complex model structures or handle data inadequacies such as heterogeneity. Research limitations/implications – While research on the PLS-SEM method has gained momentum during the last decade, there are ample research opportunities on subjects such as mediation or multigroup analysis, which warrant further attention. Originality/value – This article provides an introduction to PLS-SEM for researchers that have not yet been exposed to the method. The article is the first to meta-analyze reasons for PLS-SEM usage across the marketing, management, and management information systems fields. The cross-disciplinary review of recent research on the PLS-SEM method also makes this article useful for researchers interested in advanced concepts.

Structural equation modeling (SEM) has become a mainstream method in many fields of business research, but its use in family business research remains in its infancy. This lag in SEM's application holds especially true for partial least squares SEM (PLS-SEM), an alternative to covariance-based SEM, which provides researchers with more flexibility in terms of data requirements, model complexity and relationship specification. This article draws attention to PLS-SEM as an opportunity to advance the development and testing of theory in family business research by providing a non-technical introduction into the basic concepts and issues of PLS-SEM, bearing the needs of potential users in mind. To this end, a systematic procedure for PLS-SEM results evaluation is presented and applied to an annotated example. The article also illustrates the analysis of mediating effects, which researchers are increasingly testing in their models.

Over the past decade, the cultural intelligence construct and its underlying dimensions have been used in a number of studies. Prior research has tested the determinants and outcomes of cultural intelligence, using pooled data from different countries and cultures, and has compared the results across contexts. However, these studies often disregarded measurement invariance, which is a necessary requirement for such analyses. We assess the measurement invariance of the commonly used four-dimensional cultural intelligence scale across five countries (China, France, Germany, Turkey, and the U.S.) by means of a composite model logic, using partial least squares structural equation modeling (PLS-SEM). Our results question the scale's dimensionality concerning China and France, and reveal an item set that is invariant across the other countries. Our findings indicate that researchers should be aware of the potential lack of measurement invariance regarding the standard measurement of cultural intelligence. They should therefore be cautious when comparing the results of cross-country and cross-cultural research. (C) 2016 Elsevier Ltd. All rights reserved.

When applying the partial least squares structural equation modeling (PLS-SEM) method, the assumption that the data stem from a single homogeneous population is often unrealistic. For the full set of data, unobserved heterogeneity in the PLS path model estimates may result in misleading interpretations. This research presents the PLS genetic algorithm segmentation (PLS-GAS) method to account for unobserved heterogeneity in the path model estimates. The results of a simulation study guide an assessment of this novel approach. PLS-GAS allows for uncovering unobserved heterogeneity and identifying different groups within a data set. In an application on customer satisfaction data and the American customer satisfaction index model, the method identifies distinctive group-specific PLS path model estimates which allow for a further differentiated interpretation of the results.