Wynne W. Chin

A Partial Least Squares Latent Variable Modeling Approach for Measuring Interaction Effects: Results from a Monte Carlo Simulation Study and an Electronic-Mail Emotion/Adoption Study

The ability to detect and accurately estimate the strength of interaction effects are critical issues that are fundamental to social science research in general and IS research in particular. Within the IS discipline, a significant percentage of research has been devoted to examining the conditions and contexts under which relationships may vary, often under the general umbrella of contingency theory (cf. McKeen et al. 1994, Weill and Olson 1989). In our survey of such studies, the majority failed to either detect or provide an estimate of the effect size. In cases where effect sizes are estimated, the numbers are generally small. These results have led some researchers to question both the usefulness of contingency theory and the need to detect interaction effects (e.g., Weill and Olson 1989). This paper addresses this issue by providing a new latent variable modeling approach that can give more accurate estimates of interaction effects by accounting for the measurement error that attenuates the estimated relationships. The capacity of this approach at recovering true effects in comparison to summated regression is demonstrated in a Monte Carlo study that creates a simulated data set in which the underlying true effects are known. Analysis of a second, empirical data set is included to demonstrate the techniques use within IS theory. In this second analysis, substantial direct and interaction effects of enjoyment on electronic-mail adoption are shown to exist.

How to Write Up and Report PLS Analyses

The objective of this paper is to provide a basic framework for researchers interested in reporting the results of their PLS analyses. Since the dominant paradigm in reporting Structural Equation Modeling results is covariance based, this paper begins by providing a discussion of key differences and rationale that researchers can use to support their use of PLS. This is followed by two examples from the discipline of Information Systems. The first consists of constructs with reflective indicators (mode A). This is followed up with a model that includes a construct with formative indicators (mode B).

Extending the technology acceptance model: the influence of perceived user resources

There has been considerable research on the factors that predict whether individuals will accept and voluntarily use information systems. The technology acceptance model (TAM) has a base in psychological research, is parsimonious, explains usage behavior quite well, and can be operationalized with valid and reliable instruments. A limitation of TAM is that it assumes usage is volitional, that is, there are no barriers that would prevent an individual from using an IS if he or she chose to do so. This research extends TAM by adding perceived user resources to the model, with careful attention to placing the construct in TAMs existing nomological structure. In contrast to measures of self-efficacy and perceived behavioral control that concentrate on how well individuals perceive they can execute specific courses of action, this paper examines perceptions of adequate resources that can facilitate or inhibit such behaviors. The inclusion of both a formative and reflective set of measures provides the opportunity for the researcher and manager to decide whether to evaluate only the overall perceptions of adequate resources or also the specific underlying causes. The extended model incorporating these measures was then tested in the field. The results confirmed that perceived user resources is a valuable addition to the model.

On the use, usefulness, and ease of use of structural equation modeling in MIS research: a note of caution

Structural equation modeling (SEM) using LISREL, EOS, PLS, or other second genration data analysis techniques is increasingly being applied in MIS research. These techniques are important because they provide powerful ways to address key IS research problems such as understanding IT usage. However, they may lead to inappropriate conclusions if statistical criteria are permitted to drive analysis and override substantive understanding of a problem. The purpose of this note is to suggest the need for caution in the application of structural equation modeling and, in particular, to emphasize the need for substantive knowledge to drive modeling, exploration, and interpretation of results. The application of SEM in the absence of well-developed substantive knowledge can lead to equivocal results and may distract researchers from promising research paths.

Adoption intention in GSS: relative importance of beliefs

Models used to study information technology (IT) adoption were adapted and applied to the group support systems (GSS) domain to determine their applicability and to assess the relative importance of beliefs about GSS in the context of adoption. The beliefs examined were relative advantage, ease of use, compatibility, and enjoyment. Four methods of measuring the relative importance of the belief constructs in predicting GSS adoption intention were evaluated. The methods involved Likert-scaled measures, conjoint measures, direct ratings of importance, and paired comparisons. Compositional analyses using three distinct models, regression, molar, and molecular, were conducted to examine the data collected under the first method. The results of the study demonstrated that the models and constructs used in other IT domains are indeed applicable in the GSS context. The relative importance of the beliefs did not converge across methods, although there was some agreement among the three compositional approaches as well as among the conjoint, direct rating, and paired comparison methods. The results point to the need for caution in utilizing and interpreting regression results and demonstrate the viability of alternative methods of studying adoption.

Handbook of Partial Least Squares

The new volume of Computational Statistics represents a comprehensive overview of Partial Least Squares (PLS) methods with specific reference to their use in marketing and with a discussion of the directions of current research and perspectives. The handbook covers the broad area of PLS methods -from regression to structural equation modeling applications, software and interpretation of results. It features papers on the use and the analysis of latent variables and indicators by means of the PLS path modeling approach from the design of the causal network to model assessment and improvement. Within the PLS framework, the handbook also addresses advanced topics such as the analysis of multi-block, multi-group and multi-structured data, the use of categorical indicators, the study of interaction effects, the integration of classification issues, the validation aspects and the comparison between the PLS approach and covariance based structural equation modeling. Most chapters comprise a thorough discussion of applications to marketing and related areas, some tutorials focus on key aspects of PLS analysis with a didactic approach. This handbook serves both as an introduction for those without prior knowledge of PLS and as a comprehensive reference for researchers and practitioners interested in the most recent advances in PLS methodology.

Advancing the Theory of Adaptive Structuration: The Development of a Scale to Measure Faithfulness of Appropriation

Adaptive Structuration Theory AST is rapidly becoming an influential theoretical perspective in research on advanced information technologies. However, there still exists a paucity of methods to capture critical AST constructs. This paper describes the development of an instrument to capture the extent to which users of an advanced information technology believe they have appropriated its structures faithfully. The development of such instruments is considered critical if the theoretical base provided by AST is to be fully exploited in understanding the use of advanced information technologies. The development procedure, which occurred in the context of the use of an electronic meeting system, was carried out in three phases that began with initial item development and proceeded through an exploratory to a confirmatory phase. Three experiments, two in the exploratory phase and one in the confirmatory phase, were performed. In the final phase, structural equation modeling techniques were used to confirm the convergent, discriminant, and nomological validity of the resulting five-item scale.

A Comparison of Approaches for the Analysis of Interaction Effects Between Latent Variables Using Partial Least Squares Path Modeling

In social and business sciences, the importance of the analysis of interaction effects between manifest as well as latent variables steadily increases. Researchers using partial least squares (PLS) to analyze interaction effects between latent variables need an overview of the available approaches as well as their suitability. This article presents 4 PLS-based approaches: a product indicator approach (Chin, Marcolin, & Newsted, 2003), a 2-stage approach (Chin et al., 2003; Henseler & Fassott, in press), a hybrid approach (Wold, 1982), and an orthogonalizing approach (Little, Bovaird, & Widaman, 2006), and contrasts them using data related to a technology acceptance model. By means of a more extensive Monte Carlo experiment, the different approaches are compared in terms of their point estimate accuracy, their statistical power, and their prediction accuracy. Based on the results of the experiment, the use of the orthogonalizing approach is recommendable under most circumstances. Only if the orthogonalizing approach does not find a significant interaction effect, the 2-stage approach should be additionally used for significance test, because it has a higher statistical power. For prediction accuracy, the orthogonalizing and the product indicator approach provide a significantly and substantially more accurate prediction than the other two approaches. Among these two, the orthogonalizing approach should be used in case of small sample size and few indicators per construct. If the sample size or the number of indicators per construct is medium to large, the product indicator approach should be used.

Applying adaptive structuration theory to investigate the process of group support systems use

Adaptive structuration theory (AST) provides a sound basis from which to study the use of group support systems (GSS). The need for a theoretical approach such as AST is especially urgent because i...

Information systems outsourcing relationship factors: detailed conceptualization and initial evidence

Outsourcing is now a commonly accepted means of meeting organizational IS needs. One facet of recent outsourcing research has begun to concentrate on the relationship between customers and their service providers. However, this line of research is hampered by a lack of consistent conceptual focus, and a corresponding inconsistent treatment of constructs associated with the relationship. This paper uses Relational Exchange Theory as a conceptual basis for identifying the individual constructs that comprise a relationship. It then develops and tests a set of items to measure those constructs, using structural equation modeling and survey data collected from U.S. outsourcing customers.