The growing interest in Structured Equation Modeling (SEM) techniques and recognition of their importance in IS research suggests the need to compare and contrast different types of SEM techniques so that research designs can be selected appropriately. After assessing the extent to which these techniques are currently being used in IS research, the article presents a running example which analyzes the same dataset via three very different statistical techniques. It then compares two classes of SEM: covariance-based SEM and partial-least-squaresbased SEM. Finally, the article discusses linear regression models and offers guidelines as to when SEM techniques and when regression techniques should be used. The article concludes with heuristics and rule of thumb thresholds to guide practice, and a discussion of the extent to which practice is in accord with these guidelines.

/pdf/structural-equation-modeling-and-regression-guidelines-for-2y96bdeejv.pdf

Structural equation modeling and regression: guidelines for research practice

While the technology acceptance model (TAM), introduced in 1986, continues to be the most widely applied theoretical model in the IS field, few previous efforts examined its accomplishments and limitations. This study traces TAM’s history, investigates its findings, and cautiously predicts its future trajectory. One hundred and one articles published by leading IS journals and conferences in the past eighteen years are examined and summarized. An openended survey of thirty-two leading IS researchers assisted in critically examining TAM and specifying future directions.

/pdf/the-technology-acceptance-model-past-present-and-future-5dz0sdj5kx.pdf

The Technology Acceptance Model: Past, Present, and Future

The technology acceptance model (Davis 1989) is one of the most widely used models of IT adoption. According to TAM, IT adoption is influenced by two perceptions: usefulness and ease-of-use. Research has shown that perceived usefulness (PU) affects intended adoption of IT, but has mostly failed to do so regarding perceived ease of use (PEOU). The basic proposition of this study is that this varying importance of PEOU may be related to the nature of the task. PEOU relates to assessments of the intrinsic characteristics of IT, such as the ease of use, ease of learning, flexibility, and clarity of its interface. PU, on the other hand, is a response to user assessment of its extrinsic, i.e., task-oriented, outcomes: how IT

The Relative Importance of Perceived Ease of Use in IS Adoption: A Study of E-Commerce Adoption

The growing interest in Structured Equation Modeling (SEM) techniques and recognition of their importance in IS research raises the need to compare and contrast the different types of SEM so that research designs can be selected on the basis of appropriateness. After discussing a running example which analyzes the same dataset via three very different statistical techniques, the paper assesses the extent to which these techniques are currently being used in IS research. It then compares two classes of SEM, covariance-based SEM and partial-least-squares-based SEM. Finally, it discusses linear regression models and suggests guidelines as to when both SEM and regression techniques should be used. The paper concludes with heuristics and rule of thumb thresholds to guide practice, and a description of the extent to which practice is in accord with these guidelines. The paper is written in such a way that readers with basic knowledge of multivariate statistics can follow the logic and examples. It does not assume the reader is already conversant with LISREL, PLS, or other SEM tools. To address this audience, the paper contains straightforward examples to explain more complex topics, a glossary with clickable links, and a rudimentary structural model applying the Technology Acceptance Model (TAM) to e-Commerce which is analyzed via PLS, LISREL, and linear regression techniques.

Structural Equation Modeling Techniques and Regression: Guidelines for Research Practice

The value added by an organization’s IT assets is a critical concern to both research and practice. Not surprisingly, a large number of IS effectiveness measures can be found in the IS literature. What is not clear in the literature is what measures are appropriate in a particular context. In this paper we propose a two-dimensional matrix for classifying IS Effectiveness measures. The first dimension is the type of system studied. The second dimension is the stakeholder in whose interests the system is being evaluated. The matrix was tested by using it to classify IS effectiveness measures from 186 empirical papers in three major IS journals for the last nine years. The results indicate that the classifications are meaningful. Hence, the IS Effectiveness Matrix provides a useful guide for conceptualizing effectiveness measurement in IS research, and for choosing appropriate measures, both for research and practice.

/pdf/dimensions-of-information-systems-success-234iaxw9zt.pdf

Dimensions of information systems success

Evidence suggests that information centers (ICs) have significantly more interest in evaluating software packages and assisting in the selection of software packages than end users have. However, the selection of software packages by the information center product specialists (ICPSs) can compromise their usage. Ease of use and usefulness are believed to be fundamental predictors of usage. The question of whether ICPSs are able to correctly evaluate ease of use and usefulness of software packages for end users is posed. An insight into this issue could enhance end-user computing (EUC) policy and lead to a more effective partnership between end users and information systems (IS) professionals. A search for this insight provided the motivation for our empirical investigation of the perception of ICPSs and end users in assessing the ease of use and usefulness of thirty different software packages. Our investigation was performed in an organization with an IC that had evolved to the formalization stage. The findings show that the perceptions of the end users and ICPSs were similar in terms of assessing the ease of use of software packages. However, end users found the software packages less useful than did IC product specialists. Therefore, in sophisticated environments, end users should be empowered to develop their own user groups and suggest to IC personnel which useful software packages to acquire. Otherwise, selection of software packages without end-user participation could have an adverse effect on their usage.

An empirical study of factors affecting software package selection

An adaptive agent for case description in diagnostic CBR systems

Case-Based Reasoning (CBR) systems support ill-structured decision making. In ill-structured decision environments, decision makers (DMs) differ in their problem solving approaches. As a result, CBR systems would be more useful if they were able to adapt to the idiosyncrasies of individual decision makers. Existing implementations of CBR systems have been mainly symbolic, and symbolic CBR systems are unable to adapt to the preferences of decision makers (i.e., they are static). Retrieval of appropriate previous cases is critical to the success of a CBR system. Widely used symbolic retrieval functions, such as nearest-neighbor matching, assume independence of attributes and require specification of their importance for matching. To ameliorate these deficiencies connectionist systems have been proposed. However, these systems are limited in their ability to adapt and grow. To overcome this limitation, we propose a distributed connectionist-symbolic architecture that adapts to the preferences of a decision maker and that, additionally, ameliorates the limitations of symbolic matching. The proposed architecture uses a supervised learning technique to acquire the matching knowledge. The architecture allows the growth of a case base without the involvement of a knowledge engineer. Empirical investigation of the proposed architecture in an ill-structured diagnostic decision environment demonstrated a superior retrieval performance when compared to the nearest-neighbor matching function.

A connectionist approach for similarity assessment in case-based reasoning systems

A case-based reasoning (CBR) system supports decision makers when solving new decision problems (i.e., new cases) on the basis of past experience (i.e., previous cases). The effectiveness of a CBR system depends on its ability to retrieve useful previous cases. The usefulness of a previous case is determined by its similarity with the new case. Existing methodologies assess similarity by using a set of domain-specific production rules. However, production rules are brittle in ill-structured decision domains and their acquisition is complex and costly. We propose a framework of methodologies based on decision theory to assess the similarity of a new case with the previous case that allows amelioration of the deficiencies associated with the use of production rules. An empirical test of the framework in an ill-structured diagnostic decision environment shows that this framework significantly improves the retrieval performance of a CBR system.

A framework for retrieval in case-based reasoning systems

The objective of this study was to determine the impact of task information (TI) provided by an interface agent during the idea evaluation and integration step of the problem formulation stage of the problem solving process. The effectiveness assessment was based on solving diagnostic decision problems in the domain of complex industrial machinery. Ten domain experts participated in this study. Decision support was provided by a case-based reasoning system. Findings suggest that TI provided by the interface agent had no effect on the decision maker's performance, nor on the associated cognitive effort. However, a verbal protocol analysis revealed that the ten subjects used the interface agent to verify their decision processes. The results and their implications are discussed with respect to current findings in the area of decision support systems.

Kalyan Moy Gupta

Papers

An empirical study of factors affecting software package selection

An adaptive agent for case description in diagnostic CBR systems

A connectionist approach for similarity assessment in case-based reasoning systems

A framework for retrieval in case-based reasoning systems

On the effectiveness of cognitive feedback from an interface agent