How to perform and report an impactful analysis using partial least squares: Guidelines for confirmatory and explanatory IS research

Partial least squares path modeling (PLS-PM) is an estimator that has found widespread application for causal information systems (IS) research. Recently, the method has been subject to many improvements, such as consistent PLS (PLSc) for latent variable models, a bootstrap-based test for overall model fit, and the heterotrait-to-monotrait ratio of correlations for assessing discriminant validity. Scholars who would like to rigorously apply PLS-PM need updated guidelines for its use. This paper explains how to perform and report empirical analyses using PLS-PM including the latest enhancements, and illustrates its application with a fictive example on business value of social media.     

Effects of computer-based testing on test performance and testing motivation

Computer-based testing (CBT) is a green computing strategies used to reduce paper consumption. However, some scholars have questioned the effectiveness of CBT and suggested for conducting systematic studies on CBT to carefully check its reliability and validity before opting for it. Recently, some studies have evaluated the effectiveness of CBT by comparing it with paper–pencil testing (PPT), and the findings were inconsistent. Besides this, most of these studies have been conducted using quasi or basic experimental designs without identifying testing effects on test takers. The limitation of these design is testing effects might occur when a participant in the control or treatment group is tested at least twice on a same test, and it influences the outcomes of taking a posttest, therefore, the findings might be misinterpreted. This study employed a Solomon four-group experimental design (2 × 3 factorial design) on a group of student teachers (n = 140) to compare CBT and PPT on test performance (test scores), testing time and testing motivation. Results indicated that the CBT mode is more reliable in terms of internal and external validity. The CBT significantly reduced testing time and developed stronger self-efficacy, intrinsic and social testing motivation in the participants.     

The temporal nature of IS benefits: A longitudinal study

This paper discusses the dynamic nature of information technology (IT) benefits and reports on findings of a three-year evaluation of user productivity and organizational effectiveness following the installation of an integrated office information system (IS). For systems adopted on a voluntary basis, individual benefits occur first, while improvements in organizational effectiveness develop over a longer period of time. This suggests that the timing of IS assessment is important and that longitudinal research designs are more appropriate for IS benefits evaluation than cross-sectional studies.     

Information Systems Use: Construct conceptualization and scale development

Information Systems Use (ISU) is an essential part of the human behaviors in utilizing computers in organizations. The construct has been widely used to measure IS adoption or IS success. However, few studies attempt to understand ISU in a job and organizational setting where employees use various types of IS for different tasks. To better understand ISU, it is necessary to contextualize the construct in users’ overall work related activities. We classified the overall IS/IT use in an organization into three different types of IS; Information Reporting Systems (IRSs), Decision Support Systems (DSSs), and Group Support System (GSSs). Based on this classification, we developed four items for each type of ISU behaviors. The resulting ISU instrument was tested using a dataset of 231 responses collected in a survey. Both exploratory factor analysis and PLS are employed to successfully establish reliability, convergent/discriminate validity, and predictive validity. The contribution of this research is to provide better and more robust measurements for the ISU construct, which should help to lay a firmer foundation for further research on IS success.     

基于神经网络的混合数据的因果发现

因果推理正在成为机器学习领域一个越来越受关注的研究热点,现阶段的因果发现主要是在研究某一种假设条件下,基于纯粹的观测数据推断变量之间的因果方向。然而在现实世界中观察到的数据往往是由一些假设生成,使得传统因果推断方法的识别率不高、稳定性较差。针对当前的问题,提出了一种基于神经网络来解决混合数据因果推断的方法。该方法在混合加性噪声模型(ANM-MM)的假设下,使用梯度下降法最优化改进的损失函数得到混合数据的抽象因果分布参数,然后将分布参数看作是原因变量和结果变量之间的隐变量,通过比较原因变量和分布参数之间的HilberSchmidt独立性来确定二元变量的因果方向。在理论上证明了该方法的可行性,并通过实验表明该算法在人工数据和真实数据的表现较传统的IGCI,ANM,PNL,LiNGAM,SLOPE方法具有较好的准确率和稳定性。     

多变量非线性系统参数自调整的模糊加权控制

本文针对多变量非线性系统,提出了一种参数自调整的模糊加权信息融合方法．利用 模糊组合变量降低模糊控制系统的维数,根据不同的模糊组合变量对最后决策的作用大小, 赋予不同的权重来实现对多变量非线性系统的控制,在利用反向传播算法对量化系数和加权 系数进行自学习后,在线进行基于模糊规则的参数自调整,有效地解决了多变量模糊控制系 统中难于设计多维规则库和在线实现自适应模糊控制的问题．本文还对所提出的方法进行了 仿真实验和实际系统的实验,实验结果证明了该方法的有效性．     

Information systems resources and information security

Recent studies suggest that the number of information security incidents has increased dramatically and has caused significant economic loss worldwide. Awareness of the significance of information security is evidenced by a rapid increase in information security investments. Despite the fact that information security has taken on a new level of importance, academic research on this subject is still in its infancy. A review of literature indicated that past studies largely took a resource based view, suggesting that organizations invest and develop a variety of IS resources so as to ease potential threats caused by information security breaches. However, the resource-based perspective as used in previous studies was somewhat limited. Based on and extending from previous work, this study employed the resource-based view as a theoretical lens to examine the role that IS resources play in determining the level of information security. A field study was conducted to test the hypotheses. The results of the model testing show that IT human, relational, and infrastructure resources have significant impacts on information security.     

Improving scale adaptation practices in information systems research: Development and validation of a cognitive validity assessment method

Scale adaptation, where authors alter the wording of an already published scale, is a deeply rooted social practice in IS research. This paper argues that the time is ripe to question this activity as well as the beliefs that have progressively formed around it. We identify and challenge five fallacious scale adaptation beliefs that hinder the development of more robust measure development norms. Contributing to this area of research, this paper offers a conceptual definition of the cognitive validity concept, defined as the extent to which a scale is free of problematic item characteristics (PICs) that bias the survey response process and subsequent empirical results. Building on this conceptualization effort, a new methodological process for assessing the cognitive validity of adapted IS measures is introduced. Through a series of three programmatic studies, we find converging evidence that the method can benefit the IS field by making the scale adaptation process more robust, transparent, and consistent. Along with the method, we introduce a new index that IS scholars can use to benchmark the cognitive quality of their scales against venerable IS measures. We discuss the implications of our work for IS research (including detailed implementation guidelines) and provide directions for future research on measurement in IS.     

Revisiting IS business value research: what we already know,what we still need to know,and how we can get there

The business value of investments in Information Systems (IS) has been, and is predicted to remain, one of the major research topics for IS researchers. While the vast majority of research papers on IS business value find empirical evidence in favour of both the operational and strategic relevance of IS, the fundamental question of the causal relationship between IS investments and business value remains partly unexplained. Three research tasks are essential requisites on the path towards addressing this epistemological question: the synthesis of existing knowledge, the identification of a lack of knowledge and the proposition of paths for closing the knowledge gaps. This paper considers each of these tasks. Research findings include that correlations between IS investments and productivity vary widely among companies and that the mismeasurement of IS investment impact may be rooted in delayed effects. Key limitations of current research are based on the ambiguity and fuzziness of IS business value, the neglected disaggregation of IS investments, and the unexplained process of creating internal and competitive value. Addressing the limitations we suggest research paths, such as the identification of synergy opportunities of IS assets, and the explanation of relationships between IS innovation and change in IS capabilities.     

How have we evaluated software pattern application? A systematic mapping study of research design practices

ContextSoftware patterns encapsulate expert knowledge for constructing successful solutions to recurring problems. Although a large collection of software patterns is available in literature, empirical evidence on how well various patterns help in problem solving is limited and inconclusive. The context of these empirical findings is also not well understood, limiting applicability and generalizability of the findings.ObjectiveTo characterize the research design of empirical studies exploring software pattern application involving human participants.MethodWe conducted a systematic mapping study to identify and analyze 30 primary empirical studies on software pattern application, including 24 original studies and 6 replications. We characterize the research design in terms of the questions researchers have explored and the context of empirical research efforts. We also classify the studies in terms of measures used for evaluation, and threats to validity considered during study design and execution.ResultsUse of software patterns in maintenance is the most commonly investigated theme, explored in 16 studies. Object-oriented design patterns are evaluated in 14 studies while 4 studies evaluate architectural patterns. We identified 10 different constructs with 31 associated measures used to evaluate software patterns. Measures for ‘efficiency’ and ‘usability’ are commonly used to evaluate the problem solving process. While measures for ‘completeness’, ‘correctness’ and ‘quality’ are commonly used to evaluate the final artifact. Overall, ‘time to complete a task’ is the most frequently used measure, employed in 15 studies to measure ‘efficiency’. For qualitative measures, studies do not report approaches for minimizing biases 27% of the time. Nine studies do not discuss any threats to validity.ConclusionSubtle differences in study design and execution can limit comparison of findings. Establishing baselines for participants’ experience level, providing appropriate training, standardizing problem sets, and employing commonly used measures to evaluate performance can support replication and comparison of results across studies.     

Measuring e-learning systems success in an organizational context: Scale development and validation

Electronic learning (e-learning) has been widely adopted as a promising solution by many companies to offer learning-on-demand opportunities to individual employees in order to reduce training time and cost. While information systems (IS) success models have received much attention among researchers, little research has been conducted to assess the success and/or effectiveness of e-learning systems in an organizational context. Whether traditional IS success models can be extended to investigate e-learning systems success has been scarcely addressed. Based on previous IS success literature, this study developed and validated a multi-dimensional model for assessing e-learning systems success (ELSS) from the perspective of the employee (e-learner). The procedures used in conceptualizing an ELSS construct, generating items, collecting data, and validating a multiple-item scale for measuring ELSS are described. This paper presents evidence of the scale’s factor structure, reliability, content validity, criterion-related validity, convergent validity, and discriminant validity on the basis of analyzing data from a sample of 206 respondents. Theoretical and managerial implications of our results are discussed. This empirically validated instrument will be useful to researchers in developing and testing e-learning systems theories, as well as to organizations in implementing successful e-learning systems.     

Modern information technology in an old workforce: Toward a strategic research agenda

As the workforce ages rapidly in industrialized countries, a phenomenon known as the graying of the workforce, new challenges arise for firms as they have to juggle this dramatic demographical change (Trend 1) in conjunction with the proliferation of increasingly modern information and communication technologies (ICTs) (Trend 2). Although these two important workplace trends are pervasive, their interdependencies have remained largely unexplored. While Information Systems (IS) research has established the pertinence of age to IS phenomena from an empirical perspective, it has tended to model the concept merely as a control variable with limited understanding of its conceptual nature. In fact, even the few IS studies that used the concept of age as a substantive variable have mostly relied on stereotypical accounts alone to justify their age-related hypotheses. Further, most of these studies have examined the role of age in the same phenomenon (i.e., initial adoption of ICTs), implying a marked lack of diversity with respect to the phenomena under investigation. Overall, IS research has yielded only limited insight into the role of age in phenomena involving ICTs. In this essay, we argue for the importance of studying age-related impacts more carefully and across various IS phenomena, and we enable such research by providing a research agenda that IS scholars can use. In doing so, we hope that future research will further both our empirical and conceptual understanding of the managerial challenges arising from the interplay of a graying workforce and rapidly evolving ICTs.     

Enhancing e-learning effectiveness using an intelligent agent-supported personalized virtual learning environment: An empirical investigation

Virtual learning environments (VLEs) developed under constructivism and embedded personalization learning functions have the potential to meet different requirements of different learners and thus increase e-Learning effectiveness. We formulated internal personalized learning mechanisms by implementing intelligent agents in a VLE under a constructivist learning model and further developed an e-learning effectiveness framework by integrating educational and IS theories. An empirical field experiment involving 228 university students was conducted. The findings suggested that personalized e-learning facilities enhance online learning effectiveness in terms of examination, satisfaction, and self-efficacy criteria.     

基于偏相关性测试的递归式因果推断算法

因果推断是挖掘事物间联系的一种重要方式,但在高维数据场景下,利用因果推断算法进行条件独立性（CI）测试存在冗余测试多和测试效率低的问题,这限制了因果推断在高维数据集上的应用。提出一种基于偏相关性测试的递归式因果推断算法。采用“分治”的方法对变量集进行递归式因果分割,得到更易于处理的低维子数据集,提高对数据集的处理效率。在每个子数据集上进行局部因果推断,减少每次因果推断的计算量并提升算法的运行速度。在此基础上,通过比较显著性值的合并策略整合所有子结果并得到完整的因果关系,保证总体因果结构的准确性。在“分治”过程中,采用高效的偏相关性测试避免高复杂度的核密度估算,进一步提升算法效率。基于10个经典数据集的实验结果表明,在准确率与经典推断算法CAPA持平的情况下,该算法的运算速度提升了2~10倍,且在样本量越大的数据集中提升效果越明显,证明递归式因果推断算法可以有效处理高维数据集,在保证准确率的同时提高运算效率。     

A study of context inference for Web-based information systems

Recently, context-awareness has been a hot topic in the ubiquitous computing field. Numerous methods for capturing, representing and inferring context have been developed and relevant projects have been performed. Existing research has tried to determine user’s contextual information physically by using stereo type cameras, RFID, smart devices, etc. These are heavily focusing on external context such as location, temperature, light, etc. However, cognitive elements of the context are important and need more study. This paper confines its research domain to the Web-based information system (IS) and, first, proposes two algorithms, the context inference algorithm and the service recommendation algorithm, for inferring cognitive context in the IS domain. Second, this paper demonstrates cognitive context-awareness on the Web-based information systems through implementing prototype deploying the proposed algorithms. The proposed system deploying the context inference and service recommendation algorithm can help the IS user to work with an IS conveniently and enable an existing IS to deliver ubiquitous service. In this fashion, we show the direction of existing IS and, ultimately, shows the typical services of a ubiquitous computing-based IS.     

Using fuzzy cognitive map for the relationship management in airline service

This paper proposes the usage of fuzzy cognitive map (FCM) for the management of relationships among organizational members in airline service. The main task of relationship management demands consideration of the complex causal relationship among conflict, communication, balanced power, shared values, trust, and cooperation. It is difficult even for experts in organizational behavior to cognitively predict the causal effect of one factor on the others. FCM is used to describe the inference process for the relationship management in airline service. Initially, structural equation models are used for identifying relevant relationships among the factors and indicating their direction and strength. A standardized causal coefficient is then used to create a cognitive map illustrating the effect of the status of one component on the status of another component. The cognitive map provides preliminary insights into the direction of relationship management toward maximizing effectiveness of airline service.     

Information Resource Management — A cross-cultural perspective

A synopsis of research on the characteristics and culture of Chinese business methods is used as the basis for developing insights into the cultural aspects of information. Cultural implications in the design of Information Systems (IS) and Decision Support Systems are developed. The classical paradigms of IS studies are rooted in Western thought.Cultural differences need to be recognised in IS design: methodologies of design need a cultural dimension.     

基于数据立方体的动态推理控制方法

基于数据立方体的静态推理控制方法在联机分析处理(OLAP)系统中的访问有效性不高。为此,提出一种基于数据立方体的动态推理控制方法。该方法以提高OLAP系统访问有效性为目的,实时处理在线查询,分析查询方体的推理威胁,阻止其推理产生,并动态返回可访问方体集。实验结果表明,该方法可提高推理系统的有效性,且与静态推理控制方法有相同的安全性。     

一种基于CDC的适用于高维数据的因果推断算法

一对观测变量之间的因果关系的推断是科学中的基本问题,基于观测数据分析提出因果关系的方法对于产生假设和加速科学发现具有实用价值。利用传统的因果推断算法从高维数据中学习因果网络结构和提高学习准确率是目前研究的难点。在引入耦合相关系数(copula dependence coefficient,CDC)的基础上,提出了一种适用于高维数据的两步骤因果推断算法。首先该算法利用优于最大信息系数的CDC对变量间的关联度进行检测,寻找目标节点的父子节点集;然后使用非线性最小二乘独立回归算法,为图中的目标节点与其父子节点之间标注因果方向;最后迭代所有的节点完成完整的因果网络结构。实验结果表明,该算法提高了高维数据下因果网络结构学习的准确率。同时在大样本数据集中,该算法的时间复杂度优于传统算法,对异常值具有鲁棒性。     

Understanding determinants of information systems users’ behaviour: a comparison of two models in the context of integrated accounting and budgeting software

This study examines determinants that affect the behaviour of information systems (IS) users, and influence of the users’ decisions to continue using IS by two models: the technology acceptance model (TAM) and the expectation-confirmation model of IS continuance (ECM-IS). The behaviour of professionals who utilise complex software solutions for performing their working tasks has been in the focus of this research. Based on data gathered from questionnaires filled-out by users of the integrated accounting and budgeting software (IABS), the confirmatory factor analysis has shown that both models demonstrate good factor, convergence and discriminatory validity, respectively. The comparison of the obtained results has been performed, and it shows that ECM-IS has a larger explanatory power (R²) over TAM, explaining 49% of the dependent variable (IS continuance intention) in relation to 29%. The IS continuance intention is determined by the users’ satisfaction and perceived usefulness. The user satisfaction is influenced by perceived usefulness and confirmation. The confirmation of the user's expectations has had a positive influence on perceived usefulness. The perceived ease of use has not exerted a significant influence on the user's intention of IS continuance.