Knowledge that Transforms

Developing a large-scale institutional DSS designed to serve multiple managers in different business functions can be a more challenging task than that of developing the much more common one-user, one-function DSS that have evolved over the past few years. In this article we review some of the evidence suggesting that extra effort and rigor in the early planning and analysis stage of large-scale DSS development is worthwhile. We attempt to identify those characteristics of DSS that require different treatment than those available in traditional structured techniques. We then present, in the form of a case study, a hybrid technique which we refer to as DSA (Decision Support Analysis) which has been used effectively in developing large-scale institutional DSS. Finally, we discuss some of the positive and negative experiences that have emerged from using DSA.

MIS performance evaluation generally includes consideration of issues such as timeliness, efficiency, effectiveness, and user satisfaction. Much of this can be subjective. For individual applications more objective criteria are necessary to monitor operational occurrences that affect overall performance. This article discusses measuring a system’s performance with an application profile, describes the components that make up a profile spreadsheet, and highlights benefits derived from its use as a management tool.

This article presents special event data for use in database management systems, along with a case study of the collection and representation of such data for an organization. Special event data account for structural changes and pattern interruptions in the time series data of an organization’s performance measures, and are important for the strategic management activities of evaluating management’s past actions and forecasting exogenous variables such as demand for products and services. Most organizations do not have a practice of collecting, storing, or sharing special event data so this valuable resource is lost as employees forget the past or take new jobs. Now that organizations are widely implementing DBMSs, it is possible and important to provide special event data. Data in the hands of individual end users can lead to errors in interpretation and use. Special event data, if they are included in the database, can help to alleviate this problem by sharing knowledge on shared data.

MIS design approaches call for managers to specify their information needs to be used as requirements in the design process. To explore the merits of this practice, managers from organizations across the United States were interviewed. The managers were asked to indicate their preference for personal, interactive, report, and analytic types of information for various stages of decision making and decision familiarity. These preferences were assessed to determine if they were influenced by managers’ personal characteristics (such as psychological makeup and experience), by the decision task, or both. Analysis found that preferences were not influenced by manager characteristics but were influenced by the decision task. The managers demonstrated an understanding of information uses consistent with the decision support literature, providing support for user specification of MIS information requirements.

The new technology of computer graphics makes it possible to create images on the screen that literally appear to pop out into the room. This is accomplished by creating separate images for the left and right eye and filtering them so that each eye receives the appropriate image. In the context of a scattergram, the binocular disparity information is interpreted by the brain as the z axis of a three dimensional x-y-z scattergram. In this article a convenient anaglyph method of accomplishing this on a standard IBM PC is presented. Also, a multi-color method using polarized projection is described and tested. The effectiveness of stereoscopic presentations is investigated within the context of managerial decision-making problems. When used to present continuous data in an x-y-z scattergram, the stereoscopic presentation resulted in greater accuracy and faster answers. However, when used with 3-D block diagrams to present semi-discrete data of the type usually presented by crosstab tables, the technique was less successful than a conventional crosstab presentation.

A test of the relationship between job characteristics and data requirements was conducted. Information system users in five public sector service organizations participated in a cross-sectional field survey. Results indicate that some job characteristics are related to data manipulation requirements, while others are related to data quality. The results support past findings outside the CBIS field that complex, nonroutine tasks with low analyzability require less precise information and more information processing than simple, routine tasks.