Attention to Ethical Aspects in Decision Support Systems (DSS): An Avenue of Strategic Importance to DSS Research

Transcription

1 Attention to Ethical Aspects in Decision Support Systems (DSS): An Avenue of Strategic Importance to DSS Research Kranti V. Toraskar, Ph.D. Associate Professor, Business Information Systems Faculty of Business Administration, University of Macau, Macau-SAR (China) Abstract While considerable attention has been given to the cognitive issues in DSS research over the years, relatively little systematic research exists about the importance, or treatment, of the ethical aspects and issues related to DSS. We examine this lopsided development of DSS research in its societal context, and argue for the need for much greater attention to DSS ethics research. First, a high-level analysis of the attention given to ethical issues in today s society, and that in DSS research, highlights the enigmatic character of the research gap involved. Furthermore, the sharp contrast between the amounts of cognitive DSS research versus ethics-related research in DSS also suggests that the problem may in fact be rooted in our misperception of the true scope, and importance, of the ethical aspects and issues in DSS in general. Based on the fundamental connotations of decision-support and the DSS context, we find that the basic DSS concept logically extends into higher systems-levels, to reveal many more kinds of DSS than are commonly recognized. Similar analysis also reveals an impressive array of ethically charged actors (ECA) along the various stages of the DSS life-cycle. The resulting range and scope of DSS-like systems and associated ethical issues is indeed enormous. Given this wide-open scope of DSS ethics in reality, we argue that a comprehensive program of DSS ethics research may actually hold significant strategic importance for future DSS research as a whole. Keywords: Decision support systems, Ethics, Ethical aspects, Information systems. Introduction Presently, one merely needs to look at the major news stories of recent years to realize that moral and ethical issues are increasingly appearing at the core of many corporate, social, as well as scientific debates and controversies that seem so characteristic of these modern times. Here, we can only afford to mention just a few such news stories and related controversies: the Exxon Valdez oil spill, the Union Carbide accident, the ENRON case, the Martha Stewart case, industrial development versus rain forests and wild-life conservation, human cloning, genetically modified foods, gun-control laws, information privacy, and even as we speak, the public transparency versus a government s need for the security/integrity of its decisionmaking! The main reason for mentioning these examples here, at the outset, is to give the reader an idea of our overall frame of reference, a kind of systems-level anchoring, for the general approach of this paper and the discussion that follows. While detailed analysis of any of these stories is clearly beyond our scope here, the paper is indeed inspired by the author s conviction that, at present DSS ethics research (meaning, research on the importance, and handling, of various ethical aspects/issues related to DSS) could advance a lot by focusing on the various systems-levels in the context of the DSS (rather than on the DSS itself).

2 88 In what follows, we examine the lopsided development of DSS research in its societal context, and argue for the need for much greater attention to DSS ethics research. First, a high-level analysis of the attention given to ethical issues in today s society, and that in DSS research, highlights the pragmatic need for closing the research gap involved. Furthermore, the sharp contrast between the amounts of cognitive DSS research versus ethics-related research in DSS suggests that the problem may in fact be rooted in our misperception of the true scope, and importance, of ethical aspects and issues in DSS in general. Based on the fundamental connotations of the decision-support perspective, we find that the basic idea of a DSS logically extends itself into higher systems-levels to reveal many more varieties or types of DSS than are commonly recognized. Similar analysis also reveals an impressive array of ethically charged actors (ECA) along the various stages of DSS life-cycle. The resulting range and scope of DSS-like systems and associated ethically charged situations is indeed enormous. In the last section, we discuss the major implications of this wide-open scope of ethics in DSS for the DSS research as a whole. The Enigmatic Lack of Research on Ethical Issues in DSS The lack of research on ethical issues is well acknowledged in DSS literature (Carlson et al., 1999; Meredith and Arnott, 2003). Here, we examine this lack of DSS ethics research from some macro-level perspectives. First, it is somewhat puzzling that while ethical and moral issues have been attracting huge attention in the corporate and social arenas for quite many years, this attention remains far from reflected in DSS literature. If anything, the level of attention to the ethical issues in DSS research seems to have been chronically low. For example, nearly 20 years ago Boland (1987) recognized the fundamentally moral nature of the problem of designing information system (IS) in general. Now, ever since the inception of DSS in the early 1970s (Gorry and Scott Morton, 1971), we have known that DSS is simply a type of IS; the type that is specifically designed to support what are called semistructured decisions through an emphasis on variety of data, models and interactive userinterfaces. And yet, 26 years later, Sauter (1997, p. 26) expresses real concern whether ethical issues in DSS are given as much attention as the financial and technical issues. Even most recently, Meredith and Arnott (2003) are disappointed at the paucity of published research and debate on the ethics of decision support From a research methodology standpoint, the above lack of attention to ethics in DSS research seems rather intriguing, or enigmatic, when we realize that a lot of attention has been given in DSS literature to the cognitive aspects, which is in fact another major sibling-theme of this very Symposium. Is this pure coincidence? Not likely. Is there something more substantive than petty sibling rivalry of the themes involved here? The author thinks so. To elaborate, we should note an interesting pattern of correlation and contrasts between the various aspects of the systems being studied (here, DSS). Specifically, note (1) the inherent relatedness among the trinity of the Cognitive, Ethical and Emotive aspects in general and (2) their joint contrast with a group of fundamentally different aspects, such as, for example: Business/Economic, and Technical. From the systems holistic point of view, this pattern of contrasts and correlation between the various aspects (in DSS) clearly indicates that DSS researchers should pay much more balanced attention to the various aspects in DSS. For example, research on the Emotive Aspects is essentially non-existent in DSS today.

3 89 Clearly, there may well be a wide variety of inter-related causal factors, at various system-levels, that might fully explain the current lack of DSS ethics research. Without minimizing the importance of understanding all such factors, the author does believe that at present one such factor is simply our lack of awareness of the true scope, and importance of ethics in DSS, i.e. the various ethical aspects and issues related to DSS. Below, we elaborate an expanded conception of the scope of DSS and related ethical situations. Briefly, this conception quite logically unfolds itself in terms of two main dimensions, namely: (a) the true scope of DSS-type systems in today s society, and (b) A variety of Ethically Charged Actors (ECA) along the successive phases of DSS Life Cycle. Below, we first explain how the fundamental connotations of decision-support, and the DSS context, lead to identification of some new types, as well as levels, of DSS in today s environment. The Expanding Scope of the DSS Concept in Today s Society Thanks to the power that computers bring to a DSS, today it may be hard to imagine a DSS without computers. However, strictly speaking, we must recognize that our computerbased stereotype of DSS is not a theoretical (conceptual) restriction, but a technological (implementation) option for empowering the DSS. The decision-support framework or perspective (Gorry and Scott Morton, 1971) only requires that the decisions involved are semi- or un-structured, and that the support-system involved should only support, and not replace, the decision-maker (DM). So, in theory there is no reason why a DSS can t be a human system, a computer, or a combination of the two. Once we relax the computercondition on the DSS, and take a fresh look at the societal-level news stories and decisionmaking involved, we begin to see many examples of human DSS supporting DMs at system-levels beyond the level of a single business organization. The obvious examples of such human DSS are the economist advising an Industry Consortium, personal staff of a nation s Head, and the health-experts advising World Health Organization (WHO). The above analysis immediately expands the scope of the DSS concept along two different dimensions: 1) the nature of the DSS (human or computer or hybrid), and 2) the DM s System-levels where we clearly see the Societal-level DM in addition to the traditional levels of an Individual Manager, a Group of managers, and an entire Organization. Marakas (2003) has already introduced the concept of Organizational DSS (ODSS), above the commonly recognized DSS-types, i.e. the basic DSS for one manager, the Group Support Systems (GSS) and the Group Decision Support System (GDSS). In addition, given the powerful computing devices/networks available to private citizens today, here we also recognize that an individual DM can easily be a private citizen using a personal DSS (PDSS), or a professional (e.g. neighborhood tax consultant) in addition to the classical business manager. To conclude, in addition to the classical categories of basic DSS, GSS, and GDSS, we now see the following new DSS-types their contexts: a PDSS, a professional s DSS, the ODSS, the National DSS (NDSS), and finally a Multi-national DSS (MNDSS). Finally, at the lowest level, we cannot completely ignore the theoretical possibility of a computerized DM supported by a human DSS. As humans, we would prefer such a DSS to remain nameless!

4 90 The Variety of Ethically Charged Actors (ECA) Involved with DSS First, we should realize that the real scope of what could be labeled as an ethically charged situation, in general, indeed covers the entire realm of human activity. Whenever an individual (or an organization) makes choice(s) that can potentially affect some objectives and/or stakeholders outside the scope of authority and responsibility of that individual, we see it as an ethically charged situation, and the individual or decision-maker (DM) in question as an ethically charged actor or ECA. Bringing this ECA concept to bear upon the typical context of DSS use, makes it easy to see that this is indeed an ethically highly charged situation by its very nature and the roles of the parties involved. The parties involved in the context of DSS use minimally include the DSS-user or the DM, his/her employing Organization, any DSS-intermediary staff (Marakas, 2003), and of course the DSS itself. The ethical significance of the DSS design has been recognized in the existing literature (Carlson et al., 1999). The DM is clearly an ECA since he/she is ultimately responsible for the difficult (unstructured) decision that may affect stakeholders outside the DM s direct influence. The DM s employer organization almost always exerts a powerful influence on DM s final decision. The DSS-intermediary often has greater knowledge about the DSS and its technology, but is usually subordinate to the DM. Both these factors clearly add special ethical challenges to the job of the intermediary. In conclusion, all four actors involved in the use of a DSS are clearly ECAs. The Variety of ECAs along the Phases of DSS Life Cycle: Given the universal scope of ethically charged human activity, our search for DSS-related ECA should not end with DSS use. To begin with, the moral nature of the task of designing information system (IS), in general, is well acknowledged in the literature (Boland, 1987). For DSS in particular, Meredith and Arnott (2003) have suggested that ethical activities pervade every aspect of development, deployment and use of the system. In fact, we believe the entire DSS lifecycle is likely to contain more ethical charge than other information systems in general, because of (a) the use of the latest hardware/software technologies in DSS and (b) the rapid pace of DSS development. Thus, we have every reason to find many more DSS-related ECAs by extending our search for ECAs to all the stages of DSS life-cycle, as shown below. While the prototyping approach is commonly associated with DSS construction activities, the complete life-cycle of a large-scale DSS still covers the following general phases: Project Definition, Analysis/Design, Construction, Implementation, Normal Use, Maintenance, and Evaluation. The project definition phase concerns identifying the specific decision problem calling for a DSS solution, planning the DSS development project itself, and obtaining the necessary resources to carry it out. Even from this basic description of this phase, it is clear that the DSS Champion, internal DSS professionals, the financial officers, outside DSS consultants and vendors, will all make certain ethical choices as they contribute to the project definition. In particular, Chae et al. (2002) have elaborated on the heavily value laden nature of the problem formulation activity which is of paramount importance during this phase. Due to limitations of space, below we list only the major ECAs that can be easily recognized in other phases of the DSS life-cycle: System Analysis/Design: DSS Analyst/Designers, Database Specialists, M.S./O.R. Staff Construction: Programmers, IT Staff involved, Documentation Writers Implementation: System Testers, User Training Staff, User-Department Managers

5 91 Maintenance: Software vendors (new versions?), Troubleshooters, Industry s DSS Experts Implications for Future DSS Research The wide varieties and levels of DSS and ECAs, elaborated above, indeed project an enormous scope of ethical aspects and issues in DSS. The theoretical validity of such an allencompassing view of ethics in DSS lies in the fact that it was derived directly from the fundamental ideas of decision-support and the systems context of DSS. The wide-open scope of DSS ethics has equally far-reaching implications for future DSS research. First, the rich and comprehensive picture of the variety of DSS-types, with the many ECAs in various DSS related activities, highlights the need for a much greater attention to DSS ethics research. Secondly, the elaborate structure of the levels/types of DSS and ECAs involved can help specific researchers chart out specific areas/topics of DSS ethics research where they can contribute most effectively. Finally, the existence of the new DSS-types at personal and societal levels suggests that, in future we might see mounting public demand for more ethical decisions, and so for more ethical DSS, at these levels. Decision-makers may become reluctant to rely on their DSS, but may still blame the DSS for poor decisions. In the end, unless DSS ethics research gains momentum fast, the credibility and value of DSS as a field could itself come under scrutiny. Therefore, at present, it may be crucial for DSS researchers to speed up the research on ethical issues in DSS. In this sense, a comprehensive program of DSS ethics research seems to hold significant strategic importance for the future of DSS and DSS research in general. References Chae, B., Courtney, J.F. and D. Paradice (2002); Incorporating an ethical perspective into decision support systems design, Proc. IFIP TC8/WG 8.3 Open Conference on Decision Making and Decision Support in the Internet Age (DSI-Age 2002), Cork, Ireland, (pp ). Boland, R. (1987); The in-formation of information systems, in: Wiley Series in Information Systems (Eds. Boland, R. and R, Hirschheim), John Wiley & Sons, (pp ). Carlson, J., Carlson, D. and L. Wadsworth (1999); On the relationship between DSS design characteristics and ethical decision making, Journal of Managerial Issues, Vol. XI, No. 2 (pp ). Gorry, G.A. and M.S. Scott Morton (1971); A framework for management information systems, Sloan Management Review, Vol. 13, No. 1, (pp ). Marakas, G.M. (2003); Decision Support Systems in the 21 st Century, 2 nd ed, Prentice-Hall. Meredith, R.A. and D.R. Arnott (2003); On Ethics and Decision Support Systems Development, Proc. 7th Pacific-Asia Conference on Information Systems (PACIS2003), Adelaide, Australia, (pp ). Sauter, V. (1997); Decision Support Systems, John Wiley & Sons.