DECISIONAL GUIDANCE - Human-Computer Interaction and Management Information Systems: Foundations - page 74

Although this definition is expressed in the language of DSS—and decisional guidance may be
especially significant in the DSS realm—the underlying concept applies to any interactive system that
affords its users discretion. Use of such systems is intended to accomplish some task, so each point of
interaction that allows users discretion represents a point at which decisional guidance might be in
play, affecting the outcome of the task. Over the years researchers have applied decisional guidance
not only to individual decision support but also to such other domains as group decision support sys-
tems (GDSS), executive information systems, CASE tools, and conceptual data modeling. Moreover,
in the time since decisional guidance was introduced, use of applications software has grown dra-
matically with the popularity first of productivity tools (spreadsheets, word processors, presentation
graphics, and the like) and now of browser-based applications for e-business and other activities on
the Internet. These applications, too, demand analysis in terms of decisional guidance.
My previous work on decisional guidance (Silver, 1991a, 1991b) was intended to serve several
purposes:
• To make researchers and practitioners aware of a set of design issues that were largely being
ignored
• To provide researchers and practitioners with a structure for thinking about the design issues
• To pose a set of questions—mostly behavioral, but some engineering—whose answers would
enlighten the substantive aspects of design
This paper's purpose is to move forward our understanding of decisional guidance in three ways:
• To apply what has been learned about decisional guidance in the DSS and other domains to
clarify and refine the concept
• To broaden the scope of decisional guidance so it can be applied to any interactive computer-
based system
• To produce a more focused agenda for further decisional guidance research
The original work raised a great many questions about the design, deployment, and effects of
decisional guidance. The most central of these questions, to which all the others connect in some
way, is this:
• How does decisional guidance affect user behavior?
This question's apparent simplicity conceals its complexity. The question does not have a sin-
gular answer. Decisional guidance comes in many varieties and each may have a different effect.
Tasks, the people performing the tasks, and the environments within which they work also vary in
many ways, and the same guidance mechanism may affect each differently. Moreover, a given
guidance mechanism may produce a variety of effects (for instance, effects on performance, con-
fidence, satisfaction, and so forth). This paper will not try to answer the central question but to
position us better to address it.
The effects of an information system on user discretionary choices are not necessarily deliberate.
They may be inadvertent—unintended consequences of the design of some feature. For instance, the
order of items in a menu may affect the likelihood of a given item being selected. Although some
human factors researchers have examined issues that might be classified as inadvertent guidance,
information systems researchers studying guidance have focused almost exclusively on deliberate
guidance. Indeed, decisional guidance is often described in the literature as a means of building a bet-
ter or more effective system. So in common usage decisional guidance has implicitly referred to
deliberate guidance. But both forms of guidance are important if researchers are to understand system