Evaluation of decisional guidance in decision support systems: an empirical study

A great number of studies have been done on decision support systems. There is no work done on how a machine can intelligently help a decision maker in making appropriate choice for judgmental inputs for the decision support systems. Decisional guidance is a guidance for judgmental inputs while user is interacting with a decision support system (DSS). Traditionally the guidance provided by the help menu is related to the mechanics of the system (what should be done when and how) but decisional guidance addresses the issue of judgmental inputs (what should be done why). The primary purpose of this paper is to empirically investigate usefulness of decisional guidance in a DSS. A prototype DSS was developed to help the user select the proper forecasting model for the given historical data. Both expert systems and hypermedia technologies were employed in the development of the prototype. Four alternative decisional guidance modules with different decisional guidance strategies were developed and used with the prototype DSS. The findings suggest that although the user of decisional guidance take more time in making the decisions, they make better decisions. Furthermore, guidance strategies that deliver suggestions based on the history of user interactions are more effective than the strategies that provide canned information to the users.