Decision Support Systems

Task motivation and DSS effectiveness and efficiency are constructs in the motivational framework for understanding DSS use and decision performance. Task motivation is an important variable that influences DSS use. Since TAM does not model task (intrinsic) motivation explicitly, Venkatesh attempts to fill this gap by conceptualizing intrinsic motivation as computer playfulness. To augment these efforts, Chan proposes a research framework that links DSS effectiveness and efficiency with task motivation. In this framework, the effects of DSS effectiveness and efficiency are moderated by task motivation while task motivation has a direct effect on DSS use. In particular, the author examines whether task motivation affects use of a DSS to do a task and whether task motivation interacts with DSS effectiveness and efficiency to affect DSS use.

Chan conducts an experiment where the participants use a DSS to do one of two choice tasks that induces different levels of task motivation. The total number of iterations of the participants' use of the DSS and the total time taken on each choice task are captured and used as dependent variables. The results show that participants in the high task motivation condition use the DSS more (i.e., they have more iterations and spend more time on the task) than those in the low task motivation condition. Individuals performing a high motivation task also use a DSS more when it is more effective while DSS effectiveness does not affect the level of usage for individuals doing a low motivation task. In addition, the findings indicate that DSS efficiency has a significant impact on DSS use for individuals working on a high or low motivation task when DSS use is measured as the extent of use (i.e, the number of iterations or total time spent on a task). However, DSS efficiency does not have a significant impact on DSS use in the high task motivation condition when the DSS use construct is dichotomized as use or non-use rather than the extent of use. This result is consistent with the author's expectation that individuals performing a high motivation task are less concerned with the efficiency of a DSS.

In summary, DSS use increases (decreases) for individuals using a more (less) effective DSS to work on a high motivation task. As expected, DSS effectiveness is not a concern when individuals perform a low motivation task. The findings suggest that the strong negative impact of lack of task motivation undermines DSS use, regardless of the level of its effectiveness. The efficiency of a DSS is found to interact with task motivation to affect DSS use. That is, individuals completing a high motivation task exhibit higher tolerance for a DSS that is low in efficiency. In contrast, lack of task motivation exacerbates the users' low tolerance for a DSS that is low in efficiency.

An interesting design of the DSS in Chan's study is the built-in feature of an effortful but accurate decision strategy -- additive difference (AD). AD processing compares two alternatives simultaneously by comparing each attribute, finding the difference, and summing the differences. It requires some method for weighting each attribute, some transformation to put all the attributes into compensatory units, and a way to sum the weighted values of the attributes. After a series of alternative comparisons, the alternative with the greatest sum is chosen. AD processing is compensatory in that values on one attribute necessarily offset the values on another attribute. It makes more complete use of the available information and is normatively more accurate than non-compensatory strategies such as elimination-by-aspects. Use of the more accurate and more effortful AD strategy relative to other less accurate and less effortful strategies (e.g., elimination by aspects) may be encouraged if users are provided with a DSS that reduces the cognitive effort for using the AD strategy to complete a task. The effort required for completing a task is minimal when the DSS provides high support for the AD strategy. In the study by Chan, individuals use a DSS to select two alternatives for comparison and the DSS provides the results of how the selected alternatives differed on the attributes. Thus, the DSS in the study provides enhanced automation that reduces the effort that a user may otherwise have to expend to process information manually.

The next section describes a study by Chan et al. that examines the effects of feedback and reward on decision performance.