Individual Differences in Perceived Importance of SWAT Workload Dimensions: Effects on Judgment and Performance in a Virtual High Workload Environment

The Subjective Workload Assessment Technique (SWAT) is a numerical conjoint scaling procedure that is often used to construct estimates of overall perceived workload and the importance of component dimensions of time, effort, and stress load demands in evaluating workload. In this study initial SWAT scales were first constructed for 124 individuals who were then classified into one of six “workload prototype” groups based on a cluster analysis of their derived SWAT dimension weights. Individuals were then placed into three different virtual environment scenarios for which the task had either a high time, high mental effort, or high stress load component, respectively. Workload judgments and performance scores were obtained for each scenario. Results indicated that how individuals weighted the SWAT dimensions influenced both their judged workload for the scenarios and their performance measures in each scenario.

A Conjoint Analysis of Five Factors Influencing Heavy Vehicle Drivers Perceptions of Workload

Conjoint analysis, a multi-factor subjective scaling technique, was used in a study of heavy vehicle drivers to obtain a measure of their perceived workload demands under different driving conditions. These included combinations of low and high levels of traffic density, lighting, roadway type, visibility, and traction. A tradeoff comparison analysis was used to collect the conjoint scaling data from a subset of the complete 2−2−2−2−2 design. Results indicated that an additive factor representation fit the data very well, but that the five factors had very different importance weights. The drivers’ orderings of perceived demand appeared to be inversely related to their control over the conditions. The two most important factors (traction and visibility) are effectively environmental factors that cannot be easily controlled by the driver. The other three factors (traffic density, highway type, and lighting) can, at least to some extent, come under the control of the driver. Implications of these results and the use of conjoint scaling methodology are discussed.