upa - home page JUS - Journal of usability studies
An international peer-reviewed journal

Switching Between Tools in Complex Applications

Will Schroeder

Journal of Usability Studies, Volume 3, Issue 4, August 2008, pp. 173-188

Article Contents


We revisited whether some of the factors given in Table 1 contribute to differences in tool switching behavior such as those shown in Figures 3 and 4. Table 2 collapses Table 1 into three groups, either by averaging individual results or by expressing class (expert or novice) representation as ratios. These are the same groups that appear in the last column of Table 1.

Before examining Table 2, the experimenter naively assumed that the Low group had the best tool switching strategy and the High group the worst, based on a gut feeling that switching should be minimized. He could not have been more wrong. Not only did the Low group perform the worst (fewest segment completions, Finished (of 5) section in Table 2), but they were the most unhappy with the tools (SUS), and had the lowest in exposure to the product (3.2 files written per month). The High group, whose timelines appear almost dysfunctionally scattered, were in fact all-expert, and they proved it by completing by far the most task segments. Not only did they score the best, they also rated themselves the best and most comfortable with the software.

Table 2. Metrics by Style Grouping of Switching Statistics
Group Expert: Novice SUS Finished (of 5) Skill (MATLAB) Comfort (MATLAB) Years Programmer: Plotter Files/ Month
High 2:0 73.75 3.5 3.5 4.5 5.5 2:0 12
Mid. 2:3 80 1.8 2.4 4 1.24 2:3 11.8
Low 1:4 60.5 1.2 1.4 2.8 3.9 2:2 3.2

The manner in which this grouping of users by tool switching style organizes scores, subjective evaluations of MATLABŪ (SUS), and users' self-ratings and experience is the most important finding in the study thus far. It counter-intuitively suggests that more keystrokes and more switching "work" can be better, which goes to show how much we have yet to learn about tool switching.

Previous | Next