Switching Between Tools in Complex Applications
Journal of Usability Studies, Volume 3, Issue 4, August 2008, pp. 173-188
Abstract
Large software applications are made up of many specialized tools. In Microsoft Word the document editor is supported by tools to create and fix drawings and tables. Programming environments have custom views (difference editors) and analyses (performance reports) to help developers make robust code. Every application has tools to help users sift the documentation.
In usability, we usually test a tool at a time, yet complex work requires many tools, and this brings a new set of issues. How do I know when I should be using a different tool? What tool do I need when the one I am using is not working? How do I get to it? How quickly can I start using it?
In complex or creative work, our observations show that users seldom choose the correct tool as soon as work progress dictates. This erodes productivity and creativity and is a prime target for improved designs.
Usability practice needs a procedure to identify, record, count, and highlight tool switch events for study. This paper describes one that supports the trained observers on which User-Centered Design relies to detect problems and causes, and evaluate design changes.
Practitioner's Take Away
Methods presented in this paper serve as a foundation for the following:
- Comparison of users’ performance in complex tasks by their choice of workflow path elements (tool use) and strategies (tool choice and decision timing).
- Methods of visualizing and analyzing logged computer-user interaction data to study switching behavior.
- Transition statistics (traffic between tools, time on support tools) locate intense switching activity and compare one user’s tool strategies to another’s.
- Transition timelines (graphics) highlight problematic patterns of activity.
- Use of transition statistics as metrics to identify, zero in on, and characterize successful and unsuccessful design features.
- Use of the Standard Usability Scale (SUS) and task completion scores to indicate users’ awareness of their tool switching patterns and the effectiveness or efficiency of these patterns.
- Use of logging, counting, and plotting (visualizing) test data to extend conventional usability practice by enabling consideration of new problem areas.
- A set of baseline data against which design changes in a specific area of MATLAB®, or any other complex software toolset, can be evaluated for their impact on tool switching behavior, alongside their effect based on standard usability metrics.
Article Contents
- Introduction
- Background
- Methods of observation
- Interpreting the timelines
- Results
- Discussion
- Recommendations (What's Next)
- Conclusion
- References
Download Article
Switching Between Tools in Complex Applications