Determining where to include users in the icon development process: A reliability and validity study

Date of Completion

January 1998


Design and Decorative Arts|Psychology, Industrial|Computer Science




Computer icons are often implemented on user interfaces without being treated as interfaces themselves. The current study examines the various methods of involving users in the icon development process, whether as the designers or the evaluators for a set of icons. Along with a review of past research, the problems that currently exist with icon development are noted. Four experiments are conducted to examine the reliability of the various methods of involving users in the icon development process: the sign production method, the open-ended meaning-for-image test, the image-meaning-matching test, and the image-meaning-association test. These four methods' validity of yielding a usable set of icons is also assessed with the conduction of actual usability testing of the icons. Post-hoc analyses examine if the open-ended meaning-for-image and the image-meaning-matching tests may be used interchangeably. A final summary provides insight into the conducted experiments, implications of the findings, and potential future research. For utilization of the findings in an applied setting, the various methods reviewed in this study were compared for the time and resources needed for their implementation. ^