When It Doesn't Add Up: The Effect of Combining Command-Name Guidelines

Abstract

The purpose of the present study was to examine the effect of combining several command-name guidelines in the human-factors literature. The particular guidelines examined included the naturalness, specificity, frequency of use, and form (words or symbols) of the command names; and the relationship among the commands within a set (i.e., congruency). Four command sets were studied: a noncompliant set that violated all five guidelines; a mixed set that adhered to all five guidelines; and a symbolic and a verbal set that both violated the form guideline. Three categories of dependent variables were examined: performance (time, errors, and the use of help and cancel functions); recall; and preference measures. The results of the study indicate that the extent to which guidelines can be combined differ depending on the type of dependent variable examined. Further, the results do not support the conclusion that the more guidelines followed, the better the user's performance. The conclusions discussed include how to interpret guidelines based on past research, suggestions for combining existing guidelines, and implications for future research.