A new error metric for text entry method evaluation
Jun Gong, Peter Tarasewich · 2006
On devices such as mobile phones, text is often entered using keypads and predictive text entry techniques. Current metrics used for measuring text entry error rates have limitations in terms of the types of errors they account for, and cannot easily distinguish between different types of errors. This research proposes a new text entry error metric that addresses some of the outstanding issues that exist with current metrics. Specifically, the metric accounts in detail for the way the user handles corrections during text entry, moving beyond current keystroke level error measurement. The feasibility and usefulness of this new metric is shown through the analysis of an experiment that tests an alphabetically constrained keypad design that includes upper and lower case letters, numbers, and punctuation marks.