Recent work in human-computer interaction has demonstrated that unconstrained text entry protocols that provide a more natural environment for research participants. We demonstrate the application of this approach to the analysis of word prediction. Eleven participants (5 nondisabled/6 with disabilities) transcribed sentences using an on-screen keyboard both with and without word prediction while time-stamped keystroke data were collected. The subsequent analysis demonstrated that the entire input stream (including erroneous keystrokes and the keystrokes used to correct errors) can be included in the evaluation of performance with a text entry device or text entry rate enhancement method.
Volume 46 Number 8, 2009
Pages 1059 — 1068