User performance and attitude towards schemes for alphanumeric data entry using restricted input devices

Results are reported from two usability trials in which a total of 14 schemes for alphanumeric data entry on television screws were evaluated. The seven schemes in the first trial used only the navigation and select keys on the TV remote control unit. The second trial used the best scheme from the first trial (slightly modified), a further six schemes using different combinations of the navigation and alphanumeric keys and a portable, infra-red keyboard. In both trials task completion rates and times were measured together with the number of keystrokes. User preferences were measured using attitude questionnaires and explicit ranking of the schemes. As expected, the keyboard was, overall, the best scheme-though not the most accurate. Where only a limited range of input keys were available, grid display schemes were preferred over menus and dynamic displays. Where an alphanumeric keypad was available on the remote control unit, a multiple press scheme was marginally preferred.