The Optimal Size of Handwriting Character Input Boxes on PDAs

Abstract

This article presents an empirical study to determine the optimal size of a box for the input of handwriting characters on personal digital assistants (PDAs). The experiments involve the consideration of boxes for different kinds of characters, different box sizes and shapes (square and rectangular), different user postures, and the age differences of users. The results are assessed in terms of high performance (high character recognition rates, minimal stroke protrusions outside the character box, minimal number of error corrections, minimum writing time) and subjective ratings (e.g., ease of writing and minimum degree of fatigue). The results show that the optimal size of character boxes for the input of alphanumeric characters is 12 × 14 mm (rectangular), whereas for Kanji (Chinese characters) mixed with Kana characters and for Hiragana & Katakana characters the optimal size is 14 × 14 mm (square). We believe that knowledge of the optimal size of character input boxes will be useful for the design of user interfaces on PDAs.

This study has been partially supported by Grant-in-Aid for Scientific Research (No. 51-8103-0411) and Microsoft Research Asia Mobile Computing in Education Theme. We thank Taishi Kato and Fumiya Fukutoku for help in arranging the experiments, and the meta-reviewer and the anonymous reviewers for the valuable comments, which motivated much of the discussion and presentation in the article. We also thank the participants for their participation and patience, and the members of the Ren Lab at Kochi University of Technology for their warm support.

Notes

¹ATOK means Advanced Technology of Kana & Kanji Transfer, a Japanese input system that is the same as Microsoft input method editor. ATOK Pocket is used on mobile phones and PDAs.

²We also asked the participants to rate “readability,” however, the definition was difficult for them grasp, thus we omitted the data.

³We omitted Hiragana & Katakana in this experiment because the result for Hiragana & Katakana was the same as Kanji & Kana in Experiments 1 and 2, and it is included in Kanji & Kana; we also wanted to reduce the experimental load.

⁴We deleted the Space icon because no space was input in the experiment, and we placed the Delete icon under the two boxes to make the two input boxes bigger because we assumed the older adults would prefer bigger boxes.