Multi-Criteria Evaluation Approach of Mobile Text Entry Methods

Abstract

Current evaluations on mobile text entry methods focus only on speed and accuracy performance but fail to consider subjective preferences of users. Such evaluation is ill-defined and does not constitute a systems approach since it requires a complex multi-criteria decision process that must encompass both objective and subjective criteria. Thus, this paper proposes the use of Analytic Network Process (ANP) in the evaluation of five mobile text entry methods in Android smartphones. A hierarchical network decision problem is developed from a focus group discussion of keen smartphone users. Using ANP, new insights in research and design of text entry methods are proposed. Results show that QWERTY method is the most preferred text entry method and is influenced by criteria on perceived text entry speed and number of keys. A survey is conducted to validate the results of the evaluation process using ANP. The main contribution of this work is in introducing a multi-criteria evaluation approach for mobile text entry research that holistically addresses speed-accuracy performance along with relevant subjective criteria that may potentially influence decision-making.