Multi-Criteria Evaluation Approach of Mobile Text Entry Methods

Introduction

Increased production and development of mobile devices such as mobile phones for the past four decades revolutionized global communication modes. Integration with other advanced capabilities such as web browsing, messaging, gaming and computing among others intensifies the need to address usability issues associated with this functional complexity. Designers need to be updated with the current information on mobile phone design imperatives to keep abreast with usability issues. Such design information includes product related criteria and user related criteria (Isiklar and Buyukozkan, 2007). Part of the design decisions is the choice of mobile text entry method especially in text messaging using short messaging services (SMS). In Asia, SMS draws its success in Singapore, Malaysia and the Philippines (Balakrishnan and Yeow, 2008). Unlike making phone calls, popularity comes with SMS as it is relatively cheaper and can be created simultaneously with other tasks. This motivates a vast area of research in mobile text entry over the years (Dunlop and Montgomery Masters, 2008).

Issues such as input speed, accuracy, physical form, learning time and cost are known to be important issues in mobile text input technology (Green et al., 2004). These issues can be grouped accordingly into two types of usability measurements, namely: performance measurements and preference measurements (Koivisto and Urbaczewski, 2005). The former collects objective metrics to measure system performance which usually consists of attributes referring to speed and accuracy while the latter denotes subjective preferences and opinion data which may include physical form or aesthetics, learning time and cost, all vary with individual judgments. With this, evaluation of text entry methods must not only include speed and accuracy performance, as what former evaluations have done, but must encompass subjective judgments of users. Such an evaluation would consist of activities which are complex, multi-dimensional and may involve different users with different priorities and objectives. For instance, the evaluation of mobile text entry methods must not only consider performance measurements but must also address how subjective preferences such as prior experience, familiarity and popularity come to play. This is often a multi-criteria decision problem characterized by interrelated performance and preference issues. In this context, multi-criteria decision-making (MCDM) methods play a vital and helpful role in the choice of text entry methods. However, little work has been done in the literature on the use of MCDM tools in human-computer studies in general and mobile text entry research in particular. See the works of Chen, et al. (2009), Delice and Gungor (2009), Hsiao, et al. (2010), Hsiao (1998), Lin and Hwang (1999), Isiklar and Buyukozkan (2007) for the application of MCDM methods on usability studies.