Design and evaluation of a predictive model for smartphone selection

Abstract

Selecting a mobile phone is a very subjective process; consumers often base their decisions on advertising and their personal expectations for the device. In order to provide consumers with simpler and more objective information, a predictive model for smartphone selection has been developed. Four of the most popular mobile devices were used for the development of this model: Apple's iPhone, Google's Android, Microsoft's Windows and Research In Motion's BlackBerry. Everyday tasks, common to smartphone users, were identified and modeled, using the Keystroke Level Model. Fitts' Law was used to provide additional objective data based on the dimensions and layout of the mobile phone screen. These objective measures were integrated with user preferences, to identify which smartphone would provide superior operation and performance for the features most desired by the smartphone consumer. Research outcomes from this project include the identification of the mobile devices that performed common tasks with efficiency and a user-task model predicting user smartphone selection based on individual utility and task frequency. © 2013 Springer-Verlag Berlin Heidelberg.