A Review of Subjective Scales Measuring the User Experience of Voice Assistants

Abstract

The use of Voice Assistants (VA) in both commercial and personal contexts has experienced significant growth, emphasizing the importance of assessing their user experience (UX) for long-term viability. Currently, the development of appropriate scales that capture user viewpoints after interacting with a system has become a popular method for measuring UX of the Graphical User Interface (GUI) systems. However, the applicability of these scales that are meant for GUI systems on VA is still questionable, hence the need for analyzing the nature of previous scales used for measuring UX of VA. Additionally, in order to keep track of the state of UX research in the VA domain, it is crucial to understand the dimensions of UX that are being utilized. In this study, a comprehensive Systematic Literature Review (SLR) was carried out to identify 21 individual scales used for measuring UX of VA. Furthermore, this study present the evaluation criteria for assessing the rigor of operationalization during the development of these scales. The study analysis reveals that the scales used for measuring UX of VA extends beyond the traditional VUDA (value, usability, desirability, adoptability) principles and incorporates novel aspects such as anthropomorphism and machine personality. Future VA UX researchers should also acknowledge the variations in the rigorous measures employed during scale development, notwithstanding some common and accepted practices. Consequently, an overview is provided, along with suggestions for prospective studies in the field of VA UX research.