Social Context in Usability Evaluations: Concepts, Processes and Products

This thesis addresses social context of usability evaluations. Context plays an important role in usability evaluations. A major part of the context of a usability evaluation is the people involved. This is also often referred to as the social context of the usability evaluation, and although social context is considered important, only little research has been done to identify how it influences usability evaluations. In this thesis I explore how social context affects the process and product of a usability evaluation and explain the findings in terms of the theory of behaviour settings originating from environmental psychology. This thesis consists of five published paper contributions and a summary. In the summary I motivate three research questions addressing three aspects of social context. These research questions are answered through a literature review, four laboratory experiments and a field experiment. Findings from these activities are presented in five published paper contributions. I furthermore introduce the theory of behaviour settings as a tool to help characterise the key concepts of social context which, together with an understanding of usability evaluations, provide the framework spanning my research. I then present and discuss the research methods applied in my research, followed by a conclusion on my three research questions including limitations. The primary results of my research are: 1. Applying the concept of operatives (single leader, multiple leader and joint leader) and non‐operatives (members, spectators, neutrals and potentials) from the theory of behaviour settings to usability evaluations generates an understanding and create an awareness of the level of power possessed by each of the participants in the social context. 2. On the operative level, the verbalisation and collaboration of multiple leaders in usability evaluations are affected by acquaintance, and a break down in collaboration or a decrease in verbalisation may cause the test leader to dynamically switch role during the usability evaluation to compensate. However, the influence of non‐operatives is subject to some uncertainty. 3. A careful composition of social context can successfully support problem identification. However, problem identification differs between user groups as well as between usability evaluation setups.