4.3.5 In Situ Study
4.4 Case Study 3: The Physikit Project
4.4.1 Background
4.4.2 Theory
4.4.3 Design
4.4.4 Technology
4.4.5 In Situ Study
4.5 Overall Summary
5 Practical and Ethical Issues
5.1 Introduction
5.2 Practical Challenges
5.2.1 Managing Expectations
5.2.2 Identifying and Resolving Tensions
5.2.3 Dealing with the Unexpected
5.2.4 Overcoming the Novelty Effect
5.3 Ethics: Consent, Data Collection, and Permission
5.4 Publishing Research in the Wild
Acknowledgments
We would like to thank Eva Hornecker, Jesper Kjeldskov, and Erik Stolterman for their insightful comments on an earlier draft. We would also like to thank all our colleagues at Sussex University, Open University, and University College London who collaborated with us in our forays into the wild.
CHAPTER 1
Introduction
1.1 RESEARCH GONE WILD
It is now quite common to see the phrase “in the wild” inserted into the title of a human-computer interaction (HCI) paper. Examples include “Doing innovation in the wild” (Crabtree et al., 2013a), “Being in the thick of in the wild” (Johnson et al., 2012), and “A robot in the wild” (Williams et al., 2014), as well as abbreviated versions such as “Leaving the wild” (Taylor et al., 2013) and “Calls from the wild” (Cappadonna et al., 2016). Besides attracting eyeballs (“the wild” sounds more intriguing than the more prosaic “An in situ study of…” or “An Investigation into…”) this trend reflects a shift in how research is being carried out in HCI. Increasingly, researchers are going into people’s homes, the outdoors, and public places, to study their reactions to, use, and appropriation of a diversity of technologies that researchers have provided them with or placed in that location. Examples include exploring the co-creation of a street graph depicting changes in electricity consumption for a community (Bird and Rogers, 2010), the use of mobile devices for tracking people’s health (Consolvo et al., 2008), and exploring how robots can assist the well-being of visitors in hospital wards (e.g., Dahl and Bolous, 2014). In addition, researchers are working and participating more with communities, designing and deploying technologies in situ that address the latter’s concerns or needs. Theory has also been rethought in terms of how it can inform, extend, or develop accounts of behavior that is situated in naturalistic settings and in the context of socio-technical practices.
Research in the wild (RITW) is generally considered as an umbrella term to refer to how, what, and where research is conducted in naturalistic settings (Crabtree et al., 2013b). Its overarching goal is to understanding how technology is and can be used in the everyday/real world, in order to gain new insights about: how to engage people/communities in various activities, how people’s lives are impacted by a specific technology, and what people do when encountering a new technology in a given setting. The output can be used to inform the development of new understandings, theories, or concepts about human behavior in the real world. This includes rethinking cognitive theories, in terms of ecological concepts (e.g., situated memory) and socio-cultural accounts (e.g., the effects of digitalization on society). More specifically, RITW can be concerned with investigating an assumption, such as whether or not a technology intervention can encourage people to change a behavior (e.g., exercising more). It can be operationalized in terms of a research question to be evaluated in the wild, such as: will providing free activity trackers to employees to encourage them to develop new social practices at work (e.g., buddying up, competing with each other) that will help them to become fitter and healthier? The perspective taken for this kind of RITW is to observe how people react, change and integrate the technology in question into their everyday lives over a period of time.
RITW is broad in its scope. Some have questioned the need for yet another term for what many HCI researchers would claim they have been doing for years. Indeed, applied research has been an integral part of HCI, addressing real-world problems, by conducting field studies, user studies and ethnographies. The outputs of which are intended to inform system design, often through community engagement. So, what is the value of coining another label? We would argue that, first, it is now widely used not just in HCI, but also in a number of other disciplines, including biology and psychology, reflecting a growing trend towards pursuing more research in naturalistic settings. Second, the term is more encompassing, covering a wider range of research compared with other kinds of named methodological approaches, such as Action Research, Participatory Design, or Research Through Design. Initial ethnographic research, followed by designing a new user experience, together with the application and/or development of theory, technology innovation, and an in situ evaluation study are often conducted all in one RITW project.
Hence, while the various components involved in RITW are not new, a single project often addresses several of them. Rather than focus on one aspect, e.g., developing a new technology, advancing a new method, testing the effects of a variable or reporting on the findings of a technology intervention—research in the wild typically combines a number of interlinked strands. Technology innovation can initially inspire the design of a new learning activity that in parallel is framed in terms of a particular theory of learning. Together, they inform the design of an in situ study and the research questions it will address.
RITW is agnostic about the methods, technologies, or theories it uses. Accordingly, it does not necessarily follow one kind of methodology, where one design phase follows another, but combines different ones to address a problem/concern or opportunity, as deemed fit. Sometimes, theory might be considered central and other times only marginal; sometimes, “off-the-shelf” technology is deployed and evaluated in an in situ study. Other times, the design and deployment of a novel device is the focus. In other settings, the focus of a project is how best to work alongside a community so that a democratic design process is followed.
The multiple decisions that have to be made when operationalizing a problem are often the main drivers, shaping how the proposed research will address identified questions, what methods/technologies to use and what can be learned. In summary, RITW is broadly conceived, accommodating a diversity of methodologies, epistemologies and ways of doing research. What is common to all RITW projects is the importance placed on the setting and context, conducting research in the everyday and in naturalistic environments.
1.2 HOW DOES RESEARCH IN THE WILD DIFFER FROM LAB EXPERIMENTS?
A long-standing debate in HCI is concerned with what is lost and gained when moving research out of a controlled lab setting into the wild (Preece et al., 2015). An obvious benefit is more ecological validity—an in situ study is likely to reveal more the kinds of problems and behaviors people will have and adopt if they were to use a novel device at home, at work, or elsewhere. A lab study is less likely to show