Page 245 -

9.3 Ethnography in HCI 233

9.3 ETHNOGRAPHY IN HCI

The description of ethnography as the practice of using some form of participation
in a group to develop an understanding of the group is straight from social science
research. Social science ethnographers spend time living in traditional villages,
hanging out on inner-city street corners, and otherwise immersing themselves in un-
familiar settings to understand the dynamics of groups of interest.
As fascinating as this might sound, it may also seem a bit far removed from re-
search into HCI. After all, HCI researchers are usually trying to understand how to
build systems or how users interact with computers. How does this relate to the in-
depth study of groups and why is participation useful and helpful?
The connection becomes clearer once we consider the use of modern computer
systems. Even when we are sitting in front of a traditional computer, conducting seem-
ingly familiar tasks such as word processing, we're not really computing so much as
we are communicating. Much of our computing work that does not directly involve
communication or collaboration (e-mail, instant messaging, online calendars, virtual
worlds)—involves creating artifacts (documents, spreadsheets, presentations) that
communicate ideas to others. Mobile and ubiquitous computing tools that make com-
puting a more integrated part of daily life are even more focused on communication.
As soon as we start using computing technologies for communication and collab-
oration, we start forming groups. Whether these groups are “real” groups that have
some physical existence outside the computing environment, such as schools (Wyeth,
2006), homes (Crabtree and Rodden, 2004; Taylor and Swan, 2005), and workplaces
(Newman and Landay, 2000; Su and Mark, 2008), or are groups that would not exist
without the technological intermediary, such as virtual worlds (Ducheneaut et al.,
2007), they have their own norms and dynamics that are legitimate and important
subjects of study.
But what does the HCI researcher hope to learn about these groups? Often, the
goal is just understanding: How is a technology used? How do the features of the
design influence how people use the system? HCI researchers can use ethnographic
techniques of participating in the group to gain a detailed and nuanced understanding
that other methods cannot provide.
Lucy Suchman's study of the users of an electronic help system on a photocopier
is perhaps the most famous example of ethnography in HCI. Starting from a frame-
work that describes all action as being a product of the context in which it is taken—a
model known as situated action—Suchman observed users attempting to complete a
photocopying task with the help of an expert system designed to help them identify
problems and complete tasks correctly. Through analysis of videos and a framework
designed to demonstrate the relevant features of the interactions between the humans
and the expert system, Suchman developed a rich and detailed understanding of how
differences between the human model of the copier and the expert system's model
led to communication breakdowns and task failures (Suchman, 1987). This study re-
mains influential both as a fascinating discussion of how problems in human-machine
communication can arise and as an example of the utility of ethnography in HCI.