Page 493 -

15.5 Summary 485

Researchers involved in these studies—or, for that matter, in prior studies we now
think of as abusive or inappropriate—were not necessarily acting in bad faith or out
of lack of concern. This is exactly why we need careful, independent review to en-
sure respect for persons, beneficence, and justice (The National Commission for the
Protection of Human Subjects of Biomedical and Behavioral Research, 1979), and
to justify the trust that society requires when we ask individuals to participate in our
studies and governments and foundations to financially support our work.

15.5 SUMMARY
Working with human subjects is one of the most challenging and informative aspects
of HCI research. Finding appropriate participants; informing them of their rights;
protecting their privacy; and answering their questions can be time consuming and
often tedious, but the results are more than worth the effort. Even when study partici-
pants criticize our designs or fail to confirm our cherished experimental hypotheses,
they provide invaluable insight that provides a rigorous foundation for our work.
Whatever type of study you are running, it is never too early to plan for recruiting,
informed consent documentation, and other aspects of human participation. Proper
planning will keep your study from becoming one of the many that have been de-
layed by unforeseen circumstances including difficulty in finding participants, or
delays in IRB approval.
Recruiting entails finding the right number of the right kinds of participants. For
usability studies, ethnographic observations of users, interviews, focus groups, and
other approaches aimed at gathering requirements or evaluating design proposals,
this may mean understanding the audience of users and identifying a sample of par-
ticipants that is broad enough to reflect the needs and behavior of potential users.
Designers and professional developers conducting research of this sort might work
with collaborators, marketing teams, professional organizations, or others with ap-
propriate understanding and context to identify both the range of viewpoints that
would be needed and possible sources of the appropriate individuals.
Empirical studies require consideration of both the diversity of potential partici-
pants and any confounding factors that might contribute to performance differences.
Characteristics of desirable participants might both be informed by and influence
experimental hypotheses. Students and researchers conducting these studies should
be careful to plan their data analysis and recruiting together, to ensure that the partici-
pants will be selected to increase the power of the statistical analysis.
Appropriate respect for participants is a cornerstone of all research involving
human subjects. These issues are particularly relevant for studies that involve decep-
tion. Even when not required by institutional policy to do so, designers and develop-
ers would be advised to use formal informed consent forms to help participants make
informed decisions. Students and researchers should take the time—again, as early
in the process as possible—to understand the regulations in force in their institution,
and to make sure that their approvals are in order before starting any project.