Research Projects

My Approach to Design

My research draws on many research domains—human-computer interaction, computer-supported cooperative work, learning sciences, social psychology, management—and my approach to research methods reflects this. I believe that the types of questions I ask are necessarily multifaceted; it's difficult to capture the complexities of real-world interaction through only one approach. To answer these questions, a combination of quantitative and qualitative methods is necessary. Ethnographic techniques are particularly effective for exploring new problem domains in depth; experimental techniques allow us to begin to answer specific questions to some degree of certainty. Only through moving between these approaches, however, can we develop a sufficiently nuanced understanding of how our design decisions influence interaction.

Although I use techniques from both ends of this methodological spectrum, I rely heavily on quantitative, quasi-experimental methods supplemented with interviews and other ethnographic techniques. In this way, I can control some variables, but still allow for some of the complexities of the real world. In part of my dissertation work, for example, I compared groups of students having conversations in the face-to-face classroom with students having discussions in an online chatroom. To do this, I used two sections of one college class, which were taught by the same instructor. While one class section met online, the other section met in the classroom to discuss the same topic. The next day, the classes switched media. In doing this sort of quasi-experimental study, I was able to quantitatively compare discussions in each of these media while allowing some of the natural confounding factors that arise in real classroom situations (e.g., in multiple instances students temporarily left the online discussion to make a sandwich for lunch). I supplemented this work with interviews, observations, and design experiments conducted over multiple semesters in order to provide deeper insight into the media differences observed.

In doing this type of quasi-experimental research, I believe it's important to develop or appropriate technology to explore interesting phenomena. In studying chatrooms for educational purposes, I developed multiple versions of chat clients to support specific educational goals. In studying interruption, I helped develop a system that simulated interruptions. Since the ultimate purpose of my research is to support the design of collaborative tools and environments, technology development is an integral part of conducting these quasi-experimental studies.