CMU-HCII-07-102
Human-Computer Interaction Institute
School of Computer Science, Carnegie Mellon University



CMU-HCII-07-102

Enhancing Technology-Mediated Communication:
Tools, Analyses, and Predictive Models

Daniel Abrahami

September 2007

Ph.D. Thesis

CMU-HCII-07-102.pdf


Keywords: Interpersonal communication, technology-mediated communication, computer-supported cooperative work, CSCW, computer-mediated communication, CMC, interruption, responsiveness, availablity, interpersonal relationships, predictive statistical models, task-switching

For most of us, interpersonal communication is at the center of our professional and personal lives. With the growing distribution of business organizations and of our social networks, so grows the need for and use of communication technologies. Many of today's communication tools, however, suffer from a number of shortcomings. For example, the inherent discrepancy between one's desire to initiate communication and another's ability or desire to receive it, often leads to unwanted interruptions on the one hand, or failed communication on the other. I have taken an interdisciplinary approach to address these shortcomings, and also in order to provide a better understanding of human behavior and the use of communication tools, combining tool-building and the creation of predictive models, with investigation and analysis of large volumes of field data.

At the focus of this dissertation is my research on Instant Messaging (IM) communication, a popular, interesting, and highly observable point on the continuum between synchronous and asynchronous communication mediums. I present the creation of a set of statistical models that are able to predict, with high accuracy, users' responsiveness to incoming communication. A quantitative analysis complements these models by revealing major factors that influence responsiveness, illuminating its role in IM communication. I then describe an investigation of the effect of interpersonal relationships on communication, and statistical models that can predict these relationships. Finally, I describe a tool I have created that allows users to balance their responsiveness to IM with their ability to stay on task.

217 pages


Return to: SCS Technical Report Collection
School of Computer Science homepage

This page maintained by reports@cs.cmu.edu