Evolving Use of A System for Education at a Distance Stephen A. White, Anoop Gupta, Jonathan Grudin, Harry Chesley, Greg Kimberly, Elizabeth Sanocki Microsoft Research Redmond, WA 98052, USA {stevewh, anoop, jgrudin, harrych}@microsoft.com, greg@gak.com; a-elisan@microsoft.com ABSTRACT Networked computers increasingly support distributed, realtime audio and video presentations. Flatland is an extensible system that provides instructors and students a wide range of interaction capabilities [3]. We studied Flatland use over multi-session training courses. Even with prior coaching, participants required experience to understand and exploit the features. Effective design and use will require understanding the complex evolution of personal and social conventions for these new technologies. features. Social conventions develop slowly and can vary; for example, who speaks first when a telephone is answered differs in different cultures. Such conventions are established, agreed upon, and then learned by newcomers. SYSTEM The system used in this study consisted principally of Flatland, a flexible, synchronous tele-presentation environment. NetMeeting, a commercially available collaboration tool that supports application sharing, was used primarily for presenting demos and exercises. Keywords Distance learning, multimedia presentations INTRODUCTION Distance education can help when classroom attendance is not possible or when students would like to participate casually while attending to other work. Desktop to desktop video permits presentations without requiring anyone to travel. Researchers at Sun Microsystems and elsewhere have conducted experiments with live multimedia presentation systems [1,2]. Audience members appreciated the convenience of attending on their desktop computers, but felt it was less effective than live attendance, as did the instructors, who found the lack of feedback disconcerting. Why the mixed response, given the benefits and convenience of distance learning? In standard classroom instruction, a flexible range of communication channels is available-visual observation, voice, expression, gesture, passing notes, writing on a board, throwing an object for emphasis, walking about to view student work. Even with years of experience, effective teaching is a demanding task. Knowing which communication or feedback mechanism to employ at a given moment requires understanding their effects and social conventions governing their use. Systems that support presentations at a distance mediate all awareness and communication digitally. Users must find new ways to obtain the information or feedback they desire and to compensate for lost information, and must develop social conventions and protocols to govern technology use. Past work has focused on single-session presentations. This paper reports a study of multi-session classroom use. Users of complex, collaborative technologies experience a learning curve as they experience the range and effects of Figure 1 – Flatland Presenter Layout Flatland presented instructors and students with four major windows and several additional channels to aid communication and coordination (Fig. 1). Audio and video (upper left) went from instructor to students. The upper right window held slides (and a pointer), either PowerPoint or instructor-authored interactive slides: students can make selections, which are tallied. A chat window is in the lower right and a question queue in the lower left. A student can support another student’s question with the checkbox. Other feedback channels are above the instructor image: a pop-up attendee list, a ‘hand’ with counter (e.g., “Everyone finished with the exercise please click on the hand icon”), and too-fast/ too-slow and clear/confusing indicators. Flatland details can be found in [3]. METHOD Two technical courses normally presented in a classroom were given without modification using Flatland. The first was two 3-hour sessions with 4 students, the second four 2hour sessions with 10 students. Both included live demonstrations via NetMeeting. Each instructor presented from a usability lab, permitting logging of activity; one student in each class was in a different lab room, the others attended from their offices. The instructors received a brief demo of the system prior to their first classes. Following each session, we distributed surveys and then verbally debriefed the instructor, and reminded instructors of unused features. RESULTS Flatland use changed considerably over sessions. Confronting an array of communication and feedback channels to replace those of classroom instruction, participants tried key features, grew comfortable with them, then experimented with more effective uses and new features. Some seemingly desirable approaches were not used even when pointed out. Students were generally positive about the system; instructors’ regard increased over sessions. Both groups suggested improvements. Changes in behavior and perception over sessions “Before it was just a feature. This time it was a tool.” —Instructor 1, discussing ‘hand raising’ after the 2nd class In his third session, Instructor 2 began to make heavy use of dynamically created slides and respond promptly to the question queue. In session 4 he began monitoring the chat window during the lecture. Both instructors steadily increased pointer use. In the second class, student chat discussion shifted from a focus on Flatland to the course, rising from 27% to 60% class-related. After each class, instructors were asked how well they thought they handled student questions. Instructor 2 responded 3, 4, 4, 5 on a 1-5 scale over sessions. Asked to what extent Flatland interfered, his responses were (no response), “too early to say”, 4 (high), 2 (low). Asked how distracting Flatland was, student ratings fell from 2.8 to 1.7. What drove changes in behavior over time? We identified two major factors: ambiguity about which channel to use for particular information, and uncertainty or incorrect assumptions by the instructor about the student experience, and vice versa. The widespread failure to appreciate the other side’s experience is due to lack of full understanding of features, to differences in equipment configurations, and to the lack of normal visual and auditory feedback channels relied on in classrooms. These factors are explored below. Uncertainty about appropriate channels “Once the instructor had answered it (a question), I didn't like the feeling of not being able to say ‘Thank you.’” —A student discussing the question window feature Confusions began early. The question window, provided for students to enter and vote on questions, went virtually unused in one first session. One student assumed this window was for the instructor to pose questions. The instructors saw the hand icon as a mechanism for students to respond to their queries. Some students used it to “raise their hand” to be called upon, only to go unnoticed by the instructor. Student ‘coming and going’ was reported in the chat window, drawing early attention to it. Students then used it to report problems, and then to ask questions. Once the lecture was underway, instructors concentrated on lecturing and stopped monitoring this noisy channel, so student questions then went unanswered. There was a tendency to respond via the same channel that one is queried. In his first class, Instructor 1 responded to early chat by typing in the chat window. In the second class he responded by speaking. Similarly, when instructors asked for clarifications to question window questions, students had to choose between typing a non-question in the question window or in the inactive chat window. Uncertainty about others’ context and experience Where the interfaces differed or information was incomplete, misperceptions arose. Students did not know how the Clear/Confusing, Too Fast/Too Slow indicators appeared to the instructor and whether they were anonymous, and did not use them. One instructor posted an interactive slide inquiring about exercise status unaware that it would not be visible under NetMeeting (being used to do the exercise), and did not understand the lack of response. Which students remained engaged was unclear. “Very little feedback from students… For some reason they seemed reluctant to give comments or participate. This should be explored as the teacher needs some kind of feedback…” NetMeeting: “worked very well from my end but would be very interested how it worked for the students.” – Instructor 1 “In a conventional classroom I have lots of visual cues as to attention level and comprehension that are missing here. For instance, if I say something and get puzzled looks, I repeat myself using a different analogy or different words or a different approach. Not so easy to do that here.” - Instructor 2 CONCLUSIONS AND FUTURE WORK The development of social protocols is a major challenge for the designers and users of distance learning systems. Groups may choose sub-optimal approaches, or approaches that will clash with others when membership shifts. Should designers try to build in or guide users to effective protocols? What will work best, or at all, in different situations? The results from this initial study should be followed by more extensive studies, repeated with different designs and user populations, to identify effective combinations of features, guidance, and flexibility. REFERENCES 1. Isaacs, E.A., Morris, T., & Rodriguez, T.K. A Forum For Supporting Interactive Presentations to Distributed Audiences, Proc. CSCW’94, 1994, pp. 405-416. 2. Finn, K.E., Sellen, A.J., & Wilbur, S.B. (Eds.). VideoMediated Communication, 1997. Erlbaum. 3. White, S.A., Gupta, A., Grudin, J., Chesley, H., Kimberly, G., Sanocki, E. A Software System for Education at a Distance: Case Study Results, Microsoft Technical Report MSR-TR-98-61, November, 1998