Team-Teaching and Team-Learning on a Global Scale:

Insiders' Account of a Successful Experiment

by Debasish Dutta (Professor, UM College of Engineering, Mechanical Enginnering Department, Ann Arbor, MI 48109), Vlad Wielbut (Technology Resources Manager, Alliance for Community Technology, Ann Arbor, MI 48109)

Abstract


This paper will show that it is possible for 3 university professors in 3 different countries to  teach effectively (as a team) with the help of videoconferencing and the Internet, providing their  students (also distributed  among the 3 locations) with valuable experience of working in global, distributed, virtual teams for designing globally marketable products. We will discuss the technological infrastructure of this course and share the results  of a survey of its participants (with some surprising results). We will also describe problems likely to occur in this setting and ways of  dealing with them.

Table of Contents:

1. Introduction
After months of preparations and multiple travels between Ann Arbor (USA), Delft (The Netherlands), and Seoul (South Korea), the course "Global Product Realization" was launched in September 2000 and concluded 3.5 months later, in December. This course, first of its kind at any of the participating institutions, and probably one of only very few of its kind in the world, proved that it is possible for professors in three different countries to team-teach a group of students located in three very distant places, and make it feel as if it were all happening in a single classroom. (It proved other things as well, but we'll point those out later in this article.) Speaking from a still very fresh perspective, it was a success on many levels:
  • technological: with various sophisticated technologies working fairly seamlessly together, without major failures
  • educational: with students producing interesting ideas and models, largely on time, despite the challenge of working in a new environment of multinational, virtual teams
  • logistical: with skillful navigation through different educational calendars, different sets of requirements, and varying levels of technical support
  • perceptual: with a great majority of students considering participation in this course a highly valuable experience and willing to take more courses like this in the future
Probably a few more. There were, of course, mishaps and things we'd like to do better next time around, and we  have no intention of removing them from our account of this generally very successful course.

In this paper, our primary focus is on the technology that enabled the team teaching and team learning on a global scale. Details on the course content will be limited to the extent necessary for providing the reader an overview of the whole effort. While industry has been forced to deal with globalization issues since the early 1990s, product development for global markets is a new topic gaining attention in the academic world. This course is our first attempt to focus on the topic in a classroom setting. Needless to say the content and format are still evolving. However, collaborative engineering accross time zones and geographic regions lies at the core of this new area and provided the motivation  for us to create a new (virtual) global classroom.

2. Preparations [TOP]
Preparations were kicked off in late Fall 1999, with a meeting between Debasish DuttaFootnote1, Jongwon Kim (professors in Ann Arbor and Seoul, respectively), Joseph Hardin (Deputy Director of the University of Michigan Media Union), and Vlad Wielbut (the "chief technologist" at the Alliance for Community Technology). That is when we discussed, in very practical terms, the idea of an educational experiment involving engineering departments at the University of Michigan, Technical University of Delft, and Seoul National University. The idea was this: three professors from the three universities wanted to offer a graduate-level course in which students would learn how to design products for a global market, and where the learning itself would happen in a global environment. The question that the meeting was to answer was: Is it possible, with today's technology but without NASA's budget, to connect these three professors and their students within a single, "virtual" classroom, without losing any of the benefits of a "physical" classroom: spontaneity, eye contact, rich interactions, visual cues, etc.

The answer to that question was a cautious "yes". Cautious, because it was quite clear that some kind of video communication will be necessary, and our experience in this area was limited; we had only a vague idea of the costs involved or of the technical hurdles we would have to jump over. Not to mention some serious, long-standing misgivings about the value of video communication in online learning and collaboration. However, seeing an opportunity to fill a few holes in their knowledge base, Joseph Hardin and Vlad Wielbut decided to investigate the options and propose a tool set that could help us come as close as possible to the vision of the professors. Two more  factors contributed to our  willingness to initiate this challenging project: 

  1. the genuine enthusiasm we sensed in the teachers, who must have clearly understood the challenging nature of the endeavor they were embarking upon
  2. the fact that it was much more visionary than merely putting yet another course "on the Web". While many courses can be (and are) taught in a distributed environment with the help of Internet technologies, the real value and transforming power of the Internet shows in situations, where the distributed environment  itself becomes an educational asset. One example of this is the Global Graduate Seminar taught by Prof. Derrick Cogburn (UM School of Information), where students learn about globalization while experiencing globalization first hand through the virtual, distributed classroom they find themselves in.
The room where the course would take place (on the US side) was already equipped with a several-years-old system from PictureTel, but it quickly became clear that in order to use it, we would have  to acquire services of a commercial company to bridge (connect) our three sites. The cost of such a service, at around $30/minute would make this prohibitively expensive. Given that the other 2 sites did not yet have videoconferencing systems in place,  we saw this as an opportunity to upgrade our own equipment with technology that would not became obsolete overnight.  Hence, in the following couple of weeks we evaluated a number of videoconferencing solutions, to finally settle on Polycom's Viewstation MP. This decision was based on the following:
  1. Reasonable price. The model that would suit our needs perfectly had a sticker price of ca. $12,000.
  2. Support for IP videoconferencing. We assumed that the only way to avoid high cost of ISDN connectivity would be to connect our sites via the public Internet, or better yet Internet2, if possible.
  3. Compliance with the H.323 standard, so that other compliant clients (including Microsoft Netmeeting) could communicate with our equipment
  4. Bridging capability. The unit was able to connect up to 3 sites without costly third-party involvement.
2.1 ISDN versus IP Connectivity [TOP]
With this out of the way, the decision as to the type of connectivity (IP or ISDN) had to be made. Some of us were leaning toward IP, believing ISDN to be a technology on the way out, and way too costly for what it was capable of delivering. Unfortunately, Polycom MP does not bridge IP connections, so a third-party was still needed. Through sheer accident we found out about an ongoing experiment in distance education via IP videoconferencing between three universities in the United States and promptly contacted the people running this experiment. It turned out that a consortium of American universities purchased a sophisticated bridging equipment, now located at Ohio State University, for the members of the consortium (including the University of Michigan) to make use of at no cost. It was just a matter of giving the equipment's operator IP addresses of all the sites to be connected.

Network throughput between Seoul, Delft, and Ann Arbor was tested, with results ranging from inconclusive to good. Alas, the fear of instability and bandwidth fluctuations in the public Internet prevailed, and the decision was made to play it safe and rely on ISDN connectivity instead. (There was also concern, that we would suck up too much bandwidth from our respective university networks, in effect slowing them down or disrupting their operations.) In retrospect it probably was a smart choice, which rewarded us with problem free connections for most of the sessions. However, we still hope to run a course like that via the Internet soon, to prove that it is, indeed, a very capable medium.

Part of the thrust behind the ISDN decision was the fact, that by using 4 phone lines at each  location we could bridge them at quite sufficient - if not  impressive - speed of 256 kbps. Given the fairly low cost of making international phone calls from the US, our connectivity cost per minute would be a moderate $2.44 This was a spare change when compared with third-party bridging (although, in the end, it would still turn out to be a very costly proposition - see Appendix A.)

In most big corporations, including large universities, the path from making a decision (e.g. to buy a piece of equipment) to having the decision implemented (e.g. actually getting the piece of equipment) is long and arduous one.  There are forms to fill, signatures to collect, accounts to charge, procedures to follow, budgets, etc. Needless to say, our Viewstation MP arrived 3 days before the scheduled start of the course, not early enough for serious testing or even getting really familiar with it. On the bright side, this provided us with the opportunity to test Polycom's claims regarding their products' ease of setup and use. We must admit that these claims turned out to be largely true. We were able to establish a bridged session within hours of opening the box.

2.2. The Virtual Classroom [TOP]
Obviously, videoconferencing was not the only technological challenge we faced. Besides seeing their professors and colleagues, the students would have to receive and submit assignments, work in distributed teams, communicate frequently outside the teleconferencing room, access course materials, etc. However, this was not our first foray into the world of online learning, and so the decisions in this regard were made quickly, based on our knowledge and/or ownership of good tools to support activities in question.

Blackboard CourseInfo was selected to provide the "virtual classroom" for the course, partly because our Delft partners already had the server up and running, but partly because it is, indeed, one of the best tools in its category. In this very easy to use package of seamlessly integrated components students and instructors would find: discussion boards, e-mail system, bulletin board (announcements), space for course materials, personalized Web pages, directory of participants, whiteboard (with collaborative browsing capability), chat rooms, and more.
 

Course materials
Announcements
Discussion board
Whiteboard/Chat

While Polycom supports an impressive list of input sources (microphones, video cameras, VCRs, document cameras, computers), which enables sharing almost any type of material, including PowerPoint slides, only one source can be active at any given time. When images from a computer are being sent, live video is not, and vice versa. Since lectures with visuals were to be a main part of every class session, videoconferencing would be reduced to sending static images and speaker's voice to participating sites. This was clearly unacceptable. Therefore, Placeware Conference Center (provided by ACT) was to supplement videoconferencing with its slideshow capabilities. Placeware, a Web-based conferencing tool, would allow three sites to synchronize the PowerPoint slides during a lecture, freeing the Viewstation to serve its primary purpose, i.e. send and receive live video and audio. Placeware itself would be projected independently but in parallel with the videoconferencing.

2.3 Collaborative CAD [TOP]
While each school had a variety of CAD systems for its students to use, there was a need to establish a collaborative CAD environment. With a view towards ease of use a suggestion was made to offer students access to eVis, a fairly sophisticated tool that would allow them to collaboratively view and manipulate 3D objects (from CAD applications) in real time, over standard Internet connections. eVis is provided through the ASP model, so no investment in hardware and server software was needed - all that was required was purchasing appropriate number of relatively expensive user licenses from the provider, and installing client software on multiple machines at participating institutions. Admittedly, this was somewhat risky proposition, as it brought the number of various technologies to be used in the course to the point of oversaturation. Although we have no hard data to support this assertion, there is anecdotal evidence suggesting that introduction of more than 3 distinct technologies in a course (online conferencing, discussion boards, document sharing) is more a detriment than help, and will result in less effective use of these technologies. A course overloaded with tools runs the risk of becoming a course about these tools, regardless of what the initial theme might be.
(Click on the thumbnails below for screenshots of e-Vis.)
 
e-Vis project binder
e-Vis 3D model viewer
3.0 The Course [TOP]
Forty-eight students, 16 at each of the participating schools, took this experimental course in the Fall semester 2000 (September-December). All were graduate students,  but from various disciplines. The intent was to form cross-disciplinary teams for the project. The students in this class came from mechanical enginnering, manufacturing engineering, industrial design, and MBAs from the business school. The topic of the course was: "Global Product Realization", with the stated objective of providing the students with the experience in:
  • Applying engineering principles to solve an open-ended global product design and manufacturing problem (developed by the faculty in consultation with industry);
  • Working as an international team using the worldwide network on a globalized engineering design and manufacturing problem;
  • Providing a practical solution to the globalized problem keeping in mind performance, safety, cost, environment considerations as well as the each country's constraints such as tariff, law, marketing characteristics and so on.
The course's schedule called for two ninety-minutes lecture sessions each week, with two midterm project reviews, final project presentation, and final exam (Delft students only). Lecture sessions were held in designated rooms on the 3 campuses, each room equipped with a Polycom Viewstation unit, a monitor or a display for the Viewstation, projector and screen. The format of the lectures was standard: a speaker in front of the classroom (many of those speakers were invited experts  from the industry or from other departments) and a set of PowerPoint slides. Each lecture was followed by a brief discussion, during which students from all three locations would ask questions of the presenter or offer comments on the subject. The lectures - voice and slides but not video - were recorded using Placeware's recording capability and published on a Web site right after each session, for on-demand playback.

In this course we adopted a case-study approach to teaching (a format common in business schools). Several case studies were developed by the faculty in cooperation with companies like Ford Motor Co., Samsung, Philips, Steelcase, etc. Each case desribed the development of a product that was to be marketed in different regions of the world. In varying depths, each case highlighted new issues that product development teams in these companies had to consider. These issues included patent law differences in various parts of the world, product liability considerations in global markets, product pricing and supply chain management for global firms, trade and tariffs, etc. In addition, importance of product centric topics such as DFE, DFM, material selection, QFD, etc., were also reiterated in the case studies. Finally, the importance of team work, collaboration across timezones and cultural differences were highlighted by almost every case study.

The course content was structured into four modules. The introductory module dealt with the new competitive pressures on industry brought about by the global economy. Trends and discontinuities as a result of the Internet were also discussed. The introductory module set the stage for the course by reaffirming the importance of considering relevant global issues in product development. The remaining three modules dealt with product design, manufacturing, and logistics respectively.

3.1 Semester Project [TOP]
After the first session students were divided into 8 teams, with 6 members each, 2 from each country. The teams were to collaboratively develop a prototype of a coffee maker that could be marketed in three countries (Holland, S. Korea, and USA) with minimal or no modifications. Besides designing a 3D model of such an appliance in a CAD system, the teams were also responsible for: researching patents that might be involved in the proposed product; choosing the best location for the manufacturing plant (considering the costs of materials and labor, shipping costs, existing infrastructure, etc.); researching cultural preferences in coffee consumption; estimating the size of potential market, etc. A really complex project that, given the relatively short span of available time, required frequent and efficient communication within teams.

Recognizing the importance of communication, the teams were offered the opportunity to sign up for 30-minute videoconferencing sessions held before and after each lecture throughout the semester. Due to the high cost of videoconferencing, the initial idea was to offer this only for one or two introductory, "getting to know each other" sessions, but it quickly became apparent that students considered this an important tool for communication. However, these sessions were not sufficient and the teams relied heavily on textual chat and asynchronous discussion boards, both available within the Blackboard environment. We hoped that students would use eVis for real-time discussions of their CAD models, but they found it easier to work with drawings and diagrams which they exchanged as image files via Blackboard, reaching for CAD systems only shortly before the final presentation.

3.2. Colocation and Course Finale [TOP]
The global teams formally presented their work to the entire class twice during the semester, in a format similar to lectures, with PowerPoint slides and videoconferencing as the connecting medium. These reviews also included a written report by the team. Grading of the reviews was done collaboratively by the faculty, and each team received appropriate feedback. Before the end of the semester, students from Seoul and Delft arrived in Ann Arbor to meet their teammates in the "physical space", put final touches on their projects, and present them first to their classmates and later to the public in a publicized poster session on campus. This was different from most other experiments in virtual, distributed teams, where the team members are brought together either at the beginning of the project or during it, instead of (or in addition to) meeting at the end. This approach was not intentional but imposed by logistics and academic calendars of the participating universities.

On the last day of the week, the teams gave their final project presentations. Several people from industry attended at the invitation of the faculty. The final project presentation was accompanied by the final report submission. In the afternoon, in the GPR Exhibit, students presented their project prototypes and, using three poster boards per team, described their product development process. The exhibit was open to the public and provided a perfect forumto highlight the importance and benefits of global connectivity. In conjunction with the Exhibit, a Global Education Forum was also organized. A panel of speakers, including consular representatives from the Dutch and Korean embassies , remarked on the importance of globalization and the need for students to experience the benefit of working in global teams prior to entering industry. 

4. Lessons Learned

4.1 Connectivity & Hardware [TOP]
Being involved in this course from its conceptualization to the farewell dinner gave us an opportunity to observe it throughout and from several perspectives: that of a student listening to a lecture; a team discussing task assignment via videoconference; a technical support person; a guest speaker; a course coordinator. We diligently made mental notes of things that went wrong; things that could go wrong but didn't; things that could be done better the next time around. Of course, we also noted all the positive aspects, and there were plenty of those. It is probably safe to say that, overall, the experiment went better than expected. Despite the use of sophisticated technology, the glitches we feared were few and far between. The most serious were occasional difficulties connecting the three sites, most likely due to problems with ISDN lines (beyond our control), but these were usually resolved through repeated dialing attempts. Less serious but still quite bothersome were fluctuations in sound quality, ranging from mild to severe, and caused by a variety of factors, some easily solvable (students speaking too softly for the distant microphone to pick up their voice), some beyond our control (workers demolishing a building adjacent to the one housing the videoconferencing lab at Delft), and some totally perplexing (mysterious metallic sound, probably caused by some sort of electromagnetic interference, whose source we were unable to determine.) Appendix A contains a fairly complete list of technical and non-technical problems that appeared during the course.

On the positive side, the objectives of this course were achieved fully and on time. Global, virtual teams of widely dispersed students produced impressive results, learning a great deal in the process. Students gained experience in working in an environment most of them haven't encountered before, but are likely to encounter in their future professional lives. An impressive array of experts from all the participating countries gave guest lectures for the course - a treat nearly impossible to re-create in a traditional course; these lectures went beyond the immediate problem at hand (i.e. designing a globally marketable coffee maker), exposing students to important concepts in areas such as: patent law, online collaboration, design challenges in the automotive industry, etc.

4.2 Course Evaluation [TOP]
After the course had ended, we asked the students to take part in anonymous, online survey composed of 20-questionsFootnote 1 aimed at extracting their opinions about the course. Several surprises awaited us when the results of the survey poured in. One was the response rate: of the 48 students in the course, 36 (75 percent) filled out the questionnaire, despite having no incentive for doing so.  Another one was the fact that 81% of respondents stated that the global team approach - which must have posed significant challenges for the students - "added tremendous value to the course". Yet another one was the overwhelmingly good opinion students had about the course and its novel format:
  •  81% of  respondents would participate in a similar course (6% would not, 13% wasn't sure)
  • All but one would recommended this course (as it is) to a friend
  • None of the respondents participated in a distance education course before, and 44% had  no idea of the potentially demanding nature of such a course (36% did have an inkling), yet 86% would still enroll in it.
  • Most lectures given in the course were considered useful by 64% of students. All lectures were deemed useful by 8%, while the remaining 28% perceived only some lectures as useful.
These positive opinions are all the more satisfying that the course required more than average effort: 56% admitted that they had to spend much more time studying for this course than for others, while 31% had to work a little harder than usual. This effort apparently wasn't wasted: all students claimed that the course changed the way they saw themselves and/or the world afterwards.

Somewhat surprising was also how the various technologies were rated by the respondents: 64% rated videoconferencing as "very useful", 25% as "somewhat useful", and 11% as "not very useful". (This seems contrary to the widely held opinion of live video as unnecessary frill that adds little value to communication). Blackboard was considered "very useful" by 83% of respondents; no surprises here - this is, indeed, a well designed virtual classroom environment. Placeware fared less well: the same number of people (17%) considered it very useful and not very useful, while to 42% it was somewhat useful and 22% didn't have an opinion. This was also hardly surprising, considering that Placeware was used in a rather limited fashion in this course and that students did not have opportunities to interact directly with this tool. eVis received very negative ratings and its inclusion in future iterations of the course will likely be re-evaluated: 36% viewed it as not very useful, 33% did not have an opinion, and 17% complained that it was "more a problem than help".

(The complete results of the survey are available online.)

4.3 Closure [TOP]
We were very gratified to receive the positive evaluation from the students. For all faculty and staff involved in the course it was an enriching experience. So much so, that at the time of this writing we have begun planning for the next offering of this course in Fall 2001. Some changes are being planned in the content, structure, and delivery, with a view towards improving student experiences and enhancing the collaborative environment.
Appendix A: Problems likely to occur and ways of dealing with them [TOP]
  1. Cost of connectivity. Having avoided the enormous expense of bridging 3-way calls via a commercial service, we were still responsible for paying the cost of  calls to the other sites. Years of tough competition contributed to significant lowering of per-minute charges for international calls originating in the US, but using 4 ISDN lines meant running 4 concurrent calls to Delft and 4 to Seoul. Even at the moderate price of $0.15 and $0.46 per minute per call respectively, these costs quickly add up to a $2.44 per minute, $146.40 per hour, ca. $900 per week (including individual team sessions). Luckily, there are several ways of reducing costs of videoconferencing:
    • Connect via public Internet or, preferably, Internet2. (Polycom Viewstation can handle both ISDN and IP connections.)
    • Limit use of videoconferencing to lectures. In the course described here, student were offered two 30-minute sessions before and after each 90-minute lecture to meet in their respective teams and discuss their projects. This increased the total connectivity time (and thus cost) to over 6 hours per week. Such meetings could have been substituted with a different technology (e.g. Placeware), without detriment to the work performed by the teams.
  2. Academic calendars. Three universities in three different countries are very likely to start and finish their semesters on different dates. When differences in scheduled holidays and breaks are added to the mix, this can easily cut 3 or more weeks out of the total available time. There is no simple fix for this problem. One can only try to use the remaining time as efficiently as possible without overwhelming the students with frantic pace and heavy load of material - not an easy thing to do in a pioneering environment that is dependent upon sophisticated technology without a perfect track record.
  3. Inadequate use of technology. On several occasions we observed students in a meeting trying to show their drawings or notes to others by displaying them to the videoconferencing camera; given the low resolution of such cameras and hardly optimal lighting, these attempts were a waste of time.
  4. Language. English was the official language of the course, putting some of the students at a disadvantage. Although proficiency in English was one of the requirements for enrollment, students' abilities in this regard varied greatly. A number of Korean students had significant difficulties with spoken English, somewhat compensated with higher proficiency in the written word. The scope of this problem can be reduced by careful screening of students' proficiency in the English language and more emphasis on using text-based communication (chat rooms, discussion boards, e-mail) as opposed to video- or  audioconferencing.
  5. Old habits die hard. Every new technology, in order to be used efficiently, requires adjustments in behaviorFootnote 2. Videoconferencing is no exception. First and foremost, we have to remember that we are in front of a camera; this is particularly difficult in situations when we are speaking to local and remote audiences at the same time. Second, we have to account for the fact that a camera "sees" things differently than human eye. Most common mistakes include:
    • Moving around while speaking, to the point of walking out of the view of the camera,
    • Speaking in front of, or next to a brightly lit background. Positioning oneself in front of a display screen makes one appear to the remote audience as  dark, faceless silhouette. Moreover, the projected image is nearly unreadable to the people on the other side of the videoconferencing link, especially when detailed diagrams or dense text are involved.
  6. Poor lighting. Most videoconferencing facilities at institutions of higher learning are simply classrooms or labs "adapted" to their new role by putting the videoconferencing equipment there, without making adjustments in lighting and sound environment. This leaves to the participants the responsibility for finding the delicate balance between more light (which improves picture quality for the remote audience but makes anything displayed locally less sharp) and less light (which has the opposite effect, favoring the local audience).
Appendix B: References [TOP]

Footnotes:

1:  Preparations for this course actually started even earlier (around March 1999), when the idea for this new course was conceived by Deba Dutta. Over the next several months he discussed the idea with many companies and was reassured about the need for a course on this topic.
2: Special thanks to Derrick Cogburn for allowing us to adapt a part of his survey for this purpose.
3: A good example of that are the different strategies for efficiently drawing a series of arched windows using pen, pencil, compass, and ruler in one case, and a drawing program on a computer in the other. When drawing on paper, one would likely start with a series of half circles drawn with a compass. Next would be the vertical, parallel lines flowing down from the half circles. Finally, the horizontal "bottoms" of the windows would be drawn. In contrast, most efficient strategy for achieving the same result on a computer would be to draw a single instance of a complete window and then copy and paste it several times.


©2001 Vlad Wielbut and Debasish Dutta