Problem Based Learning on the Internet: a two year reportNikos Mattheos
Department of Periodontology, Centre for Oral Health Sciences, Malmö University, 205 06 Malmö, Sweden
Department of Periodontology, Centre for Oral Health Sciences, Malmö University, Malmö, Sweden
Department of Periodontology and Fixed Prosthodontics, University of Bern, Bern, Switzerland
Problem Based Learning (PBL) was first introduced in health education in McMaster University in Canada as early as 19691. This educational method has since then attracted an increasing interest among the academic health care community. The core of PBL philosophy is work in small groups, which go through 'thematic' modules under the guidance of a tutor-facilitator. Each session is developed around a clinical case, or 'problem'. The process of studying and 'solving' the problem follows six clearly identified steps (Table 1) that allow the students to reach the learning objectives by building their new knowledge, understanding, and skills into the context of what they already know. The first three steps are characterized by a high level of interaction and brainstorming, ending with the formulation of one or more hypothesis. The last three steps involve the collection, organization, and evaluation of information, a structured presentation of knowledge and opinions and the final testing of the hypothesis. In the undergraduate curriculum, each PBL module usually consists of two three-hour, in-classroom sessions separated by a week's interval. During the first session, the students work through steps 1-3. After the days of individual searching (step 4), the students conclude the module in a second session (steps 5 and 6).
Table 1. Six identified steps that will allow student to reach the learning objectives within the PBL study group (from Rohlin, Petersson and Svensater2)
Despite the spread of PBL in health education in-campus teaching, its adoption by Distance Learning (DL) course designers has not been similar. During the last two years, the Department of Periodontology, Centre for Oral Health Sciences, University of Malmö has been working towards developing and evaluating a PBL model for network based teaching. Among the major aims of this project is to provide quality distance education to oral health care students and professionals all over Europe, through simple and affordable technology. The pilot results of this research were recently reported in detail through a thesis3. The aim of the current report is to briefly discuss some aspects of network based PBL and some of the lessons learned so far. Further documentation, findings and details are available on the cited work3-6.
Overcoming the bandwidth problem
One of the first problems encountered when designing the Virtual Classroom for undergraduate students was related to the transmission of digital multimedia material (images, videos). Resource material for the course consisted of a large number of videos with sound, as well as series of high quality images. Technologies such as Two Ways Interactive Video, Microwave Television, Teleconference and several more7, have been employed and demonstrated positive results in the distant learning procedure. However for methodological and practical reasons a PBL virtual classroom could not utilise these demanding technologies. On the PBL methodology, a student should be able to interact as an independent unit and not as just a member of a remote group, which shares a camera. In addition, students and professionals scattered all over Europe, especially those in geographically isolated areas or developing countries, who are more likely to benefit from a distant learning course8, are not expected to have access to teleconference studios or high-speed connections to the Internet. A course addressed to students in various parts of Europe has to take into consideration the various limitations still existing in many countries, regarding bandwidth and communication networks.
The majority of the students participating in the course, used modem connections to the Internet. It is therefore reasonable to assume that, in worst cases the speed of downloading material over the Internet would vary from 1 to 3 Kb per second, depending on the time of the day and the network traffic. Bearing in mind that the discussion round a learning module might bring up the use of several videos, it is obvious that a different way of transmission had to be employed.
As the Virtual Classroom was built around web pages using relevant technology, we realised that simple HTML technology in combination with a specially designed CD-ROM might solve the problem of transmission of multimedia material effectively, thus allowing us to still maintain the Internet as the backbone of communication. All digital videos and images to be used during the course in the Virtual Classroom were included on a CD-ROM, which was mailed to the students participating. The material on the CD-ROM was arranged using an HTML address interface. The CD-ROM could not be used as an independent learning tool and students could not access the material in any structured way without interaction over the Internet. The CD-ROM was designed as an index, a digital 'storehouse' where videos and images were organised using HTML addresses. A special HTML 'index' file was prepared for the tutors. This way, the CD-ROM with the learning material was simply an extension of the network, a part of the interaction; able to assist the learning process in any direction decided at any stage. The use of the CD-ROM allowed the group tutor to use the videos during online (synchronous) discussions with the students, without any delay due to downloading times.
The designers of the learning programme have to predict in some way what will be requested from the students and the tutors during the course and include all relevant material on the CD-ROM. During the six-month course of the Virtual Classroom, we had little or no need for additional videos and images not included on the CD-ROM. On the contrary, some material was never used, as it was not accessed during topic discussion.
Figure 1. The tutor responds to the student's request with an HTML document placed on the server. The document contains a hyperlink to the relevant images and films on the student's local CD-ROM drive. The student has only to click on the hyperlinks provided to rapidly access the requested material and long downloading times are avoided.
In-classroom discussion versus 'chat'
Studies about the interaction in virtual environments are still limited, and few comparable results are currently available9,10,11. Some authors have noticed an increase in the amount of interaction during virtual classrooms compared to traditional classroom teaching12,13, even when employing highly interactive teaching methods such as PBL14.
Apart from the quantity of interaction, an increase in the quality and depth of interaction is also reported, as a consequence of employing text as a means of synchronous communication14. However, as most observations rely on subjective evaluations, the results presented so far are hard to duplicate, compare, or generalize, and therefore no safe conclusions can be drawn15. In contrast to the enthusiastic reports, several studies on various interactive media have pointed out lower levels of interaction, inferior results, frustration, or decline of the acceptance of the media among students16-19. As the medium becomes increasingly used, it is necessary to find ways to evaluate and measure both the quantity and the quality of the interaction. This will allow results to be compared and duplicated, and will allow researchers to build on the previous work.
One of the pilot studies was built around a course where classroom meetings were combined with online discussion. This fact offered the opportunity to study the interaction in both cases and identify possible differences in quantity, quality and functionality of interaction under the PBL environment. All synchronous discussions were recorded by the server. Recordings of in-classroom discussions were carried out in the respective groups. Measurements such as the total number of inputs, inputs per minute, and inputs per person were carried out both for the virtual as well as the in-classroom discussions.
One of the first findings was that interaction in the in-classroom PBL meeting was more than seven times faster than interaction in the Internet-based discussion. The fastest flow of discussion online was about 2.2 inputs per minute, in comparison to 15 inputs/minute in the classroom. In addition, students were more sceptical towards online discussions after the end of the course, with some of them mentioning that online discussions were some times frustrating and somehow superficial in comparison with the in-classroom ones.
The speed of interaction during synchronous face-to-face or computer-mediated discussions - the rate at which participants exchange meaningful communication - is a feature that was studied by several researchers in the past. 'Speed of interaction' or 'communication rate'20 or 'act rate'21 is calculated in several studies, commonly as 'communication units' or 'acts' per minute. The speed of interaction during educational discussions cannot be directly associated with the learning effectiveness of a session, however, it is observed that group performance typically involves complex trade-offs between quantity, quality, and speed. In addition, as we know from research results mainly in the field of speech therapy, a factor that significantly slows down the ability to interact has a major impact on the quality of the communication as well. Text-based computer-mediated communication is a particularly demanding cognitive task that requires the user to invest substantial processing resources in its attendance. According to Reid et al.21, this fact imposes a certain psychological and practical 'cost' on the user, which acts as a selective 'threshold' to communication. Whether or not a specific message is sent over a computer network depends on the degree to which its urgency and relevance exceeds this threshold21. The threshold imposed by text-based discussions in comparison to that in face-to-face interaction is, we believe, very well demonstrated by measurements in the speed of interaction. This threshold can also be the main reason for the students' attitude towards the asynchronous discussions observed in this study. The difference in the speed between network text-based and in-classroom interaction is indicative of how the medium restricts the flow of communication within the group. Media such as desktop videoconferencing and audio chat, when widely available, might significantly reduce this threshold, thus achieving higher functionality and acceptance by the students.
The speed of interaction observed in this study comes in agreement with previous findings10,21. We have to consider that the students in our case were not experienced in the media, and that web-based chat is not the fastest text-based communication possible. However, even with optimum chat conditions (e.g. expert users, good typing skills, fastest possible software and network), the results might not differ dramatically. It can be argued that online discussions, being text-based and not verbal, represent a more-structured way of interaction and demand a more careful expression, leading to a deeper understanding14. In this sense, computer mediated interaction might be beneficial when the methodology emphasises promoting students' ability to reflect, prioritise and better structure his/her communication. On the contrary, when a 'brainstorming' is needed, where students are expected to contribute fast and spontaneous thoughts, an in-classroom session might be more effective.
Network-based PBL, is it possible?
Within the limitations of the pilot studies of the last few years, it would be possible to conclude that a fully Internet-based PBL is possible and might be beneficial in comparison to didactic approaches in Internet-based teaching, if properly organised. However, such a model might constitute a compromise over the quality standards of in-classroom PBL, at least with the currently available knowledge and technology. A hybrid model, which will combine in-classroom sessions with interaction over the Internet, appears to be the safest and most beneficial approach right now.
In the future, the learning effectiveness of these models and Internet-based learning in general, has to be further evaluated on a research basis. In particular the student gains in terms of knowledge, skills and learning attitude have to be tested in comparison to the conventional in-classroom teaching methods.
CAL-laborate Volume 8 June 2002
Page Maintained By: PhySciCH@mail.usyd.edu.au