Developing Online Virtual Lectures for Course

Delivery: A Case Study and an Argument In Favour

Alan F. Smeaton

School of Computer Applications

Dublin City University


Dublin 9, IRELAND

Tel: +353 - 1 - 7045262

Fax: +353 - 1 - 7045442





In this paper we briefly describe how we replaced traditional lectures by an online counterpart which had audio and synchronised visuals for students to "play" in an undergraduate course driven by student-directed learning. Previous analysis of usage and exam performance has shown that our mode of virtual lectures behaves just like traditional lectures with good study habits or technical know-how not necessarily being rewarded by better exam performance. We present details of an end-of-course survey we carried out on 115 students who have completed the course and the exam, and we present results of that survey. We also discuss the costs associated with creating our virtual lectures and we argue that the virtual lecture mode of online delivery of material, as opposed to a hypertextual one, or one based on concept maps, is at least as attractive as other organisations because of the comparative ease of the task on the lecturer yet we still provide numerous benefits to the student in terms of access. Our organisation of material uses the principles of deconstruction of material into components but not the re-construction into a concept map or network and the argument of this paper is that the comparative ease with which this can be done on the part of the lecturer may encourage more teaching staff to become involved in making course material available online.

Acknowledgement: The work reported in this paper was partially supported by the Dublin City University Educational Trust



1. Introduction

It is acknowledged that increasing costs, rising student numbers and the convergence of computing and communications technologies are creating pressures on Universities to transform their mode of operation (Gaines & Shaw 97). This transformation would seem to depend on the use of information technology (IT) in teaching but at present, much of the use of IT in teaching is at the pilot level with a great and varied amount of experimentation but no consensus as to best practice. Even when that holy grail of "best practice" is discovered we will find that teachersí attitudes will be a major obstacle to the really widespread introduction of change in the way we use IT in our teaching. Some of this stems from the belief that using IT may actually increase rather than reduce the gap between student and teacher. Most of it comes from the recognition that creating and maintaining online course material is hugely demanding on lecturersí time.

In this paper we present our paradigm for using IT in teaching called "virtual lectures" which we believe are scaleable from pilot to operational projects. The paper presents our approach and the results of a survey carried out on students who completed a course using our virtual lectures only. We also discuss the costs associated with our model and argue for the many benefits it brings to the student while still remaining cost effective.


2. Virtual Lectures for Online Course Delivery

We have taken a third year undergraduate course in databases, covering topics like the relational model, SQL, database design and database implementation issues, and broken down the material into 119 bite-sized chunks or topics. For each of these topics we have recorded an audio track of 10 minutes on average and with each audio stream there are a set of visuals, equivalent to OHP slides or drawings on a blackboard. Traditional lectures for this course were cancelled and students were pointed in the direction of the online material, audio and synchronised visuals, to take their "delivery" of the content. Students access to the material was unscheduled in that it was on their own time, from campus, home or workplace, whichever was the most convenient and students were allowed to "play" and listen to as much of the course as they needed, and to repeat any parts of their virtual lectures as often as they wanted.

To navigate through the course we provided three orthogonal ways; a traditional table of contents could be used where the overall course was broken into chapters which were in turn broken into sections and sometimes sub-sections and sub-sub-sections. Each of these leaves of the hierarchical table of contents pointed to an audio track. We also created a back-of-the-book index of topics where each topic pointed to the place or places within the course, i.e. in mid-stream, where that topic was discussed. Finally, we allow users to search the audio content in such a way that the keywords they type in are searched for throughout the audio track and the parts of the course which discuss the topic of the userís query are ranked and displayed for the user to choose which to play. The latter two navigation mechanisms are useful for revision but the primary navigation is by the table of contents. The way in which the virtual lectures was created is described in (Smeaton, 1997a).

In our implementation of virtual lectures we have been able to track individual student usage of the system by issuing individual IDs and passwords. From the logfiles recording accesses to the system by 115 students we have been able to isolate individual student usage and when these were regressed against performance in a written end of course exam we found that there was no correlation between usage patterns and student performance. In another analysis of our virtual lectures we measured each studentís technical know-how and these measures were also regressed against exam mark. We found no significant correlation between a studentís technical inclination or bias, and exam mark. These results, reported in (Smeaton, 1997b) tell us that our virtual lectures are just like traditional lectures in that neither good study habits nor technical know-how necessarily imply good exam performance.

Some work similar to ours has been reported by (Anderson, 1997) and by (Dankel & Hearn, 1997) who also have made lecture material available online and a more elaborate online lectures is described in (Benest, 1997). Our work goes further than these as we use the audio and synchronised visuals as the only delivery of material and in addition we have tracked individual student usage and we have performed detailed analyses reported here and elsewhere (Smeaton, 1997b).


3. What Did Students Think of the Online Course ?

Part of our evaluation of our virtual lectures included interviews and questionnaires for students. One of these questionnaires was completed at the end of the course, just before the exam was taken. This was followed up by a focus group meeting of 30 minutes duration with up to 6 students at a time to elicit further feedback and gather comments. The questionnaire was composed of questions, each of which was answered on a scale of 1..7 with some reverse semantic differentials. The questions and the responses are shown in Table 1 and for each of these, the sample size (N) is 115.





1 (yes/true) .. 7 (no/false)

1. Course was harder than others



2. Course was very interesting



3. Continuous assessment was really hard



4. Course was badly organised



5. I work at my own pace



6. I always work with others



7. VL is a solo experience



8. As I got busy, VL were postponed



9. No technical problems with VL



10. Find VL efficient way to learn



11. I would like to take more than 1 course as VL



12. VL are less stimulating



13. If either available, I would use VL only




What do the results mean ? The standard deviation figures show that there was most agreement among respondents to the question about postponing virtual lectures when other deadlines arose (Q8). A usage analysis confirmed this result (Author, 97b). The next most agreed upon question found that the course was interesting (Q2), was not harder than others (Q1) and the continuous assessment exercises were not too difficult (Q3). Students most definitely found VLs to be a solo experience (Q7) though most work alone and at their own pace (Q5) and less so with others (Q6). Students found VLs to be effective (Q10) and neither navigation (Q4) not technical problems (Q9) were an issue. Virtual lectures were slightly less stimulating than traditional (Q12) but students would still prefer them (Q11, Q13).

All things considered, the results of this survey of our virtual lectures delivery was very encouraging.


4. Costs Associated with Virtual Lectures

The costs associated with the development of virtual lectures as we have developed them, fall into a number of categories. In our application, students had the option of using PCs on campus which had multimedia capabilities and were wired to the campus LAN. For students working in a distance education mode, client-side computing costs would require such a machine with audio playback facilities and an internet connection. Because the RealAudio server streams its data from server to client transmitting only enough data to keep the buffers for the player software full, and because of the high compression achieved with RealAudio, the bandwidth requirements to deliver our audio are low and can be handled with a modem rated at 28.8kbps and an internet service provider. The bandwidth allows the transmission of HTML pages to the userís web browser at the same time as the audio is playing, without loss of sound quality which is somewhere between AM and FM radio. Thus during the first year of operation many of our students chose to take the virtual lectures from home or from their workplace, without loss of sound quality. Apart from a machine and a network connection, the software required by the student is a web browser and a RealAudio player, downloadable over the internet and which is free for personal use.

On the server side, the costs are for a machine to run the RealAudio and WWW servers, and the cost of the server software itself. The specification of this machine is surprisingly lightweight; although we used a mid-range UNIX server a much less powerful machine would have sufficed and we were able to put all the audio, HTML and image files into about 150 Mbytes of disk space.

The big costs involved in the preparation of virtual lectures are labour costs, both the technical support and the time of the lecturer involved in the exercise. In our case we had one full-time research assistant for a 12 month period who post-processed the audio, managed the servers, marked up the visuals into HTML and generated graphics, administered user accounts for students and handled technical queries and userís problems. A great deal of the time spent on supporting the virtual lectures was devoted to finding the best way in which to do things, sourcing the software to record the audio, determining the best sampling rate and the way in which to mark up the HTML, designing the layout of the visuals, etc. Much time was also spent on developing the software to perform the search function and to mark up the index pointers. In retrospect we believe that for navigation purposes the table of contents is essential and is the primary navigational tool, the index is a nice idea which is good for revision, and the search is a bit of a luxury given the resources consumed in creating it vs. the amount of use it actually received. From a technical standpoint we have learned a lot from the experience of creating virtual lectures the first time and would be much more streamlined for a repeat of the exercise.

The final cost associated with developing our virtual lectures was the lecturerís time and this is the main point of this paper which we now discuss.


5. The Case for Online Virtual Lectures

We believe that the virtual lecture mode for online delivery of material, as opposed to a hypertextual one or one based on concept maps, is at least as attractive as other organisations because of the comparative ease of the task on the lecturer yet we still provide numerous benefits to the student in terms of access. Our organisation of material uses the principles of deconstruction of material into components but not the re-construction into a concept map or network

There is a swing towards instructional design principles and new teaching pedagogies that information technology now offers which involves deconstructing course content into atomic concepts and turning these concepts into concept maps which in turn forms the basis for a hypermedia or hypertextual organisation of material. Performing such an exercise is very difficult and time-consuming for academic staff. We know this because we performed such a task in 1990 in the pre-WWW days when there were absolutely no tools available for supporting such an exercise (Smeaton, 1991). Our experiences were that the task is hard and subsequent work by others has underlined this fact. Even now, with web creation tools emerging to make the task a bit simpler, it is hugely time-consuming and this acts as a disincentive to others from becoming involved. the argument of this paper is that the comparative ease with which virtual lectures can be done on the part of the lecturer may encourage more teaching staff to become involved in making course material available online.

Another reason preventing more use of IT in teaching is that when it comes to embracing information technology in teaching there is very little motivation for an academic to become involved in the process because there is little reward for what is undoubtedly an enormous effort, there is little support staff four course design and in academic circles research work is perceived as being a better activity. We do not claim that VLs deliver a totally learner-centered environment but only that it is a step in that direction

Our virtual lectures as described here are much easier and more comfortable for the lecturer and for the student as the concept of a lecture is embedded in our consciousness.



Anderson, T. (1997). "Integrating Lectures and Electronic Course Materials", Innovations in Education and Training Internationl, 34(1), pp.24-31, 1997.

Benest, I, (1997). "The Specification and Presentation of On-line Lectures", Innovations in Education and Training Internationl, 34(1), pp.32-43.

Dankel, D.D. and Hearn, J. (1997). "The Use of the WWW To Support Distance Learning Through NTU", in Proceedings of ACM SIGCSE/SIGCUE, Uppsala, Sweeden, pp.8-10.

Gaines, B.R. and Shaw, M.L., (1997). "Institutional Transformations to a Learning Web", in Proc ED-Media/ED-Telecom, Calgary, Canada, pp528-536.

Smeaton, A.F. (1991), "Using Hypertext for Computer Based Learning", Computers and Education, 17(3), 173-179.

Smeaton, A.F. and Crimmins, F. (1997a), "Virtual Lectures for Online Lectures:Delivery Using RealAudio and the WWW", in Proc ED-Media/ED-Telecom, Calgary, Canada, 1997, pp.990-995.

Smeaton, A.F. and Keogh, G. (1997b), "An Analysis of the Use of Virtual Delivery of Undergraduate Lectures", Computers & Education (submitted for publication).