However, it is important to ensure that the quality of the learning does not decline. So we set up a controlled classroom experiment, in which the students of an Information Society module did half their seminars face-to-face, and half over a computer conferencing system. In each week, some of the seminar groups had face-to-face seminars, while others went in each day to our computer lab, logged on the our Network Telepathy computer conferencing system (Ashmount Research 1992), looked at the own group topic, and left comments on the subject being discussed. Over two weeks the comments accumulated as the discussion continued. The face-to-face seminars were held for one hour, with groups of 10 to 20 students. The lecturer used the same approach in both, rather than trying to adapt his style to each medium, so this was a comparison of average, rather than optimised, use of each technology.
Figure 1. Browsing an on-line tutorial on Network Telepathy
The main purpose of holding seminars in Information Society was to encourage deep learning approaches among the students, in which they achieved an in-depth understanding of the subject, rather than a surface learning approach which helps pass examinations. In particular, the lecturer (Clive Cochrane) wished to encourage critical thinking about contentious issues in IT and society, such as computers and privacy. It is quite easy for face-to-face discussions to degenerate into monologues, silence filled by the teacher, or an exchange of unjustified opinions. So there is even a question of whether critical thinking takes place in face-to-face seminars, let alone computer supported ones.
This experiment was analysed using two techniques. The students were given a questionnaire to complete at the end of the semester, in which they rated how much each technology contributed to features of critical thinking as set out by Garrison. This was then analysed by factor analysis, and has been reported elsewhere (Webb et al. 1994).
We also transcribed both face-to-face and computer-assisted seminars, and analysed their contents. An earlier paper (Newman et al. 1995) describes our content analysis method. This paper reports the results of this content analysis in the first year we used computer conferencing as an alternative to conventional seminars.