Informatics Research Methodologies
Lecturer's Self Review 2004-05

Alan Bundy


The IRM module ran for the third time this year, following the pattern established in the last two years. The main change was to replace the coursework on writing a grant proposal with one on rewriting one of the papers reviewed in an earlier coursework. This change was necessitated by the introduction of IRP for all MSc students. The grant proposal and IRP would have overlapped too much and I could not retreat to the earlier literature review coursework because of overlap with IRR.

Student Numbers

20 students registered for the course --- less than half the number of last year. The numbers broke down as: 12 MSc, 5 UGs, 3 PhDs (observing). The numbers of MScs doubled since last year, but only a third of the UGs registered. It rapidly became clear that this was an artefact of the Curriculum Project. I had had to specify the level of IRM. Level 11 seemed most appropriate, but this had the unintended side-effect of removing IRM as a UG4 course. Consequently, UG4s did not get an advertisement for IRM and only took it with special permission. To cure this problem, I have proposed a level 10 version of IRM, in which the main change is to replace the paper rewriting coursework with a discussion of the evaluation of the student's project. Both versions of IRM will run in 2005-06.

How the Module Went

As last year, there were 7 lectures in the first 4 weeks, including 3 guest lectures: 2 by Frank Keller and 1 by Jennifer Tenzer. I also ran two sessions on coursework. All 20 students did 25min research presentations. With fewer students, it was not necessary to create any additional slots.

For their paper reviews, students had the options to use: the improved ERA3 expert system, the latex template or free form. Most used the ERA3 form, which seemed to work well. I again have a student project to improve ERA, so I hope to have an improved version next year.

On the whole the module went well and seemed to be very well received by the students, as witnessed, for instance, by the feedback from the questionnaires. The standard of coursework was very high. Attendence at student presentations held up well this year, for which the more humane timing of 11.10-12.00 must get a lot of the credit.

With the lower student numbers and lower coursework load, the marking burden was considerable reduced this year, although it remains high, especially for the paper-rewriting exercise. Because of this high workload, I had mixed feelings about introducing a level 10 version of the course, but my more noble principles won this battle.

I had some concerns about whether the new paper rewriting course would prove too difficult for the students or would encourage plagiarism. I was very pleased with the outcome. The standard was very high, and there was only one sigificant case of plagiarism, which was fairly innocent and was dealt with by a discussion with the student and a reduced mark. The reduction of the paper review coursework from 4/6 to 2/4 also worked well.

Along with other coursework-only courses, I was given an action by the Board of Studies to ensure that the coursework burden was not too high. I verified that the number of hours assigned to coursework was consistent and totalled 100 hours. I also asked the students for feedback on whether my estimated durations for each coursework were realistic and properly balanced. Those students who replied were very positive and verified these estimates. They did not want a reduction or change to the coursework.

Plans for Next Year

The main innovation next year will be the level 10 version of IRM. This will share the lecture/presentation slots, and the presentation and paper review courseworks with the level 11 version. The main differences will be pass/fail boundary and the final item of coursework. UG4 students will write an analysis of their project, based on what they have learnt in IRM, with the main focus on the hypothesis formation and evaluation.

I plan to use the improved version of ERA next year. This, for the first time, should allow students to backtrack, while keeping track of inferences using an ATMS. It should also use inference to tailor the referee form to different areas of informatics.

Once again, I would like to express my thanks to the 2 guest lecturers, Frank Keller and Jennifer Tenzer, who helped to make this module a success, and to Neil McGillivray, in the ITO, for being an ever present source of advice and help.

Home : Teaching : Courses : Irm : Annual-reports 

Informatics Forum, 10 Crichton Street, Edinburgh, EH8 9AB, Scotland, UK
Tel: +44 131 651 5661, Fax: +44 131 651 1426, E-mail:
Please contact our webadmin with any comments or corrections. Logging and Cookies
Unless explicitly stated otherwise, all material is copyright © The University of Edinburgh