Project Supervision Issues

At a recent meeting of the School (Undergraduate) Teaching Committee, issues arising from increasing numbers in year four projects were considered.
  1. It is clear that with increasing numbers of entrants in first year, numbers in fourth year must necessarily increase. At present, all fourth year students (whether in CS, CE or SE) undertake projects, and numbers for next year may be upward of 200. Divided between 35 or so academics Lecturer B and up, this would mean an average of around 6 projects/academic. [Note that some projects are currently supervised by persons other than Lecturer B and up.] The actual distribution of supervision at present is very uneven, with some academics (B and up) supervising as many as 16 students, and some supervising none.

  2. Projects have in the past been viewed (by some) as mini-research-projects. Such projects are seen as inappropriate for the lower echelon of fourth year students (especially in CE). In the case of CE and SE, the engineering perspective on the purpose of a project in fourth year is presumably to allow students to complete a substantial project under supervision (rather than research, say). Thus a project can be seen as one of two things - an engineering project, or a research project.

  3. Another perspective on classifying projects is according to student ability. From this perspective, there need to be challenging projects for the most able students, and relatively routine projects for less able students.

  4. Projects can be classified as group projects (that is, projects which involve collaboration by a group of students) and individual projects (that is, group size = 1). The view that group projects should be available to able students, and that able students should be encouraged

  5. They can also be classified according to the number of groups (of size 1 or larger) undertaking a particular project. For example, one might have 10 students each individually (not collaboratively) working on 10 replications of the same project - rather like a large assignment. This might be an appropriate strategy for dealing with large numbers, particularly of less able students. Such projects could be termed "replicated" projects. It was suggested that replicated projects could save on supervision time by having simultanesou meetings between the superviser and all of the students undertaking the project.

  6. One proposal for dealing with large numbers of projects involved insisting that all staff (other than those with heavy administrative loads, such as perhaps the Head of School) must supervise at least n project students, with n = about 4 initially, but inevitably rising as the number of fourth year students rises. These projects could be group projects or not.

  7. Another proposal suggested classifying students according to their average mark, and allowing those with an average < 60% (say) only to undertake a replicated project. Those in the next band, say 60% to < 70%, might be allowed to undertake an individual project. Those with an average higher than this could be encouraged to undertake a research oriented project or a group project according to their aspirations.

Bill Wilson
AndrewT adds, on the topic of estimated numbers of project students:

CE/BIOM and SE are hard to estimate as their typical progression rate isn't yet known.

CE(s1+s2)CE/BIOMCSSETotal
1999113+301100154
2000103+26141023186
200198+26141052200
200298+26251090249

These are only quick estimates.

Andrew