Teaching biology in the Collaborative Learning Centre

Learning exercise

This learning exercise was inspired by the BIOL1014 computer practicals. It has been adapted for collaborative learning, and the actual material is more advanced. It hopefully will be more interesting for lecturers to do than a first-year computer practical.

Step one - group division of task (5 minutes)

Divide the twenty amino acids up among the people present in the pod (10-20 people ideally). If you can't remember all twenty amino acids, there's a handy list here.

Step two - individual research (15 minutes)

Fill out the form for your amino acid (repeat if you have several to work with). Hint: Search for [amino acid]-rich protein in some or all of the following places:

Step three - group report (15 minutes)

Prepare a summary report of your group's discoveries, using this powerpoint presentation as a template, if you like.

The BIOL3004 Protein Structure and Evolution Elective: Semester 1, 2005

This elective, with about 33 students enrolled, was conducted in the CLC on Tuesday afternoons over a total of seven weeks. There were five weeks during semester, when the students worked on their projects. The Tuesday afternoon during swotvac, we made ourselves available to any students who wanted to polish their project. Tuesday 14th June, we held a "mini-conference" in the CLC, where the students presented their work to each other.

The aim of the elective was to familiarise the students with bioinformatics tools used in protein structure and evolutionary analysis. To enhance collaboration, the proteins the students worked on were those involved in DNA replication. Students worked in pairs on a specific protein or proteins, and were also encouraged to collaborate with students working on proteins that interacted with their protein.

The entire elective was embedded in a wiki, an interactive web site that allows registered uses to add web pages, links, etc. You may already have used Wikipedia, a free encyclopedia that is developed collaboratively by anyone who wants to participate. There are a number of wikis in more specific subject areas, with heavy coverage of computing and games, but also community and ethnic projects. Some specific wikis of interest cover areas such as: evolution and origins; Natural Killer Cells; biological field work; and the human brain.

The registered users of the PSE elective wiki consisted of the lecturers (Thomas Huber and myself) and students. The wiki thus contains a mixture of lecturer-supplied reference material (lecture notes, useful web links, details of assessment scheme) and student-created material: their results for their protein.

You are welcome to explore the wiki on your own.

Technical details of wiki-usage by students

The wiki has a number of features that make it suitable for a student project like this. The fact that the students are registered wiki users on par with the lecturer creates a sense of working on the project together. This is further reinforced by lecturer-provided reference material being on the same pages, in the same format, as the students' own work. The editing format for the wiki makes it easy to include many html features, such as adding new web pages, linking them, and linking to outside material. The students can thus easily present material in a format they're comfortable with, and which requires less linearity than traditionally. Overall, the students were more familiar with the wiki's editing features than the lecturers by the end of the project.

Students have the capacity to modify or even destroy other students' work, however, all modifications are assigned to the person who did them. Furthermore, the wiki keeps a record of earlier versions and these can be restored if necessary. Students initially had some concerns that they had the freedom to "vandalise" each other's work. In practice, no-one chose to even try this, as far as we can tell. This may partly be because they were third-year.

Outcomes

One of the implicit aims was to introduce students to research-oriented thinking. Most undergraduate practical work has been done many times before, and the outcome is well-known by the tutor and lecturer at least. One effective way to model scientific problem solving is to assign the students a problem the lecturer and tutor do not know the answer to: when the students ask for help, the tutor has to puzzle the situation out in front of the students.

It was stated up-front to the students that the lecturers and tutors were experienced at using the bioinformatics software, but didn't know particularly more than the students about DNA replication. Many students found this very unsettling at first. In retrospect, many came to appreciate that they gained a deeper understanding of the software by using it in a situation where the answer wasn't known, and also that they got a better sense of what doing science feels like.

Arranging collaborative learning

Most collaborative learning involves dividing a large task into subtasks that can be worked on individually, or in small groups. The division into subtasks can be done in two main ways, each with advantages and disadvantages.

Specialisation

The task is divided into areas of specialisation. For example, a research project can be divided into experimental design; data collection; literature research; statistical analysis; paper preparation; grant and financial administration. The advantages of this is that it is a familiar and comfortable model, for good reason, as it is probably the most efficient division of labour; and it also works effectively when there are a variety of students from different backgrounds who have different strengths and interests. The major disadvantage from a teaching perspective is that there is usually some particular subtask or specialisation we'd like all the students to gain familiarity with (for example, statistical analysis). Another disadvantage is that this type of task division may create "bottlenecks" or otherwise leave some students idle, waiting for other tasks to finish.

Parallelisation

Both the exercise at the top of the page, and the BIOL3004 elective, are examples of parallelisation. The task is divided up so that students do the same tasks in parallel to different data, that will later be combined. The advantage is that all students become familiar with specific tools (e.g. bioinformatics software), and have "something to do" most of the time. The disadvantage is that there is less impetus for the merging into a larger project unless this is pushed for, as the students "become experts" at their data, and have less motivation to join in the larger collaboration. It is also a less familiar structure, and requires more care to set up.

One important effect of the way a task is divided into subtasks is the coherence and appearance of the final, combined project. The project is likely to look more cohesive when one specialist subgroup is responsible for presentation, for example. However, there is also a risk of communication breakdown, and while the presentation may be cohesive, it may not reliably reflect other specialisations in the project; for example, the statistical analysis may be described incorrectly. The final appearance of a parallelised project is much less cohesive, unless special attention is paid to it. Compare different student project pages from the PSE elective.

An important practical issue in designing collaborative learning is that, because each student is doing something different, it is impossible as a lecturer or tutor to keep track of everything that's going on, when the collaborative session is actually happening. If you think of a linear scale:

classic lecture self-directed learning
teacher control individual freedom
structure creativity
then collaborative learning is somewhere in the middle, with enough control and structure to reassure the students and ensure the learning aims are met, but with the inevitable freedom that comes from individual subtasks.

In practice, the best way to achieve this balance is with pre-planning. If attention is paid to how the project is set up, and the majority of materials are available on a webpage such as this for reference, the majority of students will not need individual attention most of the time, and yet not feel left on their own.

System synthesis?

One aim of collaboration in general is to create something greater than the sum of its parts. The Wikipedia is an example. With the increasing interest in networks and complex systems in biology, collaborative learning exercises built on this concept would be valuable. Similarly to parallelisation, students would be working on specific units of data, but the focus would be on interactions between the units. A successful implementation of this idea would require a significant amount of pre-planning, setting up an infrastructure or framework, not necessarily visible to the students.

Assessment

It is difficult to design fair assessment of collaborative projects, because each individual's experience is different. It has been suggested that collaborative work should not be marked, beyond participation/non-participation. Particularly for earlier-year students and single-session projects, this may be the best approach.

Notably, it was important to our students in the PSE elective that the marking scheme was based on fundamental concepts like amount of research done; understanding of concepts; etc, rather than a checklist of completed tasks. This agrees well with our own assessment philosophy, but is more time-demanding to mark, and perhaps only practical for relatively small groups.

One element we introduced to make marking more manageable and exploit features of collaborative learning, was peer assessment. Each student was assigned five projects (not including their own, or their partner's) to mark on a simple 1-5 scale. This contributed 25% towards their final mark. Overall, we had good agreement between student marks and tutor/lecturer marks.

For larger, more introductory collaborative labs, I suggest either attendance-based marks, or possibly a summary quiz (which could be administered automatically using a tool like BlackBoard).

Last modified: 12th July 2005 by Ingrid Jakobsen Send email