LSE academic developer Mark Baltovic reports on the project based class work and alternative forms of assessment that have been incorporated into several of the Department of Statistics’ undergraduate courses
Statistics may not be a discipline immediately associated with active learning, but at LSE there has been a determination to build elements of it into some core courses. In the Department of Statistics’ half-unit ST201 (Statistical Models and Data Analysis) and ST312 (Applied Statistics Project) courses, for instance, project based work is used both to develop students’ learning and to contribute to overall assessment.
- In ST201, students work in groups of 3 on a project that forms 20% of their final summative assessment (the remaining 80% coming from a more traditional examination).
- In ST312, students work individually on a research question to produce a presentation for the entire group (accounting for 10% of their final summative assessment) and a written report detailing their research question, methodology and results (which accounts for 90% of their final summative assessment).
A pivotal aspect of both projects is that students must find their own data sets from real world sources to generate their research questions. While they are given some preparatory guidance on finding data (by LSE Library’s Academic Services), their biggest hurdle is often not necessarily the availability of relevant data, but rather its quality. Instead of working with the artificial and sterile data sets that are typically produced to support formative assessment activities, students have to confront the same problems that practising statisticians face in their research. They soon learn that they must make compromises in what they seek, that there are trade-offs to be made when seeking viable data. To get anywhere with their work, they are forced to make subjective judgments and possibly adapt their methodology to accommodate the constraints they come up against. As a result, students on these courses learn the ‘trade’ of statistics by effectively becoming statisticians during their projects. And the projects illustrate what Cooperstein and Kocevar-Weidinger (2004) call ‘one of the four guiding principles of the constructivist approach to active learning … that such “authentic tasks” are vital for successful and meaningful learning’.
The group work component of ST201 is an example of ‘co-operative learning’, which typically calls upon students to work in small groups to complete learning activities structured and guided by the teacher. To succeed, students need to interact socially with their peers, and thus must develop interpersonal and small group dynamic process skills. By incorporating the development of such skills directly into the nature of the activity, another of the principles of active learning – that learning is enhanced by social interaction – is brought into play.
Those same principles can be seen at work in ST327 (Market Research: An Integrated Approach). As before, real world examples are used to develop fresh case studies each year that explore the ideas discussed in lectures. Students work in groups of 5 or 6, select a case study, and then play the role of a market research company whose task is to make a bid to the company analysed in their case study. Summative assessment is comprised of:
- a group presentation of their bid (accounting for 15% of their individual final summative assessment);
- individual essays based on their group work (accounting for 25% of their final summative assessment); and
- a formal examination (accounting for the remaining 60%).
This blend of assessment formats reflects a growing belief in the department of the fairness and appropriateness of such approaches. There is an openness to still further new ideas for assessment – such as peer assessment using various e-assessment platforms – and certainly a feeling that the reduced proportion of marks accrued via formal examination (previously this was 70%) has been a positive move.
Moreover, students are responding positively to the new approaches. TQARO feedback on the courses has been favourable, and enrolment numbers this year are double those of last year. Indeed, this is indicative of a wider trend in undergraduate statistics education. A recent review of pedagogic research in this field over the past two decades (Kalaian and Kasim, 2014) identifies the significant and positive impacts that co-operational and collaborative learning methods have had in general, highlighting the principles outlined above as being central to such approaches and to successful learning. In the mean time, LSE’s Department of Statistics continues to explore new ways of delivering teaching and learning to its students.
- Cooperstein, S. and Kocevar-Weidinger, E. (2004), Beyond active learning: a constructivist approach to learning, Reference Services Review, vol. 32, no. 2, pp. 141-148.
- Kalaian, S.A. and Kasim, R.M. (2014), A Meta-analytic Review of Studies of the Effectiveness of Small-Group Learning Methods on Statistics Achievement, Journal of Statistics Education, vol. 22, no. 1.
Many thanks to Dr James Abdey (ST327) and Dr Jose Pina-Sánchez (ST201, ST312) for their information and insight into their courses.