This miniproject helped existing modules to develop into websites to provide e-delivery of exemplars of statistical methods and best practice, to showcase effective teaching and active learning tools. The two e-Learning websites are:
1. the design - and analysis - of experiments (DOE) website; and,
2. modern regression methods website.
The primary audience of these websites is undergraduate and research Chemists requiring knowledge of statistical methods for their work. As such, the authentic chemistry examples are used to motivate and illustrate the methods. However, the material is generic enough to be used in general statistics courses. It is believed that these websites provide a significant contribution to the eLearning resources available to the statistics and chemistry communities.
The websites have a particular focus on the application of the methods in chemistry research but much of the material is generic. Some key features are:
1. the use of authentic chemistry examples to stress the relevance of statistics in real applications;
2. the use of interactive examples which make use of R [1] functions and routines to allow students to engage with the material; and,
3. the provision of some formative feedback to students via ‘scenarios’ based on simulated data and also multiple-choice questions.
In order to provide formative feedback to students and allow self-assessment of learning objectives and understanding on the DOE website, two interactive exercises with feedback have been developed and implemented, along with an expanding database of multiple-choice questions.
To improve the presentation and interactivity of the regression website, interactive examples allow for both stimulation and planned participation of learners. Through collaboration with a computer scientist, the web-design incorporates best practice in design and implementation providing good structure and navigation and optimised graphics. The statistical analysis and simulation tools allow not only for student interaction but also for learning via repeating, with students able to reflect on results whilst trying different approaches [2].
At the outset of the project, the feedback offered by the regression website was limited in scope. The student can see the results of their decisions through the incorporated R functionality, but are rarely given feedback on the appropriateness of any choices that they have made. Further, there is no attempt to assess the student’s understanding of the concepts discussed by the modules. To improve this aspect, currently two forms of formative assessment are being incorporated:
1. assessment of concepts and understanding via a database of multiple-choice questions, with topic-specific questions for each section of the modules.
2. assessment of the application of statistical methods to real and simulated data.
The two modules (DOE and regression) are being incorporated in to undergraduate and postgraduate courses at various points in their development.
[1] R Development Core Team. ‘R: A Language and Environment for Statistical Computing’, R Foundation for Statistical Computing, Vienna, Austria, 2006.
[2] Paris, M. ‘Simulation authoring tools for interactive e-learning courseware’, Higher Education Academy, 2003.