Plots, Blocks and Video Tape; Experimental Design Made Easy
Dr Glenys Bishop
Department of Applied Mathematics
"I was trying to show students what an experiment actually looked like in real life, concentrating on the design rather than the analysis"
Description | Aims | Process | Evaluation | Contact
Description
This project deals with teaching large statistics classes in the first and second years of tertiary education and with the effective use of educational technology. The main aim of the project was to produce a six minute video showing the interaction between local scientific researchers and statisticians and illustrating the concepts of experimental design. The video is suitable for showing during lectures in introductory statistics courses.
The concepts of experimental design are hard to teach. Students display a lack of understanding of the basic principles. Postgraduate and Honours science students who have studied an introductory statistics subject during their undergraduate programme often fail to grasp the concepts. One reason for this is that the available teaching aids (such as commercial videos) for this level offer so much information on statistical analysis that issues of design are not clearly discernible.
Aims
The purpose of this project is to:
- improve the quality of learning and understanding statistics by introducing relevant and innovative teaching practices in our introductory courses;
Process
In the Department of Statistics, as in similar departments around Australia, we regularly use commercially available videos for motivation. One problem with these is that the subject matter is not local. The other is that the examples often involve complex statistics and so the detail is glossed over. I deliberately chose an example that was not too complex and that provided a useful context for the subject matter being taught in lectures. There is no discussion of analysis on the video because that would blur the issues of design that I was aiming to impart.
Production of the video involved:
- finding a willing research student with a suitable experiment
- liaising with the student, planning the content of the finished video
- arranging times and locations for three video shooting sessions
- writing the script for the voice-over
- constructing diagrams in PowerPoint
- one and a half days of editing.
A six minute video suitable for showing in lectures as a motivational example of the use of statistics was completed in 1996. The subject matter of the video is the design of an agricultural experiment. I chose this particular experiment for two main reasons. First, David Sloane, the postgraduate student responsible for the experiment was very cooperative. I relied upon his goodwill for assistance with the filming and for agreeing to appear on the video, discussing the reasons for his experiment. Second, his experiment used the principles that I wanted to illustrate.
After viewing my video, students understand what a plot and a block actually look like in practice (see clip below). An experimental design is no longer just a series of diagrams and numbers on a page. They can also see how an individual experiment fits into a research programme.
In summary, we have been able to teach old concepts more effectively than previously and the students have grasped these concepts better because of the practical video example. Student learning has been improved because students have been able to visualise an experiment rather than have the concepts presented in the abstract.
Evaluation
The success of this project has been monitored in several ways. Students have a written assignment, the content of which directly addresses the material covered in the video. This enables me to see if students have gleaned the important concepts from the video. Other lecturers in my department use the video and provide feedback. So far both these measures provide support for the utility of the video. Finally, more subjectively, I can judge the students' enthusiasm for the video from the mood of the class.
Comparisons with earlier student groups who did not have the benefit of the video are unfortunately not possible as there are a number of confounding variables that would undermine the legitimacy of such a comparison. One of the best ways still to evaluate teaching materials is for the lecturer to monitor what works best for her/him.
In July 1996 I showed this video at the Statistics Education Workshop in Sydney. In August 1996 I showed it at the New Zealand Statistics Association conference and in 1998 at the International conference on teaching statistics in Singapore. The video has always been reviewed favourably and numerous lecturers from Australia, New Zealand, the United Kingdom and the USA have requested copies.
Example from Experimental Design 4.0 M
The video is available for a cost of A$10, which covers the copying expense. An additional A$5 is needed if postage is required within Australia or A$10 overseas.
Enquiries to the Department of Applied Mathematics ph +61 8 8303 5418
Dr Glenys Bishop was previously on staff in the Department of Applied Mathematics, now at the Australian Bureau of Statistics.
Last updated 26.05.99
