In-class e-testing of statistics in a large cohort of diverse ability

Weir, Iain; Gwynllyw, Rhys; Henderson, Karen

Abstract

We report on our experience of running in-class e-Tests for a new Level 1 Business Decision Making module taken by 800 Business School students at the University of the West of England (UWE Bristol). The module comprises a one semester course covering statistical methodology in far greater depth than had been taught previously and modernised to include analyses using the statistical software SPSS rather than by hand calculations. The syllabus constituted a challenging amount of material to cover, especially since the student cohort was a large, diverse group of non-mathematicians.
The controlled conditions element of the module’s assessment comprises the average mark of the best two from three in-class e-Tests. In each e-Test a student receives their own unique data set to work on and performs a complete statistical analysis of it, using SPSS. We describe how we securely assessed the large number of students over fifteen 50 minute PC classes.
The main learning tools were key skills e-Assessments that provided the student with a random data set to import into SPSS to then appropriately analyse and report on. Repeated use of these would ensure that a student would thoroughly learn a key skill and cover various analysis outcome scenarios, for instance significant or not significant test outcomes. Each key skill e-Assessment has multiple embedded links to comprehensive Help pages that provide SPSS ‘how-to’ information or output interpretation. This form of support is more targeted and immediate than a student having to refer back to the notes. It also encourages the student to self-learn rather than automatically seek staff help. This allowed the staff to concentrate on giving statistical understanding and interpretation advice as opposed to say wasting time on the mechanics of producing SPSS output.
The e-Assessment system used was Dewis which is a fully algorithmic open-source e-Assessment system which was designed and developed at UWE. It is a completely stand-alone web based system used for both summative and formative assessments. Dewis’ ability to communicate with the R programming language greatly facilitates the task of generating bespoke student data and providing answers that match SPSS screen output.

Citation

Weir, I., Gwynllyw, R., & Henderson, K. (2018, September). In-class e-testing of statistics in a large cohort of diverse ability. Paper presented at 12th International Symposium on Advances in Technology Education Nurturing Professionals for Smart Cities: Way Forward for Technology Education, Hong Kong

• Presentation Conference Type: Conference Paper (unpublished)
• Conference Name: 12th International Symposium on Advances in Technology Education Nurturing Professionals for Smart Cities: Way Forward for Technology Education
• Conference Location: Hong Kong
• Start Date: Sep 18, 2018
• End Date: Sep 21, 2018
• Acceptance Date: Sep 4, 2018
• Publication Date: Sep 21, 2018
• Deposit Date: Jan 24, 2019
• Publicly Available Date: Jan 25, 2019
• Journal: Conference Proceedings of ISATE2018
• Peer Reviewed: Not Peer Reviewed
• Keywords: e-Test, e-Assessment, Statistics, Dewis, R, SPSS
• Public URL: https://uwe-repository.worktribe.com/output/860387
• Publisher URL: https://isate2018.vtc.edu.hk/files/Full%20Paper/Parallel%20Session%204.pdf
• Additional Information: Title of Conference or Conference Proceedings : 12th International Symposium on Advances in Technology Education Nurturing Professionals for Smart Cities: Way Forward for Technology Education