The idea is simple – provide staff who aren’t experts in programming with the skills to efficiently generate assessment questions, set up automated marking templates and provide personalised and timely feedback to students.
Colin Zhang outlines how this is being made possible for sessional staff teaching Statistics and Finance units and how it will be rolled out to other units and courses.
Why the need to automate marking and feedback?
Sessional teaching staff face significant workloads encompassing lesson preparation, teaching, marking and assessment development. There was a need for more support for sessional staff. This project, titled ‘Equipping Sessional Staff with Programming Skills for Automation of Assessment and Feedback Using Artificial Intelligence’ is funded by a Sessional Academic Staff Engagement Initiative Grant. It aims to empower teaching staff with AI and coding skills to allow them to automate assessment and feedback which will help them with the marking load.
The department has some large first year units which have Excel-based assessments. A set of Excel workbook tools has been developed for teaching staff which they can use across a range of different types of numerical-based assessments.
Excel workbooks form the basis of these tools – but AI makes it possible without needing to know any programming
User friendly Excel worksheets, each equipped with VBA code, have been meticulously crafted for use by staff who don’t necessarily have any coding background. VBA stands for Visual Basics Applications and is a computer programming language developed by Microsoft for applications like Excel.
The workbooks are designed for assessment questions where the answer is provided by the student in the form of an Excel spreadsheet – ie. where they do a calculation as the answer.
Once the coding is set up for one assessment question, staff can easily customise it for another question by simply copying and pasting the question into an AI platform such as ChatGPT and asking it to create new coding based on the variables they provide. All without needing to know programming or how to work with VBA coding.
AI supports the use of tools like this because AI can provide the staff member with the new code for a different question, so they don’t need to know the coding themselves. Previously this type of approach was limited because you needed to know how to program VBA. But now with AI support, people can easily do the programming by themselves by using the example we provide.
In a nutshell, this is how it works
Staff use the example Excel Workbook templates and customise them for their specific assessment question. The workbooks contain three works sheets:
- Official answer
- Student answer
- Feedback
Students are provided with a template to use for submitting their responses. Staff transfer (copy/paste) the student answers directly into the workbook. When the code is run, it will compare the student answer against the official answer, assign marks, and generate automated feedback.
It’s often the case that 80% of incorrect answers are likely to have common errors. So, in specifying the feedback scripts it is possible to provide some specific feedback e.g. “if you got this answer, you probably did this wrong“.
How staff were guided in the use of the tools
A package of resources has been created including:
- Two Excel workbooks with associated work sheets containing worked examples
- Demonstration videos to guide staff in utilising AI tools like ChatGPT to modify the code for Excel based assessments
- Comprehensive step-by-step guides on using VBA coding in automated assessment and feedback
- Learning materials shared through an online platform, fostering scalability
- Community of Practice enabling ongoing support and knowledge exchange
After initial implementation in STAT1250 it will be rolled out to other relevant units across the faculty.
Get all the nitty gritty information
READ the documentation explaining how the Excel workbooks can be used
VIEW the instructional videos
- Demonstration of the conditional formatting workshsheet
- Demonstration of automatic marking formulas in Excel spreadsheets
DOWNLOAD the example Excel worksheet template
Contact the project team to find out more
Contact the project team – Colin Zhang colin.zhang@mq.edu.au, Prashan Karunaratne prashan.karunaratne@mq.edu.au or Joshua Peate joshua.peate@mq.edu.au
ACKNOWLEDGEMENTS:
Project team: Prashan Karunaratne, Joshua Peate and Colin Zhang
Banner image: Photo by Stock-Asso on Shutterstock
Share this: