ChatGPT Should Force Higher Education to Rethink Assessments
The advent of ChatGPT and similar chatbots provides another opportunity for the sector to reflect on why and how it assesses – and what higher education is for.
The advent of ChatGPT and similar chatbots provides another opportunity for the sector to reflect on why and how it assesses – and what higher education is for.
Sioux McKenna, Rhodes University; Dan Dixon, University of Sydney; Daniel Oppenheimer, Carnegie Mellon University; Margaret Blackie, Rhodes University, and Sam Illingworth, Edinburgh Napier University
The COVID-19 pandemic was a shock to higher education systems everywhere. But while some changes, like moving lectures online, were relatively easy to make, assessment posed a much bigger challenge. Assessment can take many forms, from essays to exams to experiments and more.
Many institutions and individual academics essentially outsourced the assessment process to software. They increased their use of programs like Turnitin to check for matched wording in students’ assignments. And for closed-book, timed tests they used tools such as Proctorio, which monitor a student’s computer or phone while they write exams.
But universities did not seize this chance to reflect on what higher education is for and how assessment might be used to enhance its achievement. Instead they doubled down on the status quo, breathing a sigh of relief once isolation and lockdown orders were revoked and things could return to “normal”.
The advent of ChatGPT and similar chatbots provides another opportunity for the sector to reflect on why and how it assesses – and what higher education is for.
ChatGPT is a chatbot technology, powered by artificial intelligence (AI), that enables users to have natural, human-like conversations with a computer. It uses advanced language processing techniques to understand user input and provide natural, contextual responses. With ChatGPT, users can converse with a computer in a way that feels like talking to a real person. It scrapes information from a large database mined from the internet and uses this to create a unique response to a prompt.
So, for instance, it can write an essay on any topic – “the advantages of breastfeeding” or “the social complexity of the refugee crisis in Europe”. It can also be trained to provide context-specific essays.
We are academics from South Africa, Australia, the UK and the US, working in fields related to education, ways of learning and teaching, and academic practice. We believe ChatGPT could be a powerful impetus to shift from understanding assessment as the assurance of an educational “product” to assessment as learning.
Used properly, it could be a valuable way to teach students about critical thinking, writing and the broader role of artificial intelligence tools like chatbots in the world today.
The advent of ChatGPT has prompted a variety of reactions from universities all over the world. In the UK, for instance, the reaction towards ChatGPT and higher education has veered from the hyperbolic – will AI ruin universities? – to the more measured, such as considering what students think of the technology.
If the purpose of higher education is that students memorise and summarise a body of knowledge, and that this is then certified via assessment, then ChatGPT is an existential threat. The market value of credentials is directly threatened if universities can no longer confidently assert that the texts assessed by academics have indeed been produced by their students.
But if the purpose of higher education is to nurture a transformative relationship to a particular body of knowledge that enables students to see the world – and their place in it – in new ways, then assessment takes on a vastly different meaning.
Used well, ChatGPT and similar tools can show students the wonders and responsibilities of acquiring and building powerful knowledge. It can assist rather than being seen in opposition to their learning.
Here are four ways this might happen.
1. Students can reflect on articles produced by ChatGPT which have fabricated references and distorted information and then deliberate on the potential consequences of this in an era of fake news.
2. Students can be set assignments that require them to compare ChatGPT’s answers to ones they have developed and ascertain whether they know the material and how it might be represented differently.
3. ChatGPT can be used to support essay writing and to help foster a sense of mastery and autonomy. Students can analyse ChatGPT responses to note how the software has drawn from multiple sources and to identify flaws in the ChatGPT responses which would need their attention.
4. Students can be encouraged to consider the extent to which their use of ChatGPT has enabled or constrained their access to powerful knowledge. This is a chance to critically reflect on where and how the use of AI is taking place in society and their potential future professions.
There is already a multitude of ideas available online about how ChatGPT can be used to create prompts for assignments. Lecturers and students can explore these to see how they might be adapted for their own learning and teaching needs.
None of these ideas will be simple to implement. Academics will need support from their institutions in considering what such technological developments mean for their disciplines. And, we’d argue, that support must help academics to move beyond seeking ways to trick the software or to monitor students.
Society and the higher education sector squandered the opportunity that COVID presented to reflect on what higher education was for and how assessment might be used to enhance learning.
Rather than signalling the end of higher education, ChatGPT has instead presented the sector, and society more broadly, with another opportunity. This is a chance to develop innovative and inclusive teaching, learning, and assessment aligned to such understandings.
Sioux McKenna, Director of Centre for Postgraduate Studies, Rhodes University & Visiting Research Professor in Center for International Higher Education, Boston College, Rhodes University; Dan Dixon, Adjunct Lecturer, University of Sydney; Daniel Oppenheimer, Professor of Decision Science and Psychology, Carnegie Mellon University; Margaret Blackie, Associate Professor, Rhodes University, and Sam Illingworth, Associate Professor, Edinburgh Napier University
This article is republished from The Conversation under a Creative Commons license. Read the original article.
4 Comments
I’m thoroughly impressed by the insights shared in this article! It adeptly discusses the need for a fresh perspective on higher education assessments. The suggestions presented here are not only forward-thinking but also have the potential to reshape the learning landscape. Kudos to the author for shedding light on such a crucial topic with such clarity and depth.
This thought-provoking article highlights how ChatGPT is pushing the boundaries of education and assessment. It’s inspiring to see technology challenging the status quo and encouraging higher education to evolve. Thanks for shedding light on this important topic
This article makes a compelling case for reevaluating educational assessments in light of AI like ChatGPT. While risks exist, the author suggests this technology could catalyze positive changes, like reducing emphasis on rote memorization. With proper safeguards, ChatGPT may push educators to focus more on critical thinking and skills application. Overall, thoughtful perspective on reimagining assessments.