Stuck thinking is a great term I learned from one of my kids’ books. It describes what happens when you get an idea in your head that becomes so large that it creates resistance to change, and real problems when things don’t go as planned. I’m putting it up there with “the story I’m telling myself is…” on the list of helpful phrases for pointing out when the world is not operating quite the way you think it is.
It’s a phrase that has ironically been stuck in my head since I first read it, and it came to mind recently in a conversation with teachers about why they’re struggling to adapt to the implications of generative artificial intelligence in secondary school assessment. This conversation went much like dozens of others that I’ve had in other schools across different states and different sectors in Australia. It goes something like this:
“We can’t let students use AI because they’re not allowed to use AI in the exam. If students use AI, I can’t guarantee that they’ve learned anything, so I’ll run all of my assessments under exam conditions. Because not only does that prove they haven’t used AI, but it’s also practice for what they have to do at the end of the year.”
It’s a compelling logic, and although this conversation is seemingly about artificial intelligence, I think it represents something that runs much deeper for us in secondary education. It represents stuck thinking about what the examination is actually for.
What we think the exam is for
We think the exam is the ultimate endpoint of a mainstream secondary school student’s education. We think the exam is a pathway to university. We think the exam is a reflection on our own individual capabilities as teachers. We think the exam is a measurable outcome that can be used to justify a school’s position in the broader education community. We think the exam is a marketing tool; something that can be shown to parents in prospectuses.
And to be sure, the end of school exams in most states in Australia are all of these things. Unfortunately, when we get stuck on thinking that these things are what the exam represents, it has clear ramifications upstream in what a student learns in school. Because thinking of exams in these ways forces students and teachers to play an educational game where the goal of schooling is to win at the exam.
Here are some direct implications of playing that game:
Teachers want their students to succeed in the exams because student success is good for the individual student, good for the individual teacher, and good for the reputation of the school. In order to succeed in the exam, the most expedient thing to do is to turn every assessment into a miniaturised exam where units of work align directly to examination content.
Assessment tasks can almost be one-for-one replicas of those sections of the exams. This happens, for example, in VCE English, where Unit One lines up with Section A of the exam, Unit Two with Section C of the exam, and Unit Three with Section B of the exam. This is, of course, teaching to the exam. But it is not because teachers are lazy or can’t think of innovative teaching methods. It is because it is the best way to play the game.
Students, too, are playing the game. For a student, although the exams loom large at the end of the year, more of their day-to-day attention is taken up by classwork. Classwork and school assessed coursework point towards the exam, but are marked in the here and now. Students are also generally aware that they are being ranked. Again, this is more specific to the Australian sector than other jurisdictions, but it’s fairly common across different regions for students to be competing with one another. Therefore, it’s in the best interests of the students to play the game to the best of their abilities. Students with five to six different subject areas are, of course, going to use technology like generative artificial intelligence if it is easy to access, expedient, and does a reasonably good job. They will play the game with whatever tools are available to them.
Both teachers and students are playing the same game with the same goal in mind, but their methods of playing the game are in tension with one another. This tension is currently manifesting as teachers chasing down academic integrity concerns related to GenAI, and faculties across many schools locking down assessments to pen and paper examinations as a way to secure them against AI. But much of this is predicated on the stuck thinking about what the exam is for.
What the exam is actually for
The end of school examination for certified studies like the VCE in Victoria and HSC in New South Wales is for statistical moderation across a cohort. If we’re going to get stuck on thinking anything about exams, this should be that thought. But when I work with schools, more often than not, this is almost an alien concept.
The reason for that is fairly simple. Although staff and students are expected to play the game, it’s fairly rare to go to a school where the rules of the game are actually taught. The reason certain schools across Australia do well in the exams is not because they are more privileged, wealthier, or in a more socioeconomically affluent postcode, although, of course, all of these things contribute to student success.
The reason some schools do better than others is because they understand the real rules of the game. They understand that the purpose of the examination is to moderate the internal assessment results up or down against the performance of the rest of the state. It is therefore important that all students in a school cohort perform comparably, and that marking of school-based assessments is accurate across the cohort, so that the rank order based on coursework results reflects the performance of students in the exams.
And there are consequences for misunderstanding the rules. Treating every assessment as a miniaturised examination means that you are not getting a valid perspective on your students’ capabilities. A significant portion of your students will not be able to demonstrate their capabilities under examination conditions: autistic, dyslexic, ADHD students, students with auditory processing disorders in echoing study halls, students with physical disabilities, students with anxiety, students with temporary illnesses, migraines, gastroenteritis, students with problems outside of school, students who haven’t had breakfast on the morning of the exam…
There are a host of students who cannot perform their best work under those conditions. When we conduct all or most internal assessments under exam conditions based on the stuck thinking that we must do that because students have to do the exam at the end of the year, then you deny these students the opportunity to perform in the best conditions based on their actual capabilities.
Another consequence is the cheating behaviours that will increase as a result of students understanding that the only thing that matters is the exam.
As a student, if the only thing that matters is the exam and I’m disengaged with my subject, of course I am going to use GenAI to complete as much of the work in that subject as possible. I might know on one level that I’m shooting myself in the foot, because if I use ChatGPT all year, I’m probably not going to do that well in the exam. But as a rule, students who are disengaged in subjects are already disinterested in the exam at the end of the year, and quite possibly won’t start thinking about the exam until term four.
When we tell students the only important thing is the examination, we are essentially saying, “so you might as well use AI for everything else”.
Unsticking
It’s time that we stopped saying “But students have to do an exam at the end of year 12.” Of course they do. If you’re in Victoria, New South Wales, Queensland and other states, this is the system that you’re working in. If you’re in South Australia, some of your subjects have exams and some don’t. If you’re in the ACT, none of your subjects have exams. There’s no blanket solution to these problems, but if you are in a state where students sit exams at the end of the year, I would encourage you to unstick your thinking about the purpose of the exam.
I want you to remember what I said earlier. The purpose of the exam is for statistical moderation across the cohort. I would encourage you to learn as much as possible about how that works. The VCAA, for instance, has some very clear, short videos on the topic.
And then I want you to think about what the purpose of learning is. This, of course, has nothing to do with GenAI. I wrote an article some years ago for English teachers called You’re doing coursework wrong: Why SACs are not exams, in which I explained that for students to learn well, we have to decouple coursework from the exams.
Bringing GenAI back into this conversation looks quite different. If the purpose of most of senior schooling is to learn things and not just to perform evidence of learning in the exam, then there are places that students can use GenAI to support what they’re doing in the class, and there are places where GenAI necessarily has to be excluded.
We cannot allow stuck thinking around the purpose of exams to determine the success of students who would otherwise be better served by demonstrating their learning in other ways.
We cannot let the apparent threat of GenAI to those demonstrations of learning be another excuse for forcing us into examination-style assessments. That is an incredibly negative and unhelpful reinforcing cycle that will fundamentally undermine the purpose of senior school education.
The answer, in my mind, is:
- Clarify the real purpose of examinations
- Be incredibly transparent with students, parents, staff and the broader school community about the rules of the game
- Separate the examination from the broader goals of senior school education
If I were talking to a student, maybe one of my own children, when they’re a little older, I’d probably phrase it something like this:
In and around your lessons and the time you spend with the teacher, the purpose is to develop the skills and the knowledge you need to demonstrate that you’ve learned what this subject wants you to learn.
You’ll be asked to demonstrate those skills in a variety of ways over time, and will be assessed formally and informally, with evidence of that learning being collected across the whole course of study.
The judgements that the teacher makes about your abilities will be reflected in formally graded coursework as well as indicative grades based on how well your teacher thinks you’ll perform in the exam. The point of the exam is to ensure that all of the students in this school are awarded grades that are fair and comparable to other schools in the state.
And so, if the teachers in this school are especially lenient or especially strict, the students won’t be disadvantaged compared to a school down the road where the opposite is true.
At the end of the day, you’ll get a grade that reflects a combination of how you performed in comparison to your peers and how you performed in comparison to other students in the state. This final grade is not a reflection of everything that you have learned.
Most teachers and students respond well to this kind of language, although it is challenging. Like watching my children deal with the fact that we arrived home slightly late, so there’ll be less time to watch TV before dinner, our stuck thinking around examination practices can manifest in frustration and a sense of helplessness that the system is working against us and our students. But we can work within the system that we have, and do so in a way that means even GenAI does not automatically have to be seen as a threat.
In a future post, I’ll outline what this might look like in practice. But for now, I’d encourage you to challenge your thinking about the purpose of end of school examinations.
Want to learn more about GenAI professional development and advisory services, or just have questions or comments? Get in touch:
Your message has been sent
Leave a Reply