Rachel Forsyth

Quite right too, congratulations! 🎉

I've hesitated to post about this on BlueSky since I know how many strong opinions there are. We hope teachers will read it and feel better equipped to make choices about their own practices in relation to GenAI in higher education. One view doesn't work for all, but none of us can look away #AIinHE

Beyond a critique of grades - Higher Education Grades have been criticised for as long as they have existed. For at least a century, scholars have claimed that grades are inaccurate and unreliable; they provide overly simplistic reductions of stud...

Grades have been criticised for as long as they have existed, yet little has changed. Perhaps it is time for new approaches? Our new paper on post-critical approaches to grading with @margaretbea.bsky.social, Rola Ajjawi and David Boud is out in Higher Education! link.springer.com/article/10.1...

The blog post written with @profpeter1.bsky.social for @advancehe.bsky.social has been published.

"The costs of Generative AI: chasing the elephant round the room."

www.advance-he.ac.uk/news-and-vie...

How do teachers build trust with students in the era of GenAI? @rmforsyth.bsky.social and I discuss findings of a recent series of focus group interviews with @niyamirandabond.bsky.social on @onehe.bsky.social . Watch the video and let us know what you think! onehe.org/resources/bu...

When I taught new university lecturers/faculty (about teaching) I always used to say, in the first session, "If you don't like young people, you should probably think about another setting to do research, or you might not enjoy your career".

This seems like a strange hybrid of norm-referenced and criterion-referenced grading. Norm-referenced is easier really, since you just need to put submission in rank order. Or are they going to write new grading criteria with extra hard requirements for As? Fine, but students will aim higher.

This team are on it blog.betterimagesofai.org

I think it's just guesswork to begin with - really important to talk to students about how it actually lands in practice. (this sentence applies to everything in teaching...)

That's a great way to present it. We can explain why we have landed on a particular size range and how it links to student learning. And we can encourage feedback on how well the range worked, and can change it for next time. Thanks!

Also I dislike penalties for overlong assignments in principle but teacher workload is a factor and being succinct is usually an important skill. A range gives the student a sense that there isn't a perfect size but that we compromise in order to make the task manageable for everyone.

FWIW I think students need some guidance about the range of acceptable sizes, but that this shouldn't be decided at institutional level. It's contextual to the subject matter.

screenshot including this text: Submission size The production and maintenance of a portfolio can be time-consuming, and this would normally be taken into account when considering the length of each individual element and the reflective element. You need to decide whether you want to see all of the work that students have completed over the term, or a selection of the tasks. If it is a selection, will you choose which items are submitted (say, the lab reports from weeks 3, 7 and 9), or will you leave the choice to students (for instance, submit the four article summaries which best demonstrate your progress over the term)? You could make it time-based, and ask for one piece of evidence per week of the course. The answer will depend on whether you want to focus mostly on regular performance and acquisition of essential skills, or personal development across the module, and this will be linked to the unit learning outcomes. Make it very clear what you are expecting to see in terms of size, and if there are any penalties for submitting too much, or you may receive many appendices which you won't have time to assess properly.

Text from screenshot: Submission size You will need to specify the size of the assignment: the number of words for a written piece, or the number of minutes for a presentation, or the size of a poster or an exhibition space, for instance. It is important to give students a sense of what is expected, to help them to put parameters on what is expected. This could be the most challenging decision you make about assessment, or it could be the easiest. Your university may have set assignment sizes for different types of assignment, which makes the decision simple, unless you disagree with the recommendations, of course. If you have free rein to choose, or you have selected an assignment task which is not yet commonly used, then the decision is a little more complicated. Custom and practice is a good place to start. Assignments of the same type on the same course should normally be comparable in size, otherwise students may become very confused about expectations. You can look at similar assignments in your own university or elsewhere — many universities publish course outlines which include this level of detail. Think about how much time the students should spend on preparing their assignment submission: a rule of thumb I have used is to say that working directly on the assignment should take up about 25% of the time available for the module. In the UK system, 1 credit point is nominally associated with 10 hours of student work, so for a 10 credit (5 ECTS credits) module associated with 100 hours of student learning, the student should spend 25 hours on the assignment. This gives you a basis for estimating how much they can realistically do in that time, including all reading, data collection, analysis, and so on. Thinking through the stages which might be needed for the assignment will also help you to break down the processes you think will need to be completed and to explain them to colleagues and to students.

Text from screenshot, which follows on from previous image: However, whatever approach is taken, it will just be a rule of thumb: all the studies done on student workload shows that there is a huge variation between individual students as to the time it takes to actually complete assignments and as to students' perception of how much time it takes (Crook & Park, 2004; Kember, 2004; Kember & Leung, 2006; Sastry & Bekhradnia, 2007). It is therefore important to give students a clear idea of how long you think it might take them to complete the assigned work, sharing your thoughts about each stage of the process, but also to make it clear that there will be big individual variations.

There is guidance on deciding on submission sizes in my book, Confident Assessment in Higher Education. First screenshot is an extract from specific info about portfolios (there is info for around 18 different assessment types) and the 2nd and 3rd are general ideas. uk.sagepub.com/en-gb/eur/co...

Edtech people always find a workaround! Impressive

Re. previous, "You were capable of writing an email in 2022 and you are capable of it now" is another one that can be cross-stitched for the office

Just 'graded' something (no actual grades were used here, just feedback, I know this is a luxury) where the person wrote that they “feel increasingly confident in my own approach.” and this person has achieved all I want to see in a university teacher and no grades could ever reflect this.

This might not have been a compliment if they came from the southwest of France 🤣

Thanks, @sandra-abegglen.bsky.social - it will be available on 26 Jan! @samillingworth.com

Everyone always proposes this as the answer without understanding that an oral exam requires very different skills than a closely argued research-driven essay. If you want to argue those skills don’t matter, then fine, argue that, but don’t pretend that oral argumentation is identical.

A brass band is a non-negotiable requirement for a university function. Ditto the comedians, and there was surely also some singing.

You can do this right now:

Think of a person who wrote a paper you love, whose work influenced or helped you, or has made your professional life better.

Search up their email address. Shoot them a quick email of thanks. It means so, so much. This is a rough time of year, share some joy.

Also, the promised revolution of computers transforming universities has not yet occurred.

YouTube video by Filmer om Lund och Skåne Steve Jobs visit at Lund University in 1985

Reminds me of this amazing video 'Steve Jobs visit atLund University in 1985'. So many questions... youtu.be/NT6oUsn3ANA?...

The Christmas issue of The BMJ is coming soon!

To get us in the mood we asked comedian @michaelspicer.bsky.social to have a look at last year's issue, and respond to some articles.

Inspired by this article evaluating the cognitive abilities of AI 🤖
www.bmj.com/content/387/...
#BMJChristmas

Parking surely gets a lookin here, or is that included under admin?

I’m a Professor. A.I. Has Changed My Classroom, but Not for the Worse.

This is a long, but uplifting, article about teaching humanities with the humanities as a guiding principle (credit to Carlo Rotella) www.nytimes.com/2025/11/25/m...

👆 This. These are adults, here to learn about an interesting subject from other adults. They are not here to learn ways to beg for accommodation for the unrelated things happening in their lives that we didn't imagine during course design. They already know (probably more than enough) about that.

Re accommodations for students: if they ask they can have them, no questions asked. Stop being a gross creep about it.

With even more awesome colleagues, we're now researching how trust is built between teachers and students. As I know you agree, co-construction is the way forward! But the discourse of 'students don't always know what's best' is still often heard. (2/2)

Thanks, and we agree about not being limited to one underserved/minorities group. Hearing these students has changed my whole approach to inclusion (they were awesome throughout the project) amd my understanding of structural barriers which teachers have agency to address. (1/2)