Stop inducting students into a system that doesn’t fit them? No, start

The data from the fifteen-institution pilot is useful. Only 31 per cent of incoming students have prior VLE experience. Thirty-nine per cent have never touched generative AI. Just 4 per cent prefer exams. Nearly half want one-to-one, face-to-face feedback. Students arrive from relational, scaffolded, hard-copy learning environments – and land in a system that hands them a login and a message that says “it’s all online, go and find it.”

All true. But the framing is dangerous – because the conclusion it invites is that higher education should bend itself towards what students arrive expecting.

Some of what the PAQ surfaces is just bad design – scattered timetabling that’s indefensible when most students are working, resources dumped on a VLE with no signposting. That should be fixed, and fixing it isn’t dumbing down. But it’s a different thing entirely to hand university to students on a plate – higher education really does involve a lot more independent study, and pretending otherwise helps nobody. The real problem sits between the two. Students want to know what they don’t know – and our system neither surfaces the gap on the way in, nor credits the work of closing it, which is why no time is ever allocated to doing so.

University is supposed to be different

The gap the PAQ documents is real. The question is which side of it you build from.

Independent study, assessment autonomy, finding your way around a big and complex environment, managing your own time – these are what university is for, not design flaws for a more “humane” onboarding to sand away. A student who leaves three years later having only ever experienced the relational, teacher-guided model they arrived with has been warehoused.

Becoming a student goes beyond “transition” – beyond familiarisation with a new environment – into becoming something new. The demands themselves, the ones university makes that school didn’t, are not the scandal. We assume readiness for those demands rather than building it – and then sort students by who arrived already holding the secret codes, passed down through families with a history of higher education.

Read that way, “students prefer face-to-face feedback” and “students dislike exams” become starting positions to develop from rather than preferences to honour. And 39 per cent arriving with no AI experience makes the equity case for scaffolding AI literacy before assessment, so the hidden curriculum doesn’t just advantage the already-fluent.

The mistake is not crediting it

Where the PAQ authors are right is that induction is a pedagogic challenge, not an administrative one. But the sector has known that for years and responded with pre-arrival modules hurled at students in August, and welcome weeks so packed with orientation that they’re more overwhelming than the socialising.

The reason those initiatives stay piecemeal, hobbyist and unloved is that we refuse to give becoming a student any curriculum time or credit. If it matters – and every strand of the PAQ data says it does – it should be structured, extended across the first year, and credit-bearing, with students self-assessing against a student attributes framework on arrival, planning their own development, and submitting a portfolio at the end of the year. Some of it – consent, academic integrity, rights and responsibilities – should be tested against minimum standards. Pre-arrival data like the PAQ is the diagnostic that makes that personal rather than generic. It’s the entrance ticket to the scaffolding, not a mandate to lower the building.

Students already intend to do this – help them plan

There’s also good evidence that structured planning is where intention becomes behaviour. Students consistently report strong intentions to engage – rating their intention to attend at eight out of nine – and then don’t. A Scottish trial of a volitional help sheet found that five minutes of “if-then” planning at the start of the semester lifted attendance and kept students attending for weeks longer, without a single threat or attendance policy in sight.

Helsinki goes further – every student builds and updates a personal study plan throughout their degree, with the initial plan worth two credits. Planning to be a student is treated as part of being one. And it works best when it’s done together, in groups, with guidance – not as another solitary online module.

Thus the headline finding – “students don’t want more tech, they want more humans” – is where the piece is simultaneously most right and most risky.

Right, because the belonging evidence is overwhelming – students who don’t connect don’t continue, and the relational drought in mass higher education is real. Risky, because if “more humans” means more staff-intensive, one-to-one, school-style contact, it’s a demand the sector can’t meet. Staff-student ratios are heading the wrong way, professional services are being cut, and personal tutoring is patchy at best – some students never meet their tutor beyond a vague introductory email.

But the humans don’t have to be staff.

The other humans should be students

Partly because belonging is built peer-to-peer. Partly because there’s no money for anything else. And mostly because it’s what the evidence says works.

In Finland, every new student is assigned to a small group led by trained, credited, often paid senior students – contact starts over the summer, runs through orientation and continues all year. It’s backed by legislation, co-governed by students’ unions, and when Aalto was asked for the secret sauce behind “the world’s best student experience,” tutoring was what it picked. The meta-analyses on peer tutoring show serious effect sizes on achievement, retention and belonging – and the tutors develop too.

The multi-car pile-up of agendas the PAQ surfaces – digital capability, assessment literacy, AI, expectations, connection – calls for a universal, in-person, peer-led vehicle that can carry all of them, rather than fifteen more online modules from fifteen different departments, because students will support each other if they’re enabled and trusted to do so.

So yes – act on the PAQ data, extend induction, treat onboarding as pedagogy. But the answer to students arriving unready for a system that’s different isn’t to make the system less different. It’s to make becoming a student something we teach, credit and resource – with students themselves doing much of the teaching.