Grade BTEC & T-Level Health & Social Care with AI

GradingPal is an AI grading assistant for teachers: upload student work and a rubric, and it drafts scores and specific, evidence-based feedback for you to review, edit, and release. In this use case, we'll follow a UK BTEC Health and Social Care coursework report graded against the Pass, Merit, and Distinction criteria grid that defines the qualification itself.

The problem

BTEC and T-Level marking doesn't work like a points rubric. Each criterion, coded things like A.P1 or A.D1, is simply met or not met, and the criteria are staged: Pass has to be secure before Merit counts, and Merit has to be secure before Distinction counts. A report can explain something well enough to pass and still fall short of Distinction for a completely different reason, not writing quality, but whether it evaluates instead of just explains.

That distinction is subtle and easy to blur under time pressure. A busy marker can read solid, detailed writing and wave it through to the top band, without checking whether the student actually weighed strengths against limitations and reached a reasoned conclusion, which is what Distinction-level criteria specifically require. And for a vocational qualification, that criterion-by-criterion evidence isn't just a courtesy to the student, it's what exam-board moderation actually checks.

This is where GradingPal helps. It grades against your assessment-criteria grid, respects the Pass-before-Merit-before-Distinction structure, and can pinpoint exactly which criterion is blocking a higher grade and why.

The assignment

Task 1 Principles of Person Centred Care Evaluation Report

This BTEC Health and Social Care task asks students to write a roughly 1500-word evaluation of person-centred care, built around two provided case studies of individuals with different needs. The report has to cover the values behind person-centred care, the NHS Core Values, Skills for Care Values, and the 6Cs, explain how professionals build trust and communicate effectively with each individual, define duty of care, and weigh individual rights and choices against that duty of care for both people in the case studies.

The assignment comes with its own official case-study and checklist handouts, distributed through Google Classroom, and a teacher can choose whether GradingPal sees those materials independently of what's shared with students. Every requirement in the task maps directly to a specific assessment-criteria code, from A.P1 through A.D1.

The rubric

Scoring Rubric

The rubric is built from six official BTEC criteria: A.P1 through A.P3 at Pass, explaining values and skills, communication, and duty of care; A.M1 and A.M2 at Merit, assessing the strategies professionals use and the balance between rights and duty of care; and A.D1 at Distinction, evaluating those strategies with a reasoned, comparative judgement. Each one is a binary gate, met or not met, rather than a points scale.

GradingPal grades to that exact structure. It's still your criteria, the official assessment-criteria codes for the qualification, applied consistently to every student's report, which is exactly the kind of criterion-by-criterion evidence a BTEC internal verification or moderation process expects to see.

The graded submissions

The teacher uploads the student's report, and GradingPal checks it against each criterion in order, distinguishing what a Pass criterion actually demands from what Merit or Distinction would additionally require.

Scoring with explanation

Evidence for scores

Personalized feedback for students

Previous slideNext slide

1 / 3

On A.P1, one report scores full marks even though GradingPal notes some of its value lists are imprecise, because it correctly reasons that A.P1 only requires explaining the relevant values and skills for both case studies, not evaluating them, and the explanation is genuinely there for both individuals. That report lands at 5 out of 6 criteria met.

The feedback names exactly which criterion is holding the grade back and why. It confirms the report is secure through Pass and Merit, applying values, communication, and duty of care to both individuals rather than staying generic, and then explains precisely what's missing for Distinction: the student explains useful care approaches but doesn't yet weigh how effective they are for each specific case. A second feedback card quotes the student's own language directly, noting that a line like 'active listening can become tiring for the practitioner' is a good start toward evaluation, then shows what turning it into a genuine comparative judgement would look like: which approach is most effective for this individual, in this situation, and why.

That's coaching built from the student's own sentences, not a generic note to 'evaluate more.' And nothing goes to the student until the teacher says so; every score and comment can be copied out, edited, or released as is.

Classwide analytics

Score table

Score distribution

Strengths & weaknesses

Previous slideNext slide

1 / 3

Across the full class, the dashboard doesn't just report a mean. It recognizes and names a split: the class average sits at 88 percent, but the summary flags that two students scored below 17 percent while the other 47 scored above it, a genuinely bimodal result a flat average would hide entirely.

The per-criterion breakdown shows exactly where that split comes from. A.P1 through A.P3 sit at 94 to 96 percent, nearly every student clears Pass. Merit holds at 86 to 92 percent. Then A.D1, the Distinction criterion, drops to 63 percent, the same gap the individual report ran into. Seven of forty-nine students can explain and assess strategies but stop short of a full evaluation with weighed trade-offs and a clear conclusion, and thirteen don't fully cover how information is documented and kept confidential within the communication requirement. Every one of those numbers points at the same fix.

The outcome

Here's what changes when BTEC or T-Level coursework runs through GradingPal:

The teacher gets criterion-by-criterion evidence for every student, in the exact Pass, Merit, Distinction structure the qualification requires, which is the record a moderation process actually needs to see.

The student gets told exactly which criterion is in reach and what it would take to close it, you've explained this well, now weigh it and reach a conclusion, grounded in their own words rather than a generic push to write more.

And the class gets one consistent finding at every level, individual, evidence, and criterion percentage, so a mini-lesson on evaluative writing lands exactly where the whole cohort actually needs it.

That's the point of a criterion-referenced qualification in the first place. We don't just want a mark at the top of the page. We want a clear, defensible answer to which specific criteria a student has met, and what stands between them and the next one. GradingPal makes it practical to give that answer for every student, on every criterion, every time.