Skip to content

05 / Assurance of learning

Prove what was learned — not just what was marked.

Most programs can show a distribution of grades. Far fewer can show why a grade means what it claims. Assurance of learning is the established term; the Learning Evidence Chain is how CoAcademic intends to make it inspectable — outcome, requirement, AI-use rule, process evidence, artifact, explanation, rubric judgment, faculty decision, record.

01The problem

Grades are a summary, not evidence

Assurance of learning fails at the point where an outcome claim cannot be traced to anything a reviewer can inspect. That gap is usually a records problem rather than a teaching problem.

  • Rubrics drift

    The rubric in the handbook is often not the rubric that was marked against.

  • Band meaning is tacit

    Two markers can apply the same descriptor to visibly different work without either being wrong.

  • Moderation is undocumented

    The conversation happens; the record of what changed and why usually does not.

  • Evidence is unlinked

    An outcome report can rarely point at the passage, the process artifact, or the student explanation that justified a specific judgment.

02The record

The Assurance Record, link by link

Founding pilot scope

Founding Pilot Scope: the chain preserved end to end for one assignment, produced as a by-product of normal marking rather than a separate reporting exercise.

Intended outcome
The learning outcome this assignment was meant to evidence, stated before the work began.
Requirement and AI-use rule
What the task asked for, and whether AI use was restricted, limited, disclosed, or required.
Required process evidence
The planning, drafting, or annotation artifacts the assignment asked students to keep.
Artifact and explanation
The submitted work together with the student's account of specific choices in it.
Rubric version and calibration
Which criteria, weights, and band descriptors applied, and which anchor work set band meaning.
Moderation
Which scripts were second-read, where markers differed, and how the difference was resolved.
Faculty decisions
Every proposal a faculty member approved, revised, or rejected, recorded per criterion.
Release
Who released results, and when.

03Program level

Roll-up across courses and terms

Planned product

Program-level attainment views are planned product. They are not built, and nothing on this site should be read as a working analytics capability.

  1. 01

    Outcome mapping

    Criteria mapped to program learning outcomes so a course record contributes to a program claim.

  2. 02

    Attainment views

    Distribution of outcome attainment across cohorts, with the underlying evidence still reachable.

  3. 03

    Consistency signals

    Where marker agreement is weak, the program sees it before an external reviewer does.

  4. 04

    Review-ready export

    A package a program committee can read without asking faculty to rebuild it by hand.

Boundary
CoAcademic does not guarantee accreditation outcomes and does not represent any accreditor. It produces records; the judgment about their sufficiency belongs to your institution and its reviewers.

The Founding Assessment Assurance Pilot

One course team, one assignment, one confirmed rubric — audited, calibrated, reviewed against evidence, and decided by faculty. Enrolment is not open; the page explains the intended scope and what a pilot would require from both sides.