The assessment integrity TEQSA expects rests on two standards read together: Standard 1.4, which requires assessment to be aligned to learning outcomes and to confirm that students have achieved them, and Standard 5.2, which requires the provider to protect the integrity of the awards it confers. In practice that means constructive alignment in every unit, moderation that changes marks, at least some assessment points that are secured against outsourcing, and records showing that academic staff exercised judgment rather than processed submissions.
This article walks through what an assessor actually reads when testing those two standards, from unit outlines to moderation minutes, and what good practice looks like in a private provider. It draws on fifteen years of TEQSA registration and accreditation work, most of it with providers whose assessment was sound but whose evidence of it was not.
What assessment integrity TEQSA assesses under Standards 1.4 and 5.2
Standard 1.4 of the Higher Education Standards Framework (Threshold Standards) 2021 is about learning outcomes and assessment. It requires that learning outcomes are specified, consistent with the AQF level and field, and that assessment methods are capable of confirming that all specified outcomes have been achieved. It also requires that grades reflect the level of attainment and that assessment is conducted with integrity.
Standard 5.2 sits in a different domain, Institutional Quality Assurance, and it is about academic and research integrity at the institutional level. It requires policies, training, monitoring and action on breaches. Assessors treat the two as a pair. A provider can have beautifully aligned assessment and still fail 5.2 if nobody is watching for contract cheating, and it can have a comprehensive integrity policy and still fail 1.4 if the assessments cannot show that outcomes were met.
The distinction matters because providers tend to answer both with the same document, usually an academic integrity policy. That policy is evidence of a rule. Neither standard is satisfied by a rule. Both are written in the present tense, and both ask for evidence that the rule operates.
Constructive alignment: what the unit outline has to show
The first thing an assessor reads is the unit outline, and the first thing they look for is the line between course learning outcomes, unit learning outcomes and assessment tasks. Every assessment task should be traceable to one or more unit outcomes, every unit outcome to one or more course outcomes, and the course outcomes to the AQF level descriptors for the qualification. Our article on AQF levels for course designers sets out those descriptors in plain terms.
The common failure is a mapping table that is complete and a set of tasks that is not. A capstone outcome about independent judgment in complex settings is mapped to a multiple-choice quiz. A level 9 outcome about advanced research skills is assessed by a reflective journal with no research component. Assessors do not need to be subject experts to see this. They read the verb in the outcome, read the task, and ask whether the task could possibly produce evidence of that verb.
The second failure is weighting. If sixty per cent of a unit's marks come from a group project and the outcomes are individual competencies, the provider cannot say with confidence that any individual student has achieved them. In my experience assessors raise this more often than providers expect, because it goes directly to whether the award certifies what it claims to certify.
Moderation that leaves a trace
Standard 1.4 expects grading to be consistent and reliable, and the mechanism providers use for that is moderation. TEQSA accepts many models. What it does not accept is moderation that exists as a policy paragraph and nothing else.
Assessors ask to see moderation records for specific units in specific teaching periods. A record that says "moderation completed, no changes" for every unit in every period is read as a record of no moderation. Real moderation changes marks, tightens a rubric, sends a task back for redesign, or notes that two markers disagreed and how the disagreement was resolved. The best moderation records I have seen are short but specific: which samples were reviewed, by whom, what was found, what was changed, and who signed off.
Pre-assessment moderation is often missing entirely. Providers moderate marks after the fact but never review the task, the rubric or the marking guide before release. That is where alignment problems are cheapest to fix, and an assessor who sees only post-hoc moderation will infer that alignment is being checked only after students have already been assessed against a flawed task. The assessment standards documentation that satisfies TEQSA is usually the same documentation a conscientious unit coordinator would produce anyway; the work is in keeping it.
Secured assessment in the age of generative AI
Since 2023 TEQSA has published a series of resources on assessment in the age of artificial intelligence, beginning with Assessment reform for the age of artificial intelligence and followed in 2025 by Enacting assessment reform in a time of artificial intelligence. These are guidance, not Threshold Standards, but they tell you how TEQSA now reads 1.4 and 5.2. The regulator does not regard detection software as sufficient. It expects program-level assurance, meaning that across a course there are enough secured assessment points, such as supervised tasks, vivas, practicals or in-class work, to give confidence that the graduate has personally achieved the outcomes.
The June 2024 request for information, which required every provider to lodge a credible institutional action plan on generative AI, made this concrete. Assessors now ask where the secured points in a course are and how the provider knows they are working. A course that is assessed entirely by unsupervised written submissions, with a plagiarism checker as the only control, will draw a request for further information on both standards.
The practical response is a course-level assessment map that marks which tasks are secured, confirms that every course learning outcome is evidenced at least once through a secured task, and is approved and reviewed by the academic board. That single document answers a large part of what TEQSA is asking, and it is the document most providers do not have. Our guide to academic integrity compliance under TEQSA covers the policy and case-management side in more detail.
Academic judgment, not processing
The theme running through the assessment integrity TEQSA looks for is judgment. Standard 1.4 assumes that qualified academic staff design tasks, set standards, mark against them and adjust when the evidence warrants it. Standard 5.2 assumes that when integrity is breached, someone with academic authority decides what it means and what follows.
Assessors look for the places where that judgment is visible. Grade distributions reviewed by a course committee, with outliers explained. Integrity cases recorded with the decision, the reasoning and the penalty, and an annual report of them going to the academic board. A rubric revised because markers found a criterion unworkable. Where assessment appears to run on automated workflows with no human decision at any point, the assessor will ask who is exercising academic judgment, and the answer needs to be a named role with the qualifications to do so.
This is where the shift from Confirmed Evidence Tables to self-assurance bites. A self-assurance report that asserts robust assessment is worth nothing without the records behind it, and generic assertions, of the kind AI drafting tools produce readily, now attract more scrutiny rather than less. Evidence of assessment standards is evidence of things that happened, and the article on evidence of student learning explains what that looks like across a course.
What I would put in front of an assessor
If I were preparing a provider for a course accreditation or a renewal, I would assemble, for two or three units chosen to represent the course, the unit outline with its alignment map, the assessment tasks and rubrics, the pre-release moderation record, samples of marked work across the grade range, the post-marking moderation record and the grade distribution as reported to the course committee. Alongside that I would put the course-level assessment security map, the academic integrity policy, the register of integrity cases for the last two years and the annual integrity report to the academic board.
That is not a large body of material. It is the ordinary output of a provider that takes assessment seriously, collected in one place. The providers who struggle with the assessment integrity TEQSA tests are not those with weak assessment but those who cannot find the evidence that their assessment is strong, and the fastest path through TEQSA is a complete, specific and true account of what already happens.
Download the New Course Readiness Checklist
— a one-page test of alignment, moderation and assessment security against Standards 1.4 and 5.2, drawn from our TEQSA registration and accreditation work with private providers. Get the checklist
Want the full article?
Enter your email for free access to the rest of this guide and our TEQSA resource library.
Frequently asked questions
Does TEQSA require every assessment to be supervised?
No. TEQSA's guidance favours program-level assurance, meaning enough secured assessment points across a course to confirm that each graduate has personally achieved the learning outcomes. Unsupervised tasks remain acceptable as part of a designed mix.
Is detection software enough to satisfy Standard 5.2?
No. TEQSA has said it does not regard detection software as sufficient on its own. It expects policy, training, secured assessment design, monitoring and a documented response to breaches, overseen by academic governance.
How much moderation evidence does TEQSA want?
Assessors typically sample specific units and teaching periods. They expect to see what was reviewed, by whom, what was found and what changed, both before tasks are released and after marking. Records that show no changes anywhere are read as evidence that moderation is not occurring.
Which body should own assessment integrity in a private provider?
The academic board. Standard 6.3 places responsibility for academic quality and integrity with academic governance, so assessment maps, moderation outcomes and integrity case reports should flow to the academic board and be visible in its minutes.
Dr Brendan Moloney is CEO of Darlo Higher Education, Australia's largest specialist TEQSA consultancy. He holds a PhD from the University of Melbourne, is a Cambridge University Press author on governance in higher education, and has advised private providers on registration and course accreditation for more than fifteen years.