Clustering in VET: Starting With Similarity, Not Convenience

A comprehensive analysis of what clustering in vocational education and training actually requires: the similarity test that must come before any grouping decision, the assessment-condition filter that follows it, the principles of assessment and rules of evidence that govern every clustered design, the training package compliance framework that anchors it, and the honest case for why the sector needs better clustering rather than more of it.

A Conversation the Sector Keeps Having With Itself

There is a conversation the VET sector has been having with itself for a long time. It surfaces in professional development sessions, features in best practice guides, earns a well-meaning mention in audit feedback, and then quietly disappears when implementation requires sustained effort. The sector returns to doing things the way it has always done them, and the conversation waits for the next opportunity.

That conversation is about clustering: the deliberate grouping of units of competency for integrated delivery, integrated assessment, or both. It is not a new concept. It is not a contested one. Most practitioners, when pressed, will acknowledge that it makes sense. Yet across a significant number of registered training organisations, the unit-by-unit model remains the default, often defended by objections that are more comfortable than they are correct.

The deeper problem, however, is not the reluctance to cluster. It is the misunderstanding of what clustering actually requires. Clustering is frequently spoken about as though grouping any set of units that happen to occur in the same workplace or the same qualification is sufficient. It is not. Clustering must begin with a single, non-negotiable test: do these units share sufficiently similar skills, knowledge, and behaviours that a single, coherent assessment task can generate genuine evidence for all of them? That question must be answered before anything else. If it cannot be answered affirmatively, clustering is not appropriate, regardless of how convenient the grouping might appear.

This article sets out what clustering requires, where it works, where it does not, and the compliance framework that governs every clustering decision an RTO makes.

1. The First Principle: Similarity Comes Before Everything Else

The starting point for any clustering decision is an analysis of similarity across the units under consideration. This means examining, at the level of the training package, whether the skills being performed, the knowledge being applied, and the behaviours being demonstrated across the candidate units are genuinely close enough to be assessed together through a single task or a coherent sequence of related tasks.

This is a more demanding test than it might appear. Units that exist within the same qualification, or that are frequently delivered in the same workplace context, are not automatically candidates for clustering. The question is not whether they occur in the same environment. The question is whether the competency being demonstrated in one unit is substantively similar to the competency being demonstrated in another, to the degree that a single assessment activity can produce credible evidence for both without either becoming superficial.

Consider aged care as an example. Units that address personal care, infection control procedures, and safe manual handling techniques share a deep structural similarity. The skills involved (observing, responding, applying safe practice, and communicating with the person in care) draw on overlapping knowledge of human anatomy, care standards, and risk management. The behaviours expected of a competent practitioner are consistent across all three: attentiveness, precision, and person-centred practice. An assessor observing a learner provide personal care in a residential aged care setting can, with a well-designed tool, collect evidence across all three units simultaneously, because the task genuinely draws on all of them.

Now consider what happens if an RTO attempts to cluster those same personal care units with a unit on completing financial or administrative documentation. The skill sets are fundamentally different. One requires physical care techniques and interpersonal attunement. The other requires data entry, accuracy with figures, and compliance with administrative systems. They may both occur in the same facility. They do not occur in the same task. No single authentic workplace activity draws on both in a way that produces sufficient, valid evidence for each. Grouping them as a cluster would not reflect integration. It would reflect a misunderstanding of what integration means.

2. What Clustering Is and What It Is Not

Clustering, in the context of VET training and assessment, is the deliberate grouping of two or more units of competency for the purposes of integrated delivery, integrated assessment, or both. The defining characteristic of a legitimate cluster is that the units share genuine overlap in the skills, knowledge, and behaviours required for competent performance, such that a single, well-designed assessment task can address the requirements of all units in the cluster simultaneously.

This is the essential distinction between clustering and convenience grouping. Bringing units together because they appear in the same qualification, because scheduling them together reduces timetabling complexity, or because the workplace in which learners are employed happens to involve all of them at some point during a working week: none of these constitute a basis for clustering. The basis for clustering is demonstrated similarity in what is being assessed, grounded in the training package, confirmed through mapping, and verified by qualified assessors with current industry knowledge.

It is also important to distinguish clustering from co-delivery. Units can be delivered together, using shared resources, shared contexts, and shared learning activities, without being assessed together. Co-delivery is a scheduling and pedagogical decision. Clustering is an assessment design decision. The two are not the same, and conflating them creates compliance risk. An RTO that delivers two units concurrently but assesses them separately is not clustering. An RTO that assesses two units through a single, integrated task that has been mapped to the requirements of both is.

3. The Second Filter: Assessment Conditions, Environment, and Unit Requirements

Once the similarity of skills, knowledge, and behaviour has been established, the second analysis concerns whether the assessment conditions, assessment environment, and other specific requirements of each unit allow them to be assessed together. This filter applies after similarity has been confirmed, not instead of it. It is the second question, not the first.

Even units that share substantial skill and knowledge overlap may be difficult or impossible to cluster if their training package assessment requirements impose incompatible conditions. These incompatibilities are worth examining specifically.

3.1 Incompatible Assessment Environments

Some units require assessment in a real workplace environment as a non-negotiable condition. Others permit or require assessment in a simulated environment with specified parameters. Where two units that are otherwise similar in skill and knowledge content impose different environmental requirements, they cannot be fully assessed through the same activity. An RTO cannot satisfy a workplace observation requirement for one unit while simultaneously satisfying a simulation requirement for another through the same task. The assessment design must address each condition separately, which effectively means the cluster must incorporate distinct assessment activities rather than a single integrated one.

3.2 Incompatible Instance Requirements

Training package units frequently specify the number of times a skill must be demonstrated across assessments. Where two units require different numbers of instances, or where the range of conditions required for one unit is broader than the task can accommodate, the cluster must be designed with supplementary activities that address the outstanding requirements. A cluster that produces one workplace observation cannot satisfy a unit requiring three demonstrated instances across different client groups simply because one observation was conducted in a workplace setting. The mapping must address this gap explicitly.

3.3 Incompatible Equipment or Resource Requirements

Units within the same broad field sometimes specify assessment using particular equipment, particular materials, or in the presence of particular personnel. Where these requirements differ between units that are otherwise similar, the cluster design must either identify an assessment setting that satisfies all requirements simultaneously or acknowledge that the units cannot be fully assessed within a single cluster activity and design accordingly.

3.4 Incompatible Supervision or Third-Party Requirements

Some units specify that assessment must be conducted or verified by a person with specific industry qualifications or a particular role relationship to the learner. Where two otherwise similar units impose different supervision or verification requirements, those requirements must each be met within the cluster design. A single third-party report from a supervisor who meets the requirements of one unit but not another does not satisfy both.

4. When Units Are Too Different to Cluster

There are units that, regardless of how they appear in a qualification or how they are delivered, are not appropriate candidates for clustering because the skills, knowledge, and behaviours required are fundamentally different. Recognising this clearly is as important as recognising what clustering makes possible.

Units that operate across entirely different competency domains cannot be joined through assessment simply because a learner must eventually perform both in a job role. The fact that a business administration learner will eventually use spreadsheet software and also conduct customer service calls does not mean that units addressing those respective competencies are candidates for an integrated assessment task. The skills involved are different in kind. The knowledge base is different. The behaviours expected of a competent performer are different. No single task can simultaneously generate credible evidence for both without one or the other becoming cursory.

Units that sit at different levels of complexity within the AQF also warrant careful scrutiny before clustering is considered. A unit requiring routine application of a procedure and a unit requiring analysis and judgement under complex conditions may share some surface vocabulary, particularly in a broad area like communication or workplace safety, but the depth of knowledge, the sophistication of judgement, and the conditions under which performance must be demonstrated may be too different to support a single integrated assessment.

Similarly, units from different industry or sector contexts should not be clustered without very careful analysis. A unit from a health training package and a unit from a community services training package may share broad themes around person-centred practice, but the specific regulatory requirements, the knowledge base, and the performance conditions relevant to each sector differ sufficiently that an assessor would need to satisfy themselves, with detailed mapping, that a single task could genuinely produce sufficient evidence for both.

5. The Principles of Assessment: What They Require in a Clustered Design

The Standards for RTOs 2025 require that all assessments conducted by or on behalf of an RTO must meet the principles of assessment and the rules of evidence. These requirements apply with equal force to clustered assessment as to any other form. They are not aspirational guidelines. They are compliance requirements, and understanding them is the starting point for any cluster design, not a final check after the design is complete.

5.1 Validity

Validity requires that assessment tasks measure what they are intended to measure, under conditions that reflect the context in which competency must be demonstrated. When clustering is founded on genuine similarity of skills, knowledge, and behaviour, it strengthens validity: the integrated task reflects the integrated nature of competent performance in a way that a fragmented, unit-by-unit approach cannot. When clustering is founded on convenience rather than similarity, it undermines validity because the task no longer corresponds to a coherent piece of work that actually draws on all of the competencies being assessed.

Validity also requires full coverage. A cluster that groups units but only partially maps evidence requirements against each unit's performance criteria, knowledge evidence, and assessment conditions is not more valid for being grouped. It is deficient because it is incomplete. Clustering reorganises the mapping obligation. It does not reduce it.

5.2 Reliability

Reliability requires that an assessment produces consistent outcomes across different assessors, different learners, and different occasions. In a clustered design, reliability depends on the quality of the assessment tools and the clarity of the judgement criteria. If two qualified assessors observing the same learner performing the same task would reach different conclusions about which unit requirements have been met, the cluster has a reliability problem that the design must address.

This means that clustering demands more precise documentation of evidence requirements, not less. Assessors must know exactly which performance criteria and knowledge evidence points are being addressed through each component of the assessment activity, and the assessor guide must make those links explicit and verifiable before the assessment takes place.

5.3 Flexibility

Flexibility requires that assessment accommodate the diverse circumstances of learners, including different workplace contexts, prior experience, and evidence pathways. Well-designed clustering supports flexibility because holistic, task-based assessment is more readily adapted to different workplace settings than a rigid battery of unit-specific activities. Flexibility also requires, however, that learners who cannot participate in the standard clustered activity have access to alternative assessment pathways that are mapped with equivalent rigour.

5.4 Fairness

Fairness requires that assessment does not advantage or disadvantage learners on the basis of factors unrelated to competency. In clustered assessment, fairness considerations include ensuring that the task is accessible to all learners in the cohort regardless of their specific workplace context, that evidence requirements are communicated clearly and transparently in advance, and that learners who demonstrate competency in some but not all clustered units are assessed and recorded accurately without being penalised by the cluster structure.

6. The Rules of Evidence: How They Apply to Clustered Assessment

The rules of evidence govern the quality of evidence collected through assessment. They require that the evidence be valid, sufficient, authentic, and current. Each rule imposes specific obligations on assessors and on those who design assessment tools for clusters.

6.1 Valid Evidence

Evidence is valid when it directly addresses the requirements of the unit being assessed. In a cluster, the assessor must be able to trace each piece of evidence back to specific requirements in each clustered unit. Where a cluster is founded on genuine similarity of skills and knowledge, this tracing is natural and meaningful: the same piece of performance evidence genuinely addresses overlapping requirements across units because the skills being demonstrated are genuinely related. Where a cluster is founded on convenience, the tracing becomes forced, and the evidence is likely to be superficial for at least some of the units.

This is why similarity is the entry point, not an afterthought. Valid evidence in a cluster flows naturally from genuine skill overlap. It cannot be manufactured through clever mapping of dissimilar requirements.

6.2 Sufficient Evidence

Evidence is sufficient when it covers the full range of performance criteria and knowledge evidence required by each unit, and when the volume and nature of evidence provide confidence that competency is consistent rather than incidental. Sufficiency is one of the most common areas where clustering fails in practice, not because the approach is flawed but because assessors assume that a single observation covers more than it demonstrably does.

Sufficiency requires that the cluster design specifies, in advance, how many instances of evidence are required, what range of conditions they must cover, and how knowledge evidence is to be collected. A cluster that produces one integrated observation and no documented knowledge evidence is not sufficient for most units, regardless of how well-designed the task was or how authentic the workplace setting.

6.3 Authentic Evidence

Evidence is authentic when it can be verified as the learner's own work and as a true reflection of their performance. In clustered assessment, particularly where third-party reports or workplace supervisor verification are used, the RTO must have processes to confirm that the evidence reflects the learner's actual demonstrated performance rather than a generalised account of what someone in that role typically does. Third-party reports must be specific, dated, and tied to observed behaviour rather than character endorsements.

6.4 Current Evidence

Evidence is current when it reflects the learner's present level of competency. This is a relevant consideration in clusters that include recognition of prior learning pathways or that draw on historical workplace evidence. Evidence collected more than approximately twelve months prior, depending on the nature of the units and any relevant industry or regulatory change, may not satisfy the currency requirement without supplementary assessment activities to confirm that competency has been maintained.

7. Training Package Compliance: The Non-Negotiable Framework

Clustering must be anchored in the training package. This is not a preference: it is a compliance requirement. The performance criteria, knowledge evidence, assessment conditions, and foundation skills specifications in each unit define what must be assessed. A clustering decision that groups units without confirming that the shared assessment activity genuinely addresses all of those requirements across every clustered unit is non-compliant, regardless of how practical or well-intentioned the grouping appears.

7.1 Performance Criteria Coverage

Every performance criterion in every clustered unit must be addressable through the cluster's assessment design. The mapping document, which is a required design artefact and not an optional quality measure, must show which assessment activity, task component, or supplementary evidence item addresses each performance criterion. Where a performance criterion cannot be addressed within the core clustered activity, an additional assessment instrument must be incorporated into the cluster design. A performance criterion that is left unmapped is a performance criterion that is not being assessed.

7.2 Knowledge Evidence Requirements

Knowledge evidence requirements must be assessed for each unit within the cluster. Clustering reduces duplication where units share common knowledge requirements, and this is one of its genuine efficiencies. Where two units both require knowledge of relevant legislation or safe work procedures within the same field, a single knowledge activity addressing that requirement is sufficient for both. Where one unit requires knowledge of specific technical procedures that another does not, that knowledge must be assessed separately within the cluster design. Clustering does not permit knowledge requirements to be left unaddressed on the basis that a related unit covers something similar.

7.3 Assessment Conditions

Assessment conditions specified in the training package, including requirements for workplace observation, the number of instances of performance, the range of conditions across which competency must be demonstrated, and the use of specific equipment, environments, or personnel, must be met for each clustered unit. Where two otherwise similar units impose different assessment conditions, the cluster design must address both sets of conditions, either through a single task that satisfies both or through separate activities within the cluster. Where conditions cannot be reconciled within a coherent cluster, the units should be assessed separately.

7.4 Foundation Skills

Foundation skills, including literacy, numeracy, oral communication, and digital literacy, are embedded in training package units and must be reflected in assessment design. Clustering provides an opportunity to assess foundation skills in an integrated and contextualised way. Where units share similar foundation skill requirements, the cluster can address these naturally through the integrated task. The cluster design must confirm, however, that the foundation skills relevant to each unit are genuinely observable within the clustered task and not assumed to be covered by proximity.

8. When Clustering Is Appropriate

Clustering is appropriate when the following conditions are all present. These conditions are sequential: the first must be confirmed before the second is examined, and both must be confirmed before the third.

First, similarity of skills, knowledge, and behaviour. The units share genuinely similar skills, such that a learner performing one unit draws on the same or closely related capabilities as a learner performing another. The units share substantially overlapping knowledge requirements, meaning that the knowledge base informing competent performance in one unit is the same or closely related to the knowledge base informing competent performance in another. And the behaviours expected of a competent practitioner across the candidate units are consistent in character, even if they differ in specific application.

Second, compatible assessment conditions and requirements. The assessment conditions specified for each unit can be met within the cluster design, either through a single shared activity or through a coherent set of related activities within the cluster. The assessment environment required for each unit is compatible, and a single setting can satisfy the environmental requirements of all clustered units. And the number of instances, range of conditions, and any equipment or personnel requirements specified in each unit can all be addressed within the cluster design.

Third, design and assessor readiness. The assessor holds the qualifications, industry currency, and vocational competency to make holistic judgements across all clustered units. A mapping document exists that explicitly links every performance criterion and knowledge evidence requirement of each unit to a specific component of the cluster's assessment design, and this document has been reviewed by a second qualified assessor. And the learner has been provided with clear information about which units are being assessed through the cluster, what evidence is required, how judgements will be made, and what the outcome will be if competency is demonstrated in some units but not others.

Clustering is particularly effective in contexts such as aged care and disability support, where personal care, communication, and safety practices draw on the same skills and knowledge and occur simultaneously in practice. It is also well suited to qualifications in early childhood education, where observation, documentation, and child-directed practice share deep structural similarity, and to trade qualifications where planning, materials handling, and safe work practices within a single trade specialisation draw on closely related technical knowledge.

9. When Clustering Is Not Appropriate

Clustering is not appropriate under the following conditions. RTOs that proceed with clustering in these circumstances face genuine compliance and quality risks that an audit will identify.

The units do not share sufficiently similar skills, knowledge, or behaviours. Grouping units because they appear in the same qualification or because the learner's employer happens to require both of them is not a basis for clustering. Proximity in a timetable is not similarity in a training package.

The assessment conditions specified in the training package for individual units are incompatible and cannot be reconciled within a single cluster design. Where one unit requires a real workplace environment, and another requires a controlled simulation with specific parameters, those conditions cannot be merged without compromising validity for at least one unit.

The assessor does not hold the required industry experience, vocational competency, or currency across all units in the cluster. An assessor qualified to assess competency in one area of the cluster but not another cannot make holistic judgements across the full scope of its requirements.

The combined assessment activity required to address all units in the cluster cannot be designed as a coherent, authentic workplace task. If the instrument runs to dozens of pages of observation criteria and knowledge questions that feel assembled rather than integrated, the cluster has lost the coherence that makes it valid.

The cluster is being designed primarily to reduce assessment burden rather than to reflect genuine skill integration. Efficiency is a welcome outcome of good clustering. It is not a design objective. A cluster designed backwards from a desired reduction in the number of assessment activities will almost certainly produce insufficient evidence and weak compliance outcomes.

The learner's individual circumstances do not support the standard clustered activity. Learners completing qualifications through non-workplace pathways, or those requiring additional support and flexibility, may be better served by a modified approach that still reflects holistic practice but accommodates their specific context.

Where doubt exists about whether units are sufficiently similar to cluster, the default position should be to assess them separately and deliver them together. Co-delivery without clustered assessment is not a failure. It is appropriate professional judgement.

10. Partial Competency Within a Cluster: A Design Requirement, Not a Problem

One of the most frequently raised concerns about clustering is what happens when a learner demonstrates competency in some units within a cluster but not others. This concern is sometimes treated as an argument against clustering. It is, in fact, a design requirement that every cluster must be built to accommodate.

Competency is determined unit by unit, and a cluster does not change that requirement. If a learner completes a clustered assessment and demonstrates competency in two of three clustered units, the assessor records competency for those two units and not yet competent for the third. The learner undertakes a supplementary or alternative assessment for the third unit. The cluster structure does not override the unit-level outcome.

This requires that the cluster design anticipates partial outcomes from the beginning. Assessment tools must be structured so that evidence for each unit can be evaluated independently, even when it is collected together. The assessor guide must provide clear criteria for what constitutes a competent outcome for each individual unit. The RTO's recording systems must accommodate unit-level outcomes from a cluster-level assessment activity.

RTOs that avoid clustering because of partial competency concerns have allowed administrative complexity to override genuine compliance and quality improvement. The complexity is real but manageable. The validity and sufficiency risks of not clustering, in contexts where units are genuinely similar and assessment conditions are compatible, are not manageable by avoidance.

11. The Objections, Honestly Examined

11.1 The Funding Argument

RTOs funded per unit of competency face legitimate questions about how clustering interacts with reporting requirements. This is an operational concern worth taking seriously. It is also, in most cases, solvable. Clustering reduces per-assessment burden significantly. The administrative question of how units are reported does not require that the assessment of them be conducted in isolation. Most funding bodies do not mandate disconnected assessment design, and where operational constraints exist, they call for a creative solution rather than wholesale rejection of a better approach.

11.2 The Partial Competency Argument

What happens when a learner is competent in one unit but not the others in a cluster? As addressed above, this is a design and documentation question. The answer is better design and clearer assessor guidance, not a return to unit-by-unit assessment. Every assessment method produces partial competency scenarios. Clustering does not create this challenge. It makes the design obligation to address it explicit.

11.3 The Complexity Argument

Clustering requires stronger mapping skills, more rigorous design, and assessors who are confident in making holistic judgements. This is true. It is also an argument for investing in assessor capability, not for defaulting to a model that produces weaker outcomes. The sector has a professional development infrastructure. Using it to build clustering capability is a more productive response than using complexity as a permanent deferral.

Meanwhile, industry continues to report that graduates struggle to integrate skills in the workplace. The sector must eventually ask whether this is a learner issue or a design feature of the system it has built. Clustering, done well, is one part of the answer. Done poorly, it is another problem. The difference lies in starting with the right question: not how many units can we group together, but which units are similar enough to deserve it.

12. Clustering and the Standards for RTOs 2025

The Standards for RTOs 2025 do not use the word clustering. They do not need to. The requirements for valid, sufficient, authentic, and current evidence, for assessment that meets the principles of assessment, for training and assessment practices that are current and reflect industry conditions, and for AQF outcomes that reflect genuine competency collectively create a regulatory environment in which well-designed clustering is not merely permitted but, in many contexts, demanded by the logic of the standards themselves.

An RTO delivering a qualification in a field where skills are genuinely integrated in practice and assessing every unit in isolation is producing evidence that does not reflect the conditions of the relevant environment. That is a validity problem, regardless of how carefully each unit is mapped in isolation.

Conversely, an RTO that clusters units without confirming similarity, without mapping evidence requirements to each unit, and without producing assessor tools that support consistent and defensible judgements is not meeting the reliability, sufficiency, or training package compliance requirements. Calling an inadequately designed collection of activities a cluster does not make it one. The standard applies to the quality of the design, not the label attached to it.

13. Building Clustering Capability in Your RTO

The skills required to design effective clusters are not specialised beyond what good assessment design already demands. They include the ability to read training packages analytically, to map evidence requirements completely and precisely, to design tasks that reflect authentic workplace conditions, and to make holistic judgements that are consistent and defensible across assessors.

For RTOs seeking to develop this capability, the following approach provides a practical starting point.

Identify one qualification where units with genuine skill and knowledge similarity are already being delivered together. Do not begin with the most complex qualification in your scope. Begin with the clearest case.

Convene a working group of assessors with current industry experience in the relevant field. The analysis of skill and knowledge similarity must be grounded in knowledge of how the work actually happens, not in a reading of the qualification structure alone.

Conduct the mapping exercise before designing any assessment tools. The mapping must confirm, for each candidate unit, that the cluster's proposed assessment activity can address every performance criterion, knowledge evidence requirement, and assessment condition. If the mapping reveals gaps, revise the cluster scope before proceeding.

Design assessor guides that identify, for each unit in the cluster, exactly which evidence collected through the assessment activity addresses which requirements. The guide must enable a second qualified assessor to conduct the same assessment consistently and reach the same judgements.

Have the design reviewed by a second qualified assessor before it is used for the first time. This is not a formality. It is the reliability check that the standards require.

After the first use, review the outcomes. Where evidence was difficult to map to specific requirements, revisit the cluster design. Where partial competency outcomes occurred frequently in one unit, examine whether the assessment activity genuinely draws on that unit's requirements or whether the cluster scope needs adjustment.

Conclusion: The Compliance Case Rests on Getting the Foundation Right

The argument for clustering in VET is sometimes presented as a pedagogical preference or a productivity strategy. It is both of those things. But the more foundational case is a compliance one, and that compliance case rests entirely on the quality of the design.

Clustering that begins with genuine similarity of skills, knowledge, and behaviour, that confirms compatible assessment conditions before proceeding, that maps every unit requirement to the assessment design, and that produces assessor tools capable of supporting consistent and defensible judgements: that is clustering that fulfils the principles of assessment and the rules of evidence. It is more valid, more authentic, and more defensible than the unit-by-unit approach it replaces in those contexts where it is genuinely appropriate.

Clustering that begins with the question of which units are convenient to group together, that treats the training package as something to be mapped after the design is settled, and that produces voluminous but internally inconsistent assessment instruments: that is not clustering. It is a compliance risk wearing the name of a good idea.

The sector does not need more clustering. It needs better clustering, founded on a clear understanding of what makes units genuinely similar, what assessment conditions allow or prevent integration, and what the training package actually requires. The work does not wait for the next unit. But the design work has to come first.

References and Further Reading

Australian Skills Quality Authority. Standards for RTOs 2025: Outcome Standards, Compliance Requirements and Credential Policy. https://www.asqa.gov.au/rtos/2025-standards-rtos

Australian Skills Quality Authority. Practice Guides for the 2025 Standards for RTOs, including guidance on assessment. https://www.asqa.gov.au/rtos/2025-standards-rtos

Australian Skills Quality Authority. Principles of Assessment and Rules of Evidence (Appendices to the Users' Guide). https://www.asqa.gov.au

National Vocational Education and Training Regulator (Outcome Standards for NVR Registered Training Organisations) Instrument 2025. Federal Register of Legislation.

Australian Qualifications Framework. AQF levels and volume of learning. https://www.aqf.edu.au