Can a strong domain make up for a weak one?

Judgment is how well a person decides once AI has given the answer. AIRS-J/1 is the standard AIR APAC publishes for judgment in AI-enabled work. The Standard reads it in nine dimensions, grouped in four domains: Structure, Test, Engage and Perform. A certification decision applies six rules, and all six must be met (Section 12.2). The answers below quote AIRS-J/1 v2.3, in force since 19 October 2026. The full text is at https://airapac.org/standard/v2-3/.

Can a strong domain make up for a weak one?

No. Section 12.1 says: "Strength in one domain does not offset weakness in another". It adds that domain scores "are not averaged away into a single number for the purpose of the pass decision". The Standard calls this rule non-compensatory.

What does "non-compensatory" mean in plain words?

It means that one part of a result cannot pay for another part. A high score in one place does not cancel a low score in another place. Each part must reach its own minimum (Sections 12.1 and 12.2).

What must each domain reach?

Rule 2 of Section 12.2 sets a domain floor. Each of Structure, Test, Engage and Perform must be at or above 2.0. A domain score is the mean of its dimension scores (Section 12.2.1). Level 2 is Competent: the behaviour "appeared unprompted and was adequate to the situation" (Section 7).

Can a strong dimension make up for a weak one in the same domain?

No. Rule 1 sets a dimension floor: no dimension below 2.0. A domain mean at or above 2.0 is not enough when one of its dimensions is below 2.0. Strength in one dimension does not cover a gap in another dimension of the same domain.

Then why is there a composite at all?

Rule 3 asks for a composite at or above 2.3. The composite is the mean of all nine dimension scores (Section 12.2.1). Section 12.3 explains that the rules are not repeats of one another. A profile at exactly 2.0 on all nine dimensions has a composite of 2.0, so it fails rule 3. The floors do different work. In the Standard's words, they stop a candidate "clearing 2.3 on the strength of three high dimensions while sitting at 1.0 on, say, E1". The composite asks for more than the minimum overall. The floors stop a high average from hiding a gap.

Can a high composite lift the band past a low dimension?

No. Each band also sets a condition on every dimension (Section 12.4). Certified covers a composite of 2.3 to 2.9, with all floors met. Certified with Merit needs a composite of 3.0 to 3.4 and no dimension below 2.5. Certified with Distinction needs a composite of 3.5 or above and no dimension below 3.0. A higher composite alone does not give a higher band.

Can plenty of evidence in one dimension make up for thin evidence in another?

No. Rule 4 asks for enough evidence on all nine dimensions. Each dimension needs at least three independent observations from distinct decision points (Section 8.3). The Standard says: "A dimension below the minimum is scored Insufficient Evidence and cannot be certified". Rule 6 then asks for confidence at Moderate or High. Confidence "describes evidential sufficiency, not the candidate's ability" (Section 10).

Can strong work on one kind of stakes make up for the kinds that are missing?

No. Rule 5 asks for the stakes coverage of Section 6.4. A certifying assessment spans at least three distinct primary consequence families. At least one of them must involve irreversible or non-financial harm. The Standard says that a candidate assessed only on financial and operational scenarios "has not demonstrated transferable judgment". Section 12.3 adds that rule 5 "stops a candidate clearing on three financially-framed scenarios".

Can one impressive answer make up for a missing basic step?

No. Levels are cumulative (Section 7.4). A level requires everything in the level below it. In the Standard's words, an impressive Level 4 behaviour "without the Level 2 base is scored at the highest level whose requirements are fully met". Every observation is scored on this scale of 0 to 4 (Section 7).

What does a candidate see if one domain falls short?

The report shows every separately reportable dimension score and the four domain scores. It also shows the band, the confidence, the stakes profiles assessed and the anchors applied. It adds the model-assisted proportion and the evidence cited for each observation (Section 12.5). The Standard says: "This is provided whether or not the candidate certifies". A candidate who does not certify can therefore see where the result fell short.

What can a candidate do after a Not Certified outcome?

A candidate may sit again no sooner than 30 days after a Not Certified outcome, and receives a different form (Section 9.4). An appeal is possible within 30 days, on a ground listed in Section 18.6 (Section 18.5). One ground is "an observation not supported by transcript evidence". Disagreement with an anchor itself is not a ground for appeal. Appeal is free for the first instance. The appeal route is at https://airapac.org/standard/appeals/. When a credential is later renewed, Section 18.2 requires a full reassessment, and a prior band gives no credit.

Is the rule the same on every track?

Yes. The Standard has three tracks for people: Campus, Professional and Practitioner. Tracks "do not differ in dimensions, anchors, thresholds or decision rules" (Section 13). The scenarios differ in complexity, stakes density and stakeholder count. The six rules apply in the same way on each track.

What this does not claim

• This page explains the decision rules of AIRS-J/1 Section 12. The text of the Standard governs.
• It does not say what a lower domain or dimension score means for a person's development, worth or role.
• It gives no preparation advice and no way to raise a score before a sitting.

AIR APAC publishes the standard for judgment in AI-enabled work, certifies people against it through its official instrument, and stands behind every result.

Register for an assessment: https://airapac.org/standard/certification/apply/