The Rigors Standard v1 and Scored Specimen

Evidence status: active standard · calibrated worked examples · scheduled validity review

For: Experienced professionals deciding whether the assessment is credible, and buyers or reviewers inspecting what a future readiness signal would mean.

What version 1.0 does—and does not—establish

Version 1.0 is the scoring source for role assessments. It freezes a common 0–4 evidence anchor, weighted role criteria totaling 100 points, separate critical-condition handling, instrument bands, calibration rules, and human disposition boundaries.

The standard does not claim perfect prediction, eliminated bias, or equivalence across unlike roles. The validity register tracks reviewer agreement, subgroup and burden questions, dispositions, appeals, and downstream engagement evidence; reviewer identity, qualification, assignment, conflicts, recusal, minimum-necessary views, append-only decisions, notices, correction, and appeal routing are enforced controls.

The Rigors is modeled on the structure of ISO/IEC 17024, the international standard for bodies that certify persons. IT Modality is not accredited to it and does not claim to be.

One decision must remain reconstructable across six stages.

1. Application review. Confirm role scope, recent responsibility, and safely reviewable evidence.

2. Evidence screen. Connect claims to inspectable decisions and work without exposing confidential material.

3. Role assessment. Observe execution on a realistic, non-client task with explicit constraints and acceptance criteria.

4. Delivery conversation. Examine communication, ambiguity handling, challenge response, escalation, and professional judgment.

5. Readiness review. Reconcile evidence, critical conditions, calibration, limitations, and process integrity.

6. Decision and next steps. Communicate only the status and next action supported by the record.

Every stage records its purpose, evidence request, version, assigned reviewer, permitted disposition, rationale, notice, and audit event. A total score alone is never a readiness decision.

Review observable work on one shared 0–4 scale.

For a weighted criterion: anchor ÷ 4 × criterion weight. Each role instrument totals 100 points. Reviewers cite the exact evidence and rationale for every anchor.

A band describes one instrument, not the person.

• 85–100 / strong evidence: consistently effective or advanced evidence; material limitations still travel to readiness review.

• 70–84 / sufficient evidence: broadly effective evidence with bounded development areas.

• 55–69 / mixed evidence: material gaps require reviewer judgment under the stage policy; there is no automatic advance.

• Below 55 / insufficient evidence: the current instrument does not support the stated role level.

No final pass rate, applicant percentile, automatic rank, permanent credential, employment decision, placement decision, or client recommendation comes from these bands.

A serious process or integrity failure never disappears inside a total.

Critical conditions are recorded separately when there is confidential or unlawfully shared information; fabricated identity, experience, evidence, result, or source; undeclared material assistance; authority claimed outside the assigned role; assessment work reused for client production; an unresolved scorer conflict; an inaccessible or materially defective process; or Academy, payment, country, nationality, accent, or accommodation information leaking into a merit view.

The response may be an integrity review, evidence correction, process correction, reassignment, or closure under the applicable stage policy. A critical condition is never silently converted into zero points and averaged away.

CALIBRATED WORKED RECORD

The calibration candidate receives a calibrated application-rollout packet containing an outcome request, stakeholder roles, milestone draft, three dependencies, a test summary, an unresolved downtime decision, and two conflicting status notes. The task is to produce a bounded outcome and scope brief, corrected plan, decision-focused status, evidence gates, one escalation, and a release recommendation.

Critical-condition review: none observed in the calibrated record. This statement is part of the example, not evidence about any real applicant.

Calibration note: A second fictional reviewer scores two criteria one level lower. The reviewers retain both originals, reconcile the evidence citations, and settle at the anchors shown above. No target pass rate is used.

Readiness-review result: decision held — complete remaining instruments and process controls. The 81-point role instrument does not authorize network admission, client review, an engagement, work, employment, placement, income, or an immigration outcome.

Evidence must be safe enough to inspect and specific enough to challenge.

Allowed evidence may include redacted or recreated work products, a declared calibrated artifact, a role task created for the assessment, an explanation of decisions and tradeoffs, and references that the applicant is authorized to provide. Patient information, privileged material, client secrets, private source code, credentials, production access, or work the applicant cannot lawfully share are prohibited.

Country, nationality, accent, native-speaker identity, Academy purchase, scholarship, enrollment, performance, and completion are excluded quality factors. An accommodation owner may communicate only the adjustment needed to run the assessment; unnecessary diagnosis details never enter the scorer projection.

Scores organize evidence. Assigned people decide.

The readiness owner considers stage evidence, instrument bands, critical conditions, role and level, calibration, process integrity, accommodations, and reviewer rationale. No software rule, AI output, sales owner, Academy record, or commercial target makes the consequential decision.

Borderline, conflicted, accommodated, integrity-reviewed, and configured sample decisions receive a second review. Differences greater than one anchor level require written reconciliation while original scores remain preserved.

Freeze the version before live use; change it visibly afterward.

Version 1.0 must be frozen before a first live application. A later change records the proposal, rationale, affected instruments, review, decision owner, effective date, migration rule, notice impact, and public changelog entry. A changed rubric never silently re-scores an earlier decision.

Review governance and change control

No predictive claim exists yet.

The validity register tracks inter-rater agreement, subgroup fairness, completion burden, adverse impact, disposition rates, appeals, client acceptance, and engagement performance. Every study predefines the question, measure, denominator, protected process, owner, review date, and publication rule.

Inspect the validity and fairness plan

Inspect the process around the score.

The score is only useful if reviewer competence, conflicts, projections, audit, correction, appeal, and status governance make the record trustworthy.

Review assessor requirements Review impartiality and recusal Return to The Rigors