assessment

Assessment IS a specific kind of practice — a structured practice of information-gathering and characterization. A competent speaker would call assessment 'a practice' (assessment practice, assessment as a professional practice). Nearest kind: practice.

Evaluation IS a specific kind of assessment — it assesses with formal criteria and structured methods, whereas assessment is the broader practice of judging value or quality by any means. Competent speaker test: 'evaluation is a kind of assessment.' Direction: specific→general per Law 9.

A likert scale IS a specific kind of assessment — a structured instrument for measuring attitudes via agreement responses. Direction specific→general per Law 9.

Formative assessment is a specific kind of assessment — namely, an evaluation conducted during the learning process to inform instruction, as opposed to summative assessment at the end. A competent speaker would say 'formative assessment is a type of assessment.' Files against nearest kind (assessment). Epoch test confirms assessment is older (0.04 vs 0.23), consistent with formative assessment being a specialized development within the broader category. Law 9 and Law 11e satisfied.

A cloze procedure IS a specific kind of assessment: a method where words are removed from text and learner responses measure retention and comprehension. A competent speaker would classify it as an assessment technique. Direction tested: INSTANCE_OF (specific→general, Law 9). Files against nearest kind.

A rubric is designed and maintained for the sake of assessment. It structures how evaluative criteria are applied consistently. The servant→master direction is correct: rubric serves assessment.

TESTED: examination IS A SPECIFIC KIND of assessment — a formal evaluation through questioning or testing. A competent speaker would call an examination 'an assessment'. Direction: examination → assessment.

safety-significance is a specific kind of assessment — it assesses how significant or consequential a safety risk or event is. A competent speaker would call safety-significance 'an assessment.' It files against the nearest kind.

A safety-climate survey is a specific kind of assessment — it measures attitudes, beliefs, and perceptions about safety. A competent speaker calls it an assessment of organizational safety climate.

risk assessment is a specific kind of assessment — a systematic evaluation of potential risks and their consequences. A competent speaker would say 'risk assessment is a type of assessment.' The nearest kind is assessment, which captures the evaluative method common to both.

A credibility-assessment is a specific kind of assessment — it evaluates the trustworthiness or believability of a source, claim, or person. Competent-speaker test: 'a credibility assessment is a kind of assessment' is unambiguously true.

Model evaluation IS a specific kind of assessment: the systematic evaluation of something against criteria to determine quality or performance. A competent speaker calls model evaluation 'an assessment method.' Files against the nearest kind, avoiding the discipline-leap error from my past.

Red-teaming is an adversarial evaluation practice that depends on assessment methodology — the systematic evaluation of security posture. Remove assessment as a practice and red-teaming loses its evaluation framework. Direction correct (assessment older at 0.03 vs red-team at 0.09).

Correcting the misfiled DEPENDS_ON (struck): red-teaming is not a thing that depends on assessment — it IS a specific kind of assessment. Under assessment's accepted definition (the structured practice of gathering information to characterize a system, with parameters: target, criteria, methods of information gathering — explicitly including 'testing' — and a resulting characterization), red-teaming instantiates it with method = simulated adversarial attacks and output = a vulnerability/security-posture characterization. It is a subset/kind, so identity is not dependency (Law 8c); the correct edge is INSTANCE_OF. Nearest kind: no security-assessment or adversarial-testing entry exists, so 'assessment' is the nearest available kind (Law 11e). Pinned sense: adversarial evaluation as a species of information-gathering assessment.