ACE-Expert key terms

A managed, searchable collection of test questions stored with metadata such as content area, difficulty, and exposure history, used to assemble forms and support reuse across administrations.

Structured descriptive data attached to each question, including content classification, cognitive level, answer key, statistics, and usage history, enabling search, auditing, and informed form construction.

A specification that defines the proportion of test content allocated to each domain, task, or cognitive level, ensuring forms represent the intended construct consistently across administrations.

A document describing the overall design of an exam, including purpose, target population, length, item formats, blueprint weights, scoring rules, and administration conditions.

A systematic study of a profession that identifies the tasks performed and knowledge required, providing the empirical foundation for defining exam content and validating the blueprint.

A qualified practitioner or educator who contributes professional judgment to activities such as item writing, content review, standard setting, and defensibility of exam content decisions.

A structured evaluation of drafted questions by qualified reviewers to check accuracy, clarity, fairness, alignment to the blueprint, and absence of construct-irrelevant difficulty before field testing.

The process of composing questions that measure a defined objective with clear, concise stems, plausible distractors, and a single defensible correct answer, following established editorial guidelines.

An incorrect response option designed to attract examinees who lack the targeted knowledge while appearing plausible to those who possess it, thereby supporting score discrimination.

A statistic describing how hard a question proved for a group, commonly the proportion of examinees answering correctly, with values near one indicating an easy item.

An index of how well a question separates higher-scoring examinees from lower-scoring ones, often computed as a point-biserial correlation or an item response theory slope parameter.

A correlation between a dichotomous item score and a continuous total test score, used in classical analysis to indicate how strongly an item aligns with overall performance.

A measurement framework modeling an observed score as the sum of a true score and random error, yielding familiar statistics such as difficulty, discrimination, and reliability coefficients.

A family of probabilistic models relating the chance of a correct response to latent examinee ability and item parameters, enabling item-level information and equating across forms.

A curve describing how precisely a test measures ability across the latent trait continuum, derived from item parameters and used to evaluate measurement precision at cut scores.

The consistency of scores across items, occasions, or raters, often expressed as an internal consistency coefficient or through generalizability analysis of measurement error.

An estimate of the typical magnitude of error in an individual score, used to build confidence intervals around observed scores and to interpret score differences cautiously.

Evidence that test content adequately represents the knowledge and skills defined by the blueprint, established through expert review and alignment studies rather than statistical correlation.

The degree to which scores and their proposed interpretations reflect the intended underlying ability rather than unrelated factors, supported by accumulating evidence across multiple sources.

A statistical condition in which examinees of equal ability but different group membership have differing probabilities of answering an item correctly, signaling potential bias requiring review.

A structured process combining statistical DIF results with expert judgment to detect and remediate content that unfairly disadvantages members of a particular subgroup.

The principle that an exam and its uses are equitable across examinee groups, free of construct-irrelevant barriers, and supported by accessibility, bias, and validity evidence.

A modification to administration conditions or response format, such as extended time, intended to remove disability-related barriers without altering the construct being measured.

Policies and practices protecting exam content and scores from compromise, including controlled access, secure delivery, monitoring for leakage, and investigation of suspected irregularities.

The frequency with which a particular question has been administered, tracked to limit overexposure, reduce memorization risk, and preserve the integrity of the item pool.

The process of selecting items from the bank to construct a test form that satisfies blueprint constraints, statistical targets, and exposure limits while maintaining parallel difficulty across forms.

A statistical procedure that places scores from different test forms onto a common scale so they can be interpreted interchangeably despite differences in form difficulty.

The score threshold on a test scale that separates examinees classified as meeting a performance standard from those who do not, established through a defensible standard-setting process.

A structured judgmental procedure in which qualified panelists recommend a passing score by evaluating content difficulty and expected performance of minimally competent candidates.

A standard-setting technique in which panelists estimate, for each item, the probability that a barely qualified examinee answers correctly, with estimates summed and refined through discussion rounds.

An item-response-theory-based standard-setting approach in which panelists place a bookmark on an ordered item booklet to indicate the point where borderline candidates would likely succeed.

The hypothetical examinee whose knowledge and skill just meet the performance standard, serving as the reference point for panelists' judgments during standard-setting exercises.

A group of qualified subject matter experts convened to make passing-score recommendations, selected for relevant expertise and balanced representation across practice settings and demographics.

Embedding unscored new items in operational administrations to collect response data for evaluating difficulty, discrimination, and functioning before the items enter the scored pool.

The statistical examination of examinee responses to each question, producing difficulty, discrimination, distractor performance, and flagging of malfunctioning items for review.

Systematic assessment of whether a certification or testing program achieves its intended purposes, using metrics such as pass rates, candidate satisfaction, and stakeholder impact.

An ongoing cycle of measuring program performance, identifying weaknesses, implementing corrective actions, and verifying results to strengthen exam validity and operational efficiency.

A document specifying standardized procedures for proctors and staff, including candidate check-in, timing, security protocols, and reporting of irregularities to ensure consistent conditions.