LMP-C01 cheat sheet

Certification exams measure minimum competence against a defined blueprint; the blueprint weighting determines the number of scored items per domain.

A passing standard is set on the scored (operational) items only; unscored pretest items are embedded but never count toward the candidate's score.

Item counts, appointment length, and seat time are three distinct numbers: seat time includes check-in, tutorial, and survey; appointment time does not.

Score reports typically show pass/fail plus domain-level performance bands; scaled scores are not raw percentages and are not comparable across exam forms.

Eligibility windows and authorization-to-test periods are separate clocks: eligibility governs application validity, the ATT governs scheduling.

Rescheduling and cancellation deadlines are measured in hours before the appointment start time, not the check-in time.

Accommodations requests are handled before scheduling; approval must be on file or the seat cannot be delivered with the accommodation.

A failed attempt triggers a retake waiting period and a lifetime or rolling attempt cap defined in the candidate handbook, not by the test center.

Multiple-choice items have one keyed option and three or more distractors; each distractor should map to a plausible error or misconception.

Multiple-response items score dichotomously unless stated otherwise: all required options must be selected, with no partial credit by default.

Matching and drag-and-drop items are scored per pairing; partial-credit rules must be documented in the item specifications before use.

Case sets share one stimulus across several linked items; linked items must be independently answerable so one error does not cascade.

Ordered-response items (sequencing) require a fully correct sequence for credit unless a partial-credit rule is explicitly specified.

Constructed-response and essay items require a rubric with defined score points and exemplar responses before they enter review.

Hot-spot and exhibit-based items must be fully self-contained in the delivered form; any dependency on an external figure is a defect.

Item type selection is driven by the target cognitive level: recall suits multiple-choice, while procedure and judgment favor case sets or constructed response.

Stem should pose a clear problem or question; the correct answer plus each distractor must complete the stem grammatically.

Avoid negative stems; if a negative is unavoidable, capitalize or bold the negative word (NOT, EXCEPT) to reduce misreading.

Keep options parallel in length, grammar, and structure; the longest option should not reliably be the key.

Avoid 'all of the above' and 'none of the above'; they enable test-wise guessing and distort difficulty estimates.

Each item should test one point; a stem requiring two independent judgments conflates content and lowers diagnostic value.

Use plausible, homogeneous distractors drawn from common errors; implausible options reduce discrimination and inflate scores.

Avoid cueing, clang associations, and grammatical giveaways between stem and key; these allow correct answers without content knowledge.

Two-reviewer minimum: a content review for accuracy and a fairness/sensitivity review for bias and accessibility before field testing.

Classical item difficulty p = proportion answering correctly; item discrimination is the point-biserial correlation between item score and total score.

KR-20 and Cronbach's alpha estimate internal consistency reliability; values below roughly 0.70 warrant review for high-stakes exams.

Standard error of measurement SEM = SD × sqrt(1 − reliability); a 95% confidence band is approximately ±1.96 × SEM.

The Angoff method asks subject-matter experts to estimate the probability a minimally competent candidate answers each item correctly; the mean is the cut score.

The modified Angoff adds a second round with item-level performance data to converge expert judgments.

The bookmark method orders items by difficulty and asks panelists to place a bookmark at the point where a borderline candidate would succeed.

IRT models (1PL, 2PL, 3PL) separate item difficulty, discrimination, and guessing from person ability; item fit statistics flag misfitting items.

Equating places scores from different forms on a common scale so the passing standard is held constant across administrations.

Item exposure controls (random selection, exposure caps, multiple forms) limit how often any item is seen and reduce memorization risk.

A confidentiality agreement binds candidates and reviewers; sharing item content is a test-security violation independent of intent.

Fairness review screens for construct-irrelevant content: stereotypes, regional idioms, and cultural references unrelated to the target skill.

Differential item functioning (DIF) flags items where matched groups of equal ability perform differently; DIF is a flag for review, not automatic removal.

Accessibility review checks that item format and language do not create barriers unrelated to the construct being measured.

Incident reporting: irregularities (device failure, misconduct, disruption) are documented at the time and adjudicated under a defined policy.

Scratch paper, breaks, and personal-item rules are security controls; violations are handled by the same integrity process as content sharing.

Item banking requires version control and audit trails so retired or compromised items cannot re-enter live forms.

The candidate handbook is the controlling document for eligibility, scheduling, scoring, retakes, and appeals; it must be version-controlled.

Policies should state what changed, effective date, and which candidates are grandfathered under prior rules.

Score-report communication includes pass/fail, domain feedback, retake rules, and the appeal window in one place.

Appeals processes must specify grounds, filing deadline, evidence required, and the decision-maker to be defensible.

Recertification cycles, continuing-education requirements, and lapse/reinstatement rules belong in the published program policy.

Communication timelines (result release, retake waiting period, credential issuance) should be published as commitments, not estimates.

Candidate-facing language should avoid promises of employment or outcomes; certification attests competence, not job placement.

Program changes affecting the blueprint require advance notice and a transition plan so candidates are not tested on retired content.