Explore historical practice percentages across exam parts, domains and subjects. Every report explains its method, evidence limits and a practical next step.
Community accuracy weights every eligible learner equally. The top-performer line shows the highest-accuracy quartile within each part, not the learners with the most attempts.
The top-quartile cutoff is the minimum learner accuracy within that part's reproducible top cohort. Scroll horizontally to review every column.
| Part | Community accuracy | Top accuracy | Top cutoff |
|---|---|---|---|
| Part 1 | 63.4% | 80.1% | 71.8% |
| Part 2 | 63.7% | 81.9% | 74.8% |
| Part 3 | 70.1% | 86.9% | 78.7% |
Data through August 13, 2026. The snapshot contains aggregate results only and excludes names, email addresses, notes, question text, session identifiers, and learner-level rows.
Public methods package v1.1.0: data dictionary and reproducibility manifest. The public package documents percentage outputs and qualitative methods without publishing retained cohort, activity-threshold, or blank-rate details. The internal snapshot v1.0.0 remains partially, not independently, reproducible.
Original evidence about difficult subjects, exam pressure, review flags, and common misconceptions.
Compare subject accuracy with the top quartile to find rules that deserve focused review. Results describe the historical practice cohort and are not a forecast of official exam difficulty.
Read the reportStudy priority combines the exam domain's blueprint weight with its observed miss rate. The percentage is that domain's share of weighted difficulty within its part, not a probability of failing.
Read the reportThe comparison follows the same eligible learners in question practice and completed practice exams. Answered-item accuracy describes the response quality in each mode; the observational gap does not establish a cause.
Read the reportCompare accuracy on flagged and unflagged answered items. A flag can help organize review, but these percentages do not establish that changing a flagged answer improves the result.
Read the reportDominant wrong-answer share measures how often one distractor attracts incorrect responses within qualifying subjects. The report exposes a pattern without publishing proprietary question or answer wording.
Read the reportObserved practice behavior, repeat work, simulations, pacing, study time, and notes.
The first and latest recorded results describe repeat-question recovery. The platform deliberately resurfaces difficult items, so improvement in this selected group cannot establish the effect of the course.
Read the reportFirst and second response accuracy is grouped by the delay between them. Starting performance differs between delay groups, so the percentages cannot identify one optimal review interval.
Read the reportFirst and latest scores compare completed practice exams of the same length and part. Cohorts vary in evidence strength; limited groups can change substantially and must not support broad improvement promises.
Read the reportHalf and full practice exams serve different study needs. Answered accuracy measures responses submitted; overall score reflects the entire assigned exam. The comparison describes practice behavior and does not establish that one length causes better preparation.
Read the reportBuild a pacing routine with timed mixed practice, a planned review pass and a short post-exam review. Use your own practice experience to adjust the routine and check current PSI instructions before exam day.
Read the reportActivity shares describe retained browser-event duration after invalid and unusually long events are excluded. Open tabs, inactivity and overlap may remain, so the result does not measure verified active study time.
Read the reportThe percentages compare accuracy with and without recorded note activity. The cohorts are not matched and differ in study behavior, so the comparison cannot show that notes caused a change in performance.
Read the reportCoverage, item analysis, editorial review, and transparent performance benchmarks.
The top quartile is defined by each learner's overall accuracy within the same part. Domain-level differences describe performance separation, not a causal advantage or an official pass rate.
Read the reportThe distribution shows the share of well-sampled questions in each observed accuracy band. Very hard and very easy questions both need editorial context; difficulty alone does not establish question quality.
Read the reportThese percentages describe the distribution of the question bank in the August research snapshot. They are historical coverage shares, not today's product inventory or proof of complete exam readiness.
Read the reportEditorial confidence describes authoring certainty while observed accuracy describes practice difficulty. Their percentages should be reviewed separately; one does not validate the other.
Read the reportAnnual candidate trends and careful validation of Passboard readiness signals.
The historical percentages compare practice accuracy across years and exam parts. Content changes, cohort composition and question exposure can move the results. Year-to-date observations have limited evidence and cannot establish a trend.
Read the reportPassboard bands are compared with current practice accuracy and historical practice-exam scores. The comparison is not time-ordered and contains no verified official SEE outcomes; it cannot establish a first-attempt pass rate.
Read the report