A single-case analysis worksheet audit traces every phase value to source evidence, verifies phase labels and improvement direction, recalculates formulas, and checks missing-state, tie, trend, and time rules. It also confirms that the selected metric fits its stated purpose and remains supplemental to visual, clinical, contextual, and design review. Preserve corrections and unresolved items.

Reconcile phase records

Match every plotted value to its date, phase, source, unit, and validity state. Flag duplicates, impossible order, and silent imputation.

Verify analysis choices

Check improvement direction, metric version, reference rule, tie handling, time scale, trend method, and exclusions.

Recalculate independently

Rebuild medians, extremes, pair counts, projections, and denominators from locked raw values. Compare results with the worksheet.

Release with scope

Document corrections, unresolved items, visual-analysis findings, context, client input, and the qualified review decision. A clean formula does not validate the underlying design.

Assemble an audit packet before recalculating

Preserve the source export, graph, data dictionary, phase-change record, calculation workbook, software version, and prior result as separate artifacts. The audit copy should be read-only. Work from a controlled duplicate so correcting a formula cannot erase the original evidence or its timestamps.

Create one row for every expected observation. Include the source identifier, timestamp, phase, measure and unit, context, validity state, exclusion reason, and correction owner. This row-level inventory exposes duplicate imports, silent exclusions, incompatible units, and points assigned to the wrong phase before they reach a formula.

Classify duplicates without deleting evidence

Two identical values are not automatically duplicates. Confirm whether they represent the same source event, a legitimate repeated observation, or two records with an unresolved relationship. A true duplicate can be held from the calculation while remaining in the audit trail with the basis, reviewer, and disposition.

When identity is unresolved, keep the record in a held state. Do not select the value that produces the more favorable result. Regenerate every dependent graph and calculation after the source status is resolved, and preserve the earlier output as superseded.

Recalculate from the locked valid cohort

Build each metric from the valid row inventory instead of copying a prior numerator. Show the baseline median, baseline extreme, NAP pair ledger, tie rule, and valid denominators. Calculate with full precision and round only the displayed result.

For Ravi's valid values, PEM is 3 of 4, or 75%; PND is 1 of 4, or 25%; and NAP is 8.5 of 12, or 70.8% after half credit for one tie. The NAP ledger contains eight favorable, one tied, and three unfavorable comparisons. These counts provide an independent reconciliation point for the worksheet.

Apply single-case analysis worksheet audit to one phase comparison

The supplemental-analysis audit sheet for Ravi identifies the phase labels, improvement direction, raw observations, missing-state rule, reference value or line, formula, tie rule, calculation owner, and review date before a supplemental result is interpreted.

Work the single-case example for Ravi

Ravi's export lists baseline 2, 4, and 6 and intervention 3, 5, 6, 8, and a duplicate 8 at the same timestamp. The duplicate is held, leaving four valid intervention values. Recalculation gives PEM 75%, PND 25%, and NAP 70.8%. The original five-point denominators are rejected.

The audit reports data coverage separately: four valid intervention observations out of five exported rows, or 80%, with one duplicate held. Metric denominators use the four valid observations or 12 valid cross-phase pairs. Coverage and metric results answer different questions and remain separate.

Audit reproducibility for Ravi

Ravi's worklist accounts for the held duplicate, source timestamp, correction owner, valid phase counts, formulas, tie rules, and regenerated graph. The original export stays preserved.

Maintain a discrepancy and release log

For every difference between the prior worksheet and the recalculation, record the affected cell or output, prior value, corrected value, cause, downstream artifacts, owner, reviewer, and release state. A corrected percentage should not be released while an old graph, report, or clinical summary still carries the superseded denominator.

The qualified clinician reviews what the correction changes in interpretation and whether any clinical decision needs reconsideration. Data and calculation staff can trace and repair evidence within their roles. They do not acquire authority to select treatment, determine consent, or make a clinical recommendation by fixing a workbook.

Protect client relevance for Ravi

During Ravi's reproducible single-case calculations review, keep AAC, interpreters, mobility, food, water, bathroom use, prescribed care, health support, rest, relationships, and emergency help available. Use accessible communication to ask whether the measured outcome and magnitude matter to the person. A favorable statistic cannot repair an unwanted goal or inaccessible plan.

Use current single-case evidence for Ravi

For Ravi's calculation, the BCBA Test Content Outline supplies examination scope for graphing, interpretation, experimental design, and data-based evaluation. The BACB ethics hub and CASP summary provide professional context. The WWC Version 5.0 handbook is an education-research evidence standard, not a clinical authorization rule. A single-case methods review documents PND, NAP, and extended-line formulas. Research on structured visual-analysis aids and decision accuracy across overlap measures supports using calculations as transparent supplements while retaining metric and design limits. ASHA supports continuous AAC access.

Close Ravi's analysis review

Review the supplemental-analysis audit sheet with Ravi and the responsible clinician. Preserve raw data, graph, phase definitions, calculation version, coverage, ties, limitations, client input, decision, owner, and next review date. Reopen the analysis when data, context, access, health, goal, phase, or method changes.

Clinician audit checklist

  • Does every expected observation have a source identifier, phase, unit, and validity state?
  • Are duplicates distinguished from legitimate repeated values and preserved in the audit trail?
  • Do valid phase counts reconcile to the coverage report and every metric denominator?
  • Can another reviewer reproduce PEM, PND, and the 8-favorable, 1-tied, 3-unfavorable NAP ledger?
  • Are formulas, software versions, rounding rules, and tie rules stored with the output?
  • Have graphs, reports, and downstream decisions been checked for superseded values?
  • Are unresolved discrepancies held with an owner, next step, and review date?

Related resources

Sources