To audit descriptive ABC data, reconcile planned, observed, valid, response, mapped-sequence, missing, truncated, and excluded units before recalculating probabilities. Verify event definitions, timestamps, sequence windows, code versions, observer coverage, and source records. Release the table with raw cells and descriptive limits. Route clinical interpretation to the qualified professional responsible for the assessment.

Reconcile the observation universe

Start with planned units and account for observed, missed, invalid, and outside-scope states. Keep the valid analysis denominator and coverage denominator separate.

Trace every response sequence

Link each response to its timestamp, eligible antecedent or consequence window, mapped events, and truncation state. Flag duplicate event IDs and impossible order.

Recalculate from raw cells

Compare worksheet formulas with source counts. Review direction, interval length, background denominator, rounding, empty cells, and overlapping categories.

Use a release gate

Require an owner, completed corrections, unresolved-item list, source version, and qualified review. Preserve the original evidence and correction history.

Assemble an audit packet

Include the observation plan, current definitions, codebook, raw timestamps or interval records, planned and actual coverage, exclusion log, sequence-mapping table, calculations, software or spreadsheet version, and prior corrections. The packet should let a qualified reviewer reproduce every reported numerator and denominator without relying on an undocumented analyst judgment.

Use stable record identifiers instead of copying sensitive narratives into a broad worklist. Link access or health context only when it is needed to interpret the sample, and restrict those fields to appropriate roles. Preserve who entered, corrected, calculated, and approved each element.

Classify discrepancies by their effect

Separate source errors, definition mismatches, timestamp problems, missing coverage, duplicate mappings, wrong sequence direction, formula errors, rounding differences, and interpretation concerns. Some problems require a recalculation. Others require new observation because the underlying sample cannot answer the question.

Give every discrepancy a state such as open, corrected, accepted with limitation, recollection required, or withdrawn. Record the owner, decision authority, due date, evidence, and affected outputs. Keep the original result and correction history available.

Release a result only when its limits are visible

The final table should show planned, observed, valid, response, mapped-sequence, missing, invalid, truncated, and excluded counts that reconcile to the locked cohort. Include definitions, sequence window, background denominator, calculation, coverage, and unresolved limitations.

Clinical release means the qualified professional has reviewed the evidence within scope. It does not turn a descriptive association into a demonstrated behavioral function. If critical mappings remain unresolved, hold the affected result and state what can still be reported safely.

Frame one clear audit question

The descriptive-data audit worklist for Wren names the observable events, sequence direction, time or interval window, planned exposure, valid denominator, missing-state rule, raw cells, calculation owner, and review date before percentages are interpreted.

Worked example: reconcile 80 planned intervals

Wren has 80 planned intervals, 72 observed, and 70 valid. Observation coverage is 72 of 80, or 90%. Valid coverage is 70 of 80, or 87.5%. The response occurs in 14 valid intervals; 12 have a complete mapped consequence window, so sequence mapping completeness is 12 of 14, or 85.7%. Two remain open, not negative.

Reconcile every denominator

Wren's worklist accounts for all 80 planned intervals and all 14 response intervals. It records the reason, owner, and disposition for eight missed observations, two invalid observations, and two incomplete response sequences without double-counting them.

Review representativeness after arithmetic

A clean calculation can still rest on an unrepresentative sample or ambiguous definitions. Review where missingness occurred, which contexts were observed, whether access was preserved, and whether the descriptive method can answer the clinical question.

Protect access during observation

During Wren's release-ready ABC evidence review, keep AAC, interpreters, mobility, food, water, bathroom use, prescribed care, health support, rest, relationships, and emergency help available. Record access or health changes and use accessible communication to ask about relevant events, comfort, assent, dissent, and priorities.

Use evidence within its stated scope

The BACB ethics hub and CASP public summary provide professional context. The BCBA Test Content Outline includes direct, indirect, and product measures; validity and reliability; representative measurement; and indirect, descriptive, and experimental functional assessment as examination content.

For Wren's release-ready ABC evidence question, primary descriptive-assessment research compares conditional with background probabilities and shows why event direction, sequence windows, cell counts, and context matter. Antecedent versus consequent predictors, contingency space analysis, and comparisons of descriptive methods document limited correspondence between descriptive and experimental findings. Research on reducing functional-assessment ambiguity distinguishes observed association from experimental demonstration. ASHA says AAC users should always have access to their tools or devices.

Close the descriptive-data review

Review the descriptive-data audit worklist with Wren and the responsible clinician. Record raw observations, definitions, windows, coverage, calculations, limitations, the selected next assessment step, its owner, and a review date. Reopen the method when the response, context, access, health, observer, software, or decision changes.

Clinician release checklist

  • Does the audit packet reproduce every numerator, denominator, and coverage value?
  • Are missing, invalid, truncated, excluded, and unresolved records visible?
  • Have duplicate mappings, direction, windows, and code versions been tested?
  • Is each correction attributable to an owner and preserved with the original result?
  • Does the released table identify limitations that affect interpretation?
  • Has a qualified clinician reviewed the result within scope?
  • Can the client access and comment on the outcome without losing communication or support?

Related resources

Sources