An ABA maintenance probe data sheet needs the elapsed time, the person's actual exposure between checks, and the initial response before programmed help. A date label alone is insufficient. A missed check is not a failed skill, while a correct later response cannot demonstrate retention from an earlier phase if intervening teaching is omitted.

Clinicians & ABA Professionals / Assessment and Treatment Planning.

This worksheet helps a team schedule and review observations. It does not establish mastery, choose a maintenance interval, authorize a procedure, or define a relapse response. Those decisions require qualified clinical judgment, client participation, current evidence, contextual fit, risk review, and applicable requirements.

Maintenance needs a fixed reference point

The current BCBA Test Content Outline, Sixth Edition treats response maintenance, representative measurement, maintenance procedures, generalization, contextual fit, procedural integrity, relapse mitigation, and data-based decisions as related but separate competencies. The BACB test-content outline page identifies the certification outlines currently in use. An outline is not a client plan and does not validate this form.

Name the comparison point for every later observation. The last teaching phase, a transition decision, or another qualified review point can serve as the reference only when its version, date, target definition, and context are preserved. If teaching restarts, the next observation may remain useful, but it no longer answers an unchanged maintenance question.

Maintenance contract

Complete one contract for each target and procedure version. Start a new version when the target, probe arrangement, ordinary supports, initial-response rule, consequence, or reference phase changes materially.

Maintenance contractEntryClient or authorized case codeClient-selected goal and practical reasonTarget response and measurement definitionReference phase, version, and end dateEvidence at the reference pointScheduled maintenance intervals and rationaleEligible opportunity and start/end boundaryInitial-response windowCorrect, incorrect/no-response, invalid, unavailable, and missed codesPrompt, feedback, correction, and consequence rule during a probeHow interim teaching, practice, and natural exposure will be loggedOrdinary AAC, communication, mobility, sensory, and environmental supportsClient choice, assent, dissent, break, and stop processSafety and escalation conditionsObserver, integrity, and agreement planAuthorized record location and correction ruleResponsible clinician and next review trigger

Keep the scheduled label and actual elapsed time side by side. A “Week 6” check might occur 39 or 51 days after the reference point. That difference belongs in the evidence instead of being normalized away.

Scheduling and exposure ledger

The exposure history makes the later observation interpretable. It need not be perfect to be useful, but known teaching and practice should not be hidden.

Planned intervalPlanned date/windowActual probe dateActual days since referenceAvailable?Reason missed or delayedTeaching since referenceNatural practice or exposureProcedure/access changesReviewer noteCheck 1Yes / noCheck 2Yes / noCheck 3Yes / noCheck 4Yes / no

Natural exposure between checks is often expected for a meaningful skill. Recording it does not invalidate the review; it clarifies the question. Daily use under typical conditions and no contact for three months represent different durability histories.

Trial-level maintenance sheet

Record the initial response first. If teaching or correction follows, preserve it as a later event.

DatePlanned intervalOpportunityActual condition and materialsInitial responseCorrect?Prompt before initial outcome?Available?Valid probe?Teaching/correction afterwardClient communication/contextObserver1Yes / noYes / noYes / noYes / no2Yes / noYes / noYes / noYes / no3Yes / noYes / noYes / noYes / no4Yes / noYes / noYes / noYes / no5Yes / noYes / noYes / noYes / no6Yes / noYes / noYes / noYes / no

Use separate codes for missed, unavailable, and invalid observations. A scheduled event that never occurred supplies no performance outcome. An event can occur but be invalid for the intended question if the cue, support, timing, or consequence did not match the contract.

Denominators and timepoint summaries

Review quantityCalculationWhat it describesOpportunity availabilityAvailable or presented opportunities / planned opportunities x 100How much of the planned schedule produced an opportunityValid probe coverageValid probes / planned opportunities x 100How much of the schedule supports the intended maintenance questionCorrect responding in valid probesIndependently correct initial responses / valid probes x 100Recorded performance within the valid subsetTimepoint-specific correct respondingCorrect initial responses at the timepoint / valid probes at the timepoint x 100Performance at that elapsed-time checkIntegrity contaminationInvalid probes / presented opportunities x 100A procedure-quality signal, not a client-performance score

Do not use the planned count as the performance denominator when probes were not presented or valid. Conversely, do not erase the planned count. The ABA maintenance probe data sheet preserves both because coverage and performance answer different questions.

TimepointActual days since referencePlannedAvailable/presentedValidIndependently correctIncorrect/no responseInvalidCorrect percentageInterim exposure note

Retain each timepoint even when the team also calculates a pooled percentage. Unequal elapsed times are not interchangeable replications, and the direction of a short sequence cannot identify its cause.

Fictional completed example

The following record is invented for arithmetic review. Case M is not a real client, and the routine, reference phase, intervals, consequences, and interpretation are not recommendations.

Case M chose a fictional transit-card routine as a practical goal. Procedure version 3 defined the initial response and ordinary communication access. Six probe opportunities were planned at each of three checks: 2, 6, and 12 weeks after the reference phase, for 18 planned opportunities.

TimepointPlannedAvailable/presentedValidIndependently correctIncorrect/no responseWhat happened to the remainder2 weeks66651All planned opportunities occurred6 weeks66541One early prompt invalidated the initial-response observation12 weeks64422Two planned transit activities did not occurTotal181615114Two unavailable and one invalid record remain visible

Opportunity availability is 16 divided by 18, or 88.9%. Valid probe coverage is 15 divided by 18, or 83.3%. Correct responding within valid probes is 11 divided by 15, or 73.3%.

The timepoint values remain separate: 5 divided by 6, or 83.3%, at 2 weeks; 4 divided by 5, or 80.0%, at 6 weeks; and 2 divided by 4, or 50.0%, at 12 weeks. The count check is 5 + 4 + 2 = 11 correct initial responses and 1 + 1 + 2 = 4 incorrect or no-response outcomes across 15 valid probes.

The 73.3% aggregate does not establish durable maintenance. It does not make the three elapsed times equivalent, treat the two missed opportunities as failures, or repair the invalid 6-week observation. A reviewer would also need the exposure ledger. If Case M practiced the routine between checks, that history must accompany the data rather than being interpreted as “no intervention.”

Nothing in the fictional sequence proves mastery, relapse, causation, generalization, treatment effect, social validity, or client benefit. The apparent decrease may be a review signal, but it is not an automatic decision rule.

Research context for probe records

A descriptive analysis of applied skill-acquisition research found substantial variation in whether and how maintenance was reported. That variation cautions against assuming one universal interval, criterion, or meaning for the word “maintenance.”

An intraverbal prompting study reported participant-specific maintenance timing and also described a scheduling omission. A behavioral skills training study arranged maintenance without prompts or feedback under its procedure. A safety-skills study conducted checks at 2 weeks, 1 month, and 3 months after its study-specific criteria.

These examples make timing, procedure, and missing observations material to interpretation. Their populations, skills, risks, designs, and outcomes neither validate this worksheet nor prescribe a schedule, prompt rule, criterion, deterioration response, or expected result for someone else.

The CASP ABA Practice Guidelines access page identifies Version 3.0 and the terms for accessing the licensed guidance. This page neither reproduces that guidance nor represents this worksheet as compliant. Use the authorized current document and applicable requirements when relevant.

Ordinary supports and meaningful conditions

Current ASHA AAC practice guidance describes multimodal communication and access considerations. Keeping a skill available should not require removing the person's usual AAC, interpreter, positioning, vocabulary, mobility support, sensory support, corrective lenses, or prescribed equipment. An access change should be documented and reviewed, not mistaken for loss of a skill.

Choose probe conditions that matter to the client rather than manufacturing deprivation or avoidable burden. Record assent, dissent, requests to pause, and reasons a planned opportunity did not occur. Do not withhold food, water, bathroom access, prescribed care, emergency communication, or another ordinary right to complete a schedule.

Record integrity and scope

The BACB Ethics Codes hub is the starting point for current controlling documents. Relevant duties in the Ethics Code for Behavior Analysts span working within competence, protecting welfare and confidential information, obtaining required consent, maintaining accurate records, and representing services and outcomes carefully. Which provisions apply depends on the role and facts. This worksheet offers neither ethics nor legal advice.

Keep identifiable probe and exposure details in the organization's authorized record environment. HHS Security Rule risk-analysis guidance frames risk analysis for electronic protected health information but does not approve a spreadsheet, personal device, messaging channel, retention period, or disclosure. Use the applicable controls for access, corrections, auditing, exports, backups, retention, and disposal.

Do not backfill a missed check from memory or infer an unobserved response. Label late entries. When the target or procedure changes, close the old series and preserve its record. When teaching resumes after a probe, retain the initial result and start the appropriate new teaching or reference phase rather than rewriting the observation.

Missed-probe and revision log

DatePlanned check or versionOriginal schedule/entryMissed, corrected, or revised entryReason and source evidenceAuthorized byOpportunities affectedSummary affected

Do not continue interpreting when the reference phase, target definition, actual interval, interim exposure, ordinary support, valid-probe rule, client authorization, safety plan, responsible clinician, or record destination is unresolved. Escalate an urgent safety concern through emergency and clinical channels. Send distress, withdrawn assent, inaccessible communication, suspected record alteration, privacy or security exposure, serious integrity problems, and out-of-scope work to the responsible organizational, clinical, compliance, legal, payer, or regulatory reviewer as applicable.

Related resources

Sources