An ABA self-management worksheet should help a client use information for a purpose they understand and value. It should not quietly become a surveillance log, an honesty test, or a demand to perform without accommodations. A useful worksheet records what the client is observing, how they can respond, which supports stay available, what data are shared, and how the team will notice burden or a reason to change course.

This artifact separates opportunity availability, target performance, self-recording, sampled agreement, fidelity, and support-fading questions. It cannot select a target or prescribe a self-management package. The fictional example shows the arithmetic without supplying a benchmark or real-person decision.

Clinicians & ABA Professionals / Assessment and Treatment Planning.

Clinical boundary: Use only within an authorized, individualized process led by qualified professionals. Preserve ordinary communication, AAC, sensory, motor, health, safety, privacy, relationship, and access supports. A self-record belongs to an agreed clinical purpose; it is not permission for unlimited observation or disclosure. Pause when participation changes, recording creates distress or stigma, privacy cannot be protected, the context is unsafe, or the tool interferes with the activity it is meant to support. The BACB ethics resources provide professional context, but this worksheet cannot establish consent, necessity, effectiveness, payer coverage, privacy compliance, or legal compliance.

Self-management is a supported arrangement, not a character trait

Self-management can include noticing an event, recording it, evaluating the record, arranging a consequence, following a prompt, or using another selected support. Those parts do not always appear together. “Independent” is also not a sufficient definition. A person can independently use a visual checklist, speech-generating device, timer, sensory support, or help request. Removing a useful tool to make performance look less supported may reduce access rather than demonstrate progress.

The Sixth Edition BCBA Test Content Outline distinguishes measurement, intervention components, generalization, maintenance, integrity, and data-based evaluation. The BACB test-content access page identifies the current certification outlines. These materials help organize review questions; they do not define a self-management program for a particular client.

A review of interventions intended to increase independence for autistic people describes self-monitoring as one way to shift control from continuous adult management toward another cue or support. The review of self-monitoring, video modeling, and work systems does not imply that all adult help should disappear. Instead, it reinforces the need to say which support is changing, which remains available, and what outcome is actually observed.

Purpose, participation, and data boundaries

Begin with the client's purpose in language and a communication form that work for them. A target such as “be responsible” is too broad to observe fairly. Name the routine, the exact opportunity, accepted response variations, the unit of measurement, and the environmental conditions required for the opportunity to count.

The record should also explain who can see the data, why they need access, where the record is stored, when it is deleted or archived, and how the client can ask for a correction. If a phone, wearable, picture card, tally counter, or paper form is proposed, record whether the client wants it and whether it is accessible in the actual setting. The ASHA AAC Practice Portal describes AAC as individualized and multimodal. Self-monitoring must not displace a person's ordinary way of communicating choice, help, discomfort, a break, dissent, or a change request.

Blank purpose and participation card

FieldCase-specific entryClient-chosen or agreed purposeRoutine, setting, people, and meaningful outcomeObservable target and accepted variationsOpportunity start, end, and direct measureConditions required for a valid opportunitySelected self-observation cue and recording responseOrdinary AAC, visual, sensory, motor, hearing, and processing supportsHow choice, questions, help, break, dissent, and change are communicatedWho may view each data element and whyStorage, correction, retention, and deletion rulesAlternatives considered and client feedbackQualified owner, version, effective date, and review date

Do not convert private thoughts, feelings, bodily states, or identity into a public score simply because they matter clinically. When self-report is relevant, distinguish it from direct observation and preserve the client's communication as its own evidence. A discrepancy between records is a review signal, not a finding that one person is dishonest.

The versioned plan map

A compact client-facing tool may sit beside a fuller clinical record. The two should share a version identifier so a team does not fade prompts from one plan while scoring another. Each revision should identify the changed element and preserve the prior version.

Plan elementCurrent versionClient-facing descriptionAccess or supportExceptionEvidence or sourceOwnerPurpose and targetCue to notice or checkRecording response and toolSelf-evaluation step, if anyReinforcement arrangement, if anyHelp-seeking or repair responseObserver-comparison samplePrivacy and data-access ruleSupport-retention or fading questionHold, rollback, and stop conditions

A research review of technology-based self-monitoring found varied technologies, targets, and study participants and noted a limited adult evidence base. A familiar app or smartwatch is therefore not a default. The tool should fit the person, purpose, environment, access needs, and privacy limits, and the record should show when that fit was last checked.

Event records keep availability and performance apart

If the planned routine never occurred, the recording device failed, materials were unavailable, communication access was missing, or the definition changed mid-event, the entry may be invalid for target-performance analysis. Repeated invalid events can identify a system problem.

Date/timePlan versionContext and opportunityAvailable and accessible?Initial target responseMeets case definition?Self-record made?Cue or support usedHelp/dissent/unwanted effectValid? reasonObserver, if sampledyes / no / not scorableyes / no / not scorableyes / no / not scorable

Keep the initial response visible even when later support helps the person finish. The initial measure and final supported outcome can both be useful, but they answer different questions. A missing self-record also should not rewrite an observed target response as a failure; it indicates that one component of the arrangement was absent or not completed.

Observer comparison without turning the plan into surveillance

Observer checks can answer a narrow question about agreement under sampled conditions. They should be planned, proportionate, and transparent. Record how comparison moments are selected, what the observer can access, whether the client knows when observation can occur, and how discrepancies will be discussed. Constant hidden observation is not the default implied by a column labeled “accuracy.”

Comparison IDSampling ruleTarget/opportunityClient recordObserver recordAgreement under definition?Difference typeAccess or context issueClient responseFollow-upyes / no / not comparableyes / no / not comparable

Agreement means the records meet a predefined matching rule. That comparison cannot establish ground truth. Both observers can agree under a poor definition or disagree because they saw different parts of the event, the cue was ambiguous, the tool lagged, or the opportunity was not comparable.

A study of staff self-monitoring with accuracy feedback directly compared staff estimates with observer records. The study of training, prompting, and self-monitoring demonstrates one bounded way comparison data have been collected. Its participants, criteria, feedback, and classroom context do not establish a required observer-check frequency or accuracy threshold here.

Measurement layers and their denominators

Review questionNumeratorDenominatorResultWhat it does not showValid opportunity coveragevalid practice blocksplanned practice blockstarget performance or benefitPerformance among valid blocksvalid blocks meeting the target definitionvalid practice blocksplanned coverage or causationPlanned-block yieldvalid blocks meeting the target definitionall planned blockswhy blocks were invalidSelf-record completioncompleted self-recordsvalid record opportunitiesrecord accuracy or target performanceSampled record agreementsampled comparison opportunities meeting the agreement rulecomparable sampled opportunitiesground truth or unsampled accuracyImplementation fidelitycorrectly implemented componentsobservable component opportunitiesclient outcome, assent, or effectiveness

Preserve raw counts and definitions. A percentage without its denominator can make unavailable routines look like poor performance or make a small observer sample look comprehensive. No row in this table supplies a universal criterion.

Fictional community library example

The fictional client chose a media-shelving routine at a community library because it supported participation in a preferred volunteer activity. Twelve practice blocks were planned. In one block the assigned cart never arrived; in another, the library changed the activity. Those two blocks were invalid for the case-defined routine, leaving 10 valid blocks. The client independently completed the defined routine in eight.

  • valid block coverage: 10 / 12 × 100 = 83.3%;
  • performance among valid blocks: 8 / 10 × 100 = 80.0%; and
  • planned-block yield: 8 / 12 × 100 = 66.7%.

The fictional review also sampled 30 individual comparison opportunities selected under a written rule. The self-record and observer record agreed on 27 and differed on three. Sampled agreement was 27 / 30 × 100 = 90.0%. That is a property of this comparison sample and definition, not proof that every self-record was accurate or that the observer was correct.

Implementation was reviewed separately across 18 observations and seven components, creating 126 observable component opportunities. One hundred seventeen were correct and nine were mismatches, reconciling to 126. Fidelity was 117 / 126 × 100 = 92.9%.

Fictional result table

Evidence layerFictional resultBounded interpretationPlanned coverage10 valid of 12 plannedTwo blocks could not test client performanceTarget performance8 of 10 validDescribes this definition and these blocks onlyPlanned yield8 of 12 plannedKeeps unavailable blocks visibleSampled record agreement27 of 30 comparisonsDoes not establish truth or unsampled agreementImplementation fidelity117 of 126 componentsNine mismatches require reviewBurden and reactivitydescriptive record neededThe percentages do not show effort, stigma, or disruptionGeneralization or maintenancenot establishedNo durable or cross-setting claim follows

The numbers invite questions about the two unavailable blocks, the three comparison differences, the nine fidelity mismatches, client feedback, and whether the recording step helps the chosen purpose. They do not automatically select continue, fade, revise, or end.

Burden, reactivity, and help-seeking belong in the review

Self-monitoring can change the activity being observed. That reactivity may be useful, burdensome, distracting, stigmatizing, or mixed. Ask whether prompts interrupt concentration, whether recording attracts unwanted attention, whether the tool is physically or cognitively difficult to use, and whether the client would rather use another method. A high completion rate cannot answer those questions.

Review dateClient feedback and communication formEffort or interruptionPrivacy or stigma concernHelp request available and honored?Tool still useful?Change requestedOwner

The BACB Ethics Code for Behavior Analysts places client rights, involvement, informed consent, assent when applicable, confidentiality, documentation, competence, and data use within the responsible professional's duties. A completed ABA self-management worksheet cannot demonstrate that those duties were met.

Support changes require an explicit question

Fading can refer to the cue to check, a prompt for the target response, an observer comparison, feedback, reinforcement, the recording form, or adult proximity. Those are not interchangeable. The plan should name one element, identify what stays constant, and state what evidence would support a pause or return to the earlier version.

A small study used self-management to support reciprocal conversation and progressively changed prompts and recording supports. The self-management conversation study illustrates that support changes can be measured, but its participant-specific criterion is not a transferable fading rule. Some tools may remain useful long term. Retention can be the right outcome when the client wants the support and it improves access.

Change versionElement under reviewPrior levelProposed levelElements held constantClient feedbackDirect resultFidelityBurden or unwanted effectRollback ruleDecision owner

The CASP ABA Practice Guidelines access page provides professional context for individualized care. The page neither endorses this worksheet nor decides whether a support should be faded, retained, redesigned, or replaced.

Implementation-fidelity record

ObservationCurrent version availableValid opportunity identifiedCue delivered as writtenOrdinary access preservedInitial response recordedPrivacy rule followedDissent/help/exception honoredCorrect componentsOpportunitiesNotes

Define each component before collecting fidelity data. A mismatch may identify an unclear plan, inaccessible tool, competing demand, training gap, version problem, or environmental barrier. It is not evidence that the client lacks motivation.

Decision and exception record

FieldReview entryVersions and date range reviewedClient purpose, feedback, assent/dissent, privacy, burden, and unwanted effectsPlanned coverage, target performance, and recording completionObserver sample and known limitsFidelity mismatches and environmental barriersHelpful supports that should remain availableGeneralization, maintenance, or support-change evidence still missingConsent, rights, safety, setting, payer, or legal issue for separate reviewHold / revise / retain support / consider change / end questionQualified decision owner and next review date

The decision record should preserve uncertainty. Self-management is not established by a high percentage, tool use is not assent, agreement is not truth, and fewer adult prompts do not necessarily mean better access. The responsible team still needs the current assessment, authorized plan, direct client response, setting requirements, risks, alternatives, and qualified judgment.

Related resources

Sources

The sources span different participants, settings, targets, technologies, intervention packages, supports, measures, and designs. They support explicit definitions, direct measurement, individualized tools, bounded comparison, and visible support changes. They do not provide a universal target, device, accuracy criterion, observer schedule, prompt hierarchy, fading sequence, stopping rule, or prediction for a particular client.