An ABA discrete trial teaching worksheet should make every trial reconstructable: what was presented, when the response window opened, what the learner did first, which prompt or correction followed, and which consequence occurred. A session percentage alone can hide invalid trials, unbalanced targets, prompted success, or repeated implementation errors across an entire teaching session.

The plan and ledger below are designed for qualified review of one individualized program. They do not make discrete-trial teaching appropriate for every learner, goal, setting, or service. Target selection, teaching format, trial dose, response definitions, prompting, correction, consequences, mastery criteria, and generalization plans stay under the current assessment and approved clinical plan.

Clinicians & ABA Professionals / Assessment and Treatment Planning.

Write the anatomy and boundaries of one valid trial

Write the observable boundaries that separate a trial from surrounding activity. A trial may include an antecedent or instruction, a response opportunity, the initial learner response, a planned prompt or correction path, a consequence, and an intertrial arrangement. The exact components vary by program, so list the components used here.

Program identityComplete before implementationWhich meaningful use did the client select?Which observable and accessible responses count?Where do the trial, response window and trial end?Which material, array, position and rotation rules apply?Which antecedent or instruction variations are allowed?How are initial correct, incorrect, no-response and unavailable states defined?Which prompt or correction sequence applies?Which consequences follow each response category?Which ordinary communication, sensory, motor and medical supports remain?Which generalization and maintenance questions will be reviewed?How are consent, assent, dissent, withdrawal and stopping handled?Which version applies, who approved it, and when?

The BACB Sixth Edition BCBA Test Content Outline distinguishes trial-based and free-operant procedures, prompting, generalization, maintenance, contextual fit, integrity, and data-based review. That separation supports explicit program components; it does not supply this client's procedure. The BACB Test Content Outlines hub identifies the current outline.

Preplan target exposure and material placement

Specify which target appears on each trial when target balance, position, order, or exposure matters. Then preserve any departure from that plan. A session total cannot show that one target received most of the opportunities, followed a predictable sequence, or always appeared in the same location.

TrialWhich target or exemplar?Which materials and position?Which antecedent?Which response window?Which prompt or correction?Which consequence?Which context variation?123456

Do not require an unrelated posture, gaze pattern, or speech form merely to make a trial appear standardized. Accepted responses may include speech, gesture, sign, AAC, writing, selecting, eye gaze, movement, or another accessible form defined by the team. The ASHA AAC Practice Portal states that people who use AAC should always have access to their communication tools or devices. Access is not an extra prompt and should not be withdrawn to manufacture an independence score.

Keep the first response separate from teaching that follows

Use one row per scheduled trial. Record the initial response before any extra prompt, then record the prompt, correction, and final outcome in separate fields. Mark a trial invalid when a required component was unavailable or delivered outside the plan, and state why.

Date/timeWhich version?Which trial and target?Was the antecedent delivered as written?What was the initial response?Which prompt or correction followed?What was the final response?Which consequence occurred?Was the trial valid?What access, assent/dissent or context mattered?Yes / No / UnclearCorrect / incorrect / no response / unavailableCorrect / incorrect / no response / not applicableYes / No, reason

An initial error followed by a correct prompted response remains an initial error and a final correct response. The two fields answer different questions. A trial with the wrong materials, instruction, timing, or target is evidence about implementation; it should not be silently counted as a learner error.

Research comparing methods for collecting DTT performance data found that first-trial-only data sometimes gave premature indications of mastery and were less sensitive to early performance changes. The study does not require one collection system, but it supports stating exactly which trials and prompt information a summary represents.

Separate trial coverage, initial accuracy and final outcome

The fictional example schedules 24 trials across three symbol-selection targets. Four trials are invalid: two use the wrong array, one instruction does not match the active version, and one prompt begins before the response window ends. That leaves 20 valid trials. Eleven initial responses are correct, six become correct after the planned prompt or correction, and three remain incorrect or have no response.

Trial categoryScheduled trialsValid trialsInitially correctCorrect after planned prompt/correctionIncorrect or no response after sequenceDelivered under the active version.20201163Antecedent, material or timing mismatch.40Not applicableNot applicableNot applicableReconciled total.24201163

  • Valid trial coverage is 20 / 24 × 100 = 83.3%.
  • Initial accuracy is 11 / 20 × 100 = 55.0%.
  • Final correct responding is (11 + 6) / 20 × 100 = 85.0%.

Coverage describes how much of the planned session can answer the defined learner question. Initial accuracy excludes prompted or corrected responses. Final correctness includes them. These percentages cannot be added or averaged, and 85.0% final correct responding is not 85.0% independent accuracy. None proves mastery, treatment effect, generalization, maintenance, benefit, assent, or readiness for a program change.

Use the ABA discrete trial teaching worksheet to retain the event rows behind each result. If the four invalid trials cluster around one target or implementer, that pattern may be more important than the session percentage.

Measure fidelity by component, not only by session

A global implementation score can look acceptable while the same critical component fails repeatedly. A study of DTT integrity found that global measures may mask component errors. Another paper compared procedural-fidelity measurement methods after defining intervention components. Those studies support a component record, not a universal list or threshold.

Choose the components that actually define this plan. In the fictional review, 19 valid trials contain all six scored components: correct materials, correct antecedent, full response window, planned prompt or correction, programmed consequence, and intertrial arrangement. That creates 19 × 6 = 114 component opportunities. Implementers complete 105 as written.

ComponentOpportunities observedImplemented as writtenMismatchesPattern or consequenceMaterials/array.19181One placement mismatch.Antecedent or instruction.19172Review target-specific wording.Response window.19181One shortened window.Prompt or correction.19172Timing differed from the version.Programmed consequence.19172Consequence category mismatches.Intertrial arrangement.19181Materials remained visible.Reconciled total.1141059Review component rows.

Component implementation is 105 / 114 × 100 = 92.1%. This value describes implementer behavior, not learner performance. Review the nine mismatches by component and trial before interpreting the 92.1% total. Nineteen fully scorable trials also differ from the 20-trial learner denominator; do not force the fidelity and outcome denominators to match.

A laboratory analysis of commission errors during DTT manipulated consequence errors under a specified task and small sample. It demonstrates why an incorrect consequence can be recorded separately. It does not establish a correction procedure, consequence, or tolerance for errors in clinical care.

Compare the teaching format with access and natural use

Discrete-trial teaching is one possible instructional format. Review whether the target is meaningful, whether the format fits the skill and context, and whether the person has accessible ways to respond, pause, request, reject, or stop. Examine pace, repetition, fatigue, affect, avoidance, prompt exposure, reinforcer access, and opportunities for spontaneous or natural use.

Review questionEvidence and datesWhat the evidence can supportMissing or conflicting informationOwner/next actionIs the target still meaningful and client-informed?Does the trial format fit the target better than a naturalistic alternative?Are response forms and materials accessible?What assent, dissent, withdrawal or burden was observed?Are prompting and correction effects acceptable?Has independent performance appeared with new people, materials or contexts?Are interdisciplinary or safety questions unresolved?

Do not use loss of basic communication, food, water, bathroom access, medical care, sensory regulation, mobility support, or emergency assistance as a programmed consequence. Do not reinterpret withdrawal or inaccessible responding as noncompliance.

Prewrite review triggers without inventing a universal criterion

Review triggerEvidence requiredOptions the reviewer may considerLimits or stop conditionsQualified owner/dateContinue the current version.Hold and collect missing evidence.Revise a target, trial component or teaching format.Probe generalization or maintenance.Stop or refer.

Specify who may change the plan and how a new version becomes effective. Do not apply a new response window, target definition, prompt sequence, or mastery rule backward to earlier trials. When cross-version data are shown together, preserve separate version labels and explain why comparison is valid.

The live BACB Ethics Codes hub and Ethics Code for Behavior Analysts provide current professional boundaries concerning competence, dignity, client participation, consent and assent when applicable, risk, records, data accuracy, written behavior-change programs, and ongoing evaluation for covered certificants. They do not approve this worksheet or a specific DTT plan. The CASP ABA Practice Guidelines access page identifies Version 3.0 and licensing terms; this page neither reproduces licensed content nor claims compliance.

A two-minute trial-set check

  • The current program version is visible.
  • Today's target set matches that version.
  • Planned exemplars are available.
  • Material positions follow the stated rule.
  • The antecedent wording is clear.
  • Response-window timing can be observed.
  • Accepted response forms are known.
  • AAC remains available for every trial.
  • Other ordinary supports are ready.
  • The initial response has its own field.
  • Prompted responses cannot overwrite that field.
  • The correction path matches the plan.
  • Consequences are available by response category.
  • Intertrial arrangements can be implemented.
  • Invalid-trial reasons have distinct codes.
  • Target order can be reconstructed later.
  • Assent and dissent indicators are understood.
  • A withdrawal response will be honored.
  • Pace and burden can be documented.
  • Component fidelity can be observed feasibly.
  • Unresolved safety questions are held.
  • The implementer knows who may change the program.

Preserve exceptions and version changes

DateWhat triggered review?Which trial and component evidence matters?What input was received?Which options were considered?What was decided, and why?Which version takes effect?Who owns the decision?Continue / hold / revise / stop / refer

Log equipment failures, unavailable AAC, material changes, interruptions, health changes, distress, withdrawal, target imbalance, and procedural mismatches without recoding them as learner errors. A decision record should point to the evidence reviewed and the version that applies next.

Use this worksheet alongside the current assessment, treatment plan, consent and assent process, safety plan, authorization, applicable law, payer rules, and interdisciplinary recommendations. It is a planning and review aid, not a prescribed curriculum, automated mastery engine, consent record, or substitute for qualified oversight.

Related resources

Sources