An ABA massed vs distributed practice worksheet has one job: make the schedule visible enough to review. “Massed” and “distributed” are relative labels; neither tells a reviewer how many opportunities occurred, how far apart they were, whether the same dose was delivered, or when retention was tested.

Use this ABA massed vs distributed practice worksheet to separate spacing, planned dose, valid exposure, initial responses, teaching time, calendar time, retention probes, learner experience, and integrity. The fictional example reconciles the arithmetic without identifying a preferred schedule.

Clinicians & ABA Professionals / Assessment and Treatment Planning.

Clinical boundary: Use practice-spacing decisions only within an authorized, individualized process led by qualified professionals. Keep ordinary speech, AAC, gesture, sensory, motor, visual, medical, safety, relationship, and access supports available during every schedule. A pause, refusal, help request, pain report, or change in participation calls for an accessible response and review, not for extra trials. The BACB ethics resources provide current professional context; this worksheet cannot establish consent, assent, necessity, effectiveness, coverage, or legal compliance.

A schedule label is not a schedule

Massed practice concentrates comparable opportunities within a shorter period. Distributed practice separates them by longer intervals, sessions, routines, or days. That contrast operates at several scales. A program might distribute teaching across four days while presenting six closely spaced trials within each visit. Another might hold one visit but separate trials with naturally occurring activities. Calling both plans “distributed” would conceal an important difference.

Write the actual timing. Record trials per block, minutes between trials, blocks per session, sessions per day, days per week, total planned opportunities, and the calendar span. If a break occurs, distinguish a planned spacing interval from an interruption caused by illness, missing materials, staff availability, or the learner asking to stop.

Spacing is observable. So is dose.

Skill-purpose and comparison card

A review begins with a meaningful skill and the reason it is being taught. Next, define what must remain comparable before anyone interprets a schedule difference. Target difficulty, response form, materials, prompts, correction, reinforcement, instructors, settings, prior exposure, and mastery rules can all change the result. When they differ, record the difference rather than presenting spacing as the only variable.

FieldCase-specific entryMeaningful skill purpose and authorized plan referenceTarget set A and prerequisite evidenceTarget set B and comparability limitsObservable initial-response definition and latencyPrompt, correction, and consequence rulesValid-opportunity requirementsOrdinary AAC, sensory, motor, visual, and health supportsHelp, break, refusal, alternative, and stop responsesAcquisition, generalization, and retention measuresQualified owner, version, effective date, and review date

A comparison need not be experimental. Routine care may permit only a descriptive review. Say so plainly: matched targets still differ, and calendar order can introduce history or carryover. Document any stronger design separately.

Versioned schedule and dose map

Give each schedule its own row. Describe what should happen and how departures are counted. “Twenty trials per week” is incomplete when one plan schedules all twenty on Monday and another schedules five on four days. Equal planned totals can still produce unequal delivered dose.

Schedule ID/versionTrials per blockBlocks per sessionWithin-block spacingBetween-block spacingSessions/daysCalendar spanPlanned opportunitiesActive teaching-time ruleMissed/invalid ruleOwner

The study by Haq and colleagues compared all weekly opportunities on one day with opportunities spread across several days for three autistic children learning selected verbal skills. Distributed practice was more efficient in that study. A related study with typically developing children also reported more efficient acquisition and stronger maintenance for its distributed arrangement. Both studies are informative, but neither supplies a universal calendar or guarantees the same effect for a different learner, target, dose, or service context.

Counterevidence matters. A six-participant comparison of massed trials, distributed trials, and task interspersal found that five children acquired targets first in the massed-trial condition tested. The arrangements and measures differ from the calendar-day comparisons above. Taken together, these findings support defining the actual procedure and measuring individual outcomes rather than treating “distributed” as a synonym for better.

Exposure ledger: preserve what was actually available

One row represents one planned acquisition opportunity. Invalid opportunities remain in the schedule-delivery record but leave the valid learner-performance denominator. Missing materials, unavailable activities, incorrect stimuli, an inaccessible instruction, a device failure, or an instruction presented after the defined window are not learner errors.

Date/timeSchedule/versionTarget IDPlanned position and gapActual gapValid? reasonInitial response/latencyPrompt or correctionFinal supported responseActive minutesHelp/refusal/breakFatigue/access/contextObserveryes / noyes / noyes / no

Do not overwrite the initial response after a successful prompt or correction. Both data points may matter, but they answer different questions. A later correct response can show that support worked without being independent acquisition. Likewise, elapsed time since the previous valid opportunity is different from active teaching time and should not be reconstructed from a visit label alone.

Schedule-delivery summary

Schedule/versionPlanned opportunitiesDelivered opportunitiesInvalid opportunitiesValid opportunitiesCalendar daysActive teaching minutesMissed-session reasonsUnplanned outside exposure

Operational failure can coexist with strong-looking learner accuracy. A distributed plan may lose visits to transportation or staffing. A concentrated plan may deliver every opportunity but produce fatigue, long pauses, escalating prompts, or reduced communication. Keep feasibility and learner performance adjacent without merging them.

Learner experience and access are outcomes to review

Practice spacing changes the shape of a person's day. Some learners may value shorter repeated contacts; others may prefer one coherent block or fewer transitions. Preference may vary by task, partner, setting, energy, pain, or communication access. Ask in a form the person can use, and do not infer preference from compliance.

Review dateSchedule/versionAccessible options offeredLearner selection or descriptionBreak/help honored?Fatigue, pain, or sensory effectTransition/routine burdenRequested changeFollow-up owner

A small learner-preference study found that two autistic children selected alternating mastered and unknown trials over a massed sequence in a computerized task. That comparison concerns within-session order, not distribution across calendar days. It does show why choice should be measured directly rather than assigned from an adult's assumptions.

The ASHA AAC Practice Portal describes AAC as individualized, flexible, and multimodal. A device, signs, gestures, speech, writing, partner support, and sufficient processing time remain available across schedule conditions. A distributed schedule with inconsistent device access is not a fair test of spacing, and a concentrated schedule cannot condition communication access on completing a block.

Retention and generalization require their own clocks

Acquisition data tell what happened during teaching. Retention asks what occurs after a stated interval under a stated level of support. Generalization asks about new examples, people, settings, or response forms. Schedule records should identify which question a probe addresses, the elapsed time since the last teaching contact, any intervening exposure, and whether the probe was truly unprompted.

Probe date/timeSchedule/target setLast teaching contactElapsed intervalProbe type/contextValid? reasonInitial responsePrompt/support after initial scoreIntervening exposure known?Interpretation limitretention / generalizationyes / noretention / generalizationyes / no

A recent study of different follow-up-session distributions with nine college-aged learners illustrates why follow-up timing and repeated measurement matter. Some practice arrangements produced better maintenance than the control, but outcomes were not uniform across individuals. The population and tasks differ substantially from child ABA services. Use the article to sharpen measurement questions, not to select a clinical schedule.

Do not pool training responses with later probes. If four retention probes occur one day after one schedule and four occur three weeks after another, the raw percentages are not directly comparable without qualification. If a prompt is delivered before the initial probe response, record that probe as prompted or invalid under the plan rather than calling it retained independently.

Separate denominators answer different review questions

Review questionNumeratorDenominatorResultDoes not establishValid exposure coveragevalid acquisition opportunitiesplanned acquisition opportunitieslearner accuracy or schedule benefitAcquisition initial accuracyinitial correct responsesvalid acquisition opportunitiesretention, generalization, or causationPlanned-opportunity yieldinitial correct responsesall planned acquisition opportunitieswhy exposure was lostRetention probe accuracyinitial correct retention responsesvalid retention probesgeneralization or durability beyond that intervalGeneralization probe accuracyinitial correct generalization responsesvalid generalization probesretention or causal effect of spacingImplementation integritycorrectly implemented componentsobservable component opportunitieslearner benefit, assent, or preference

Keep the counts next to each percentage. “Seventy-five percent” means little without the denominator, timing, schedule version, response definition, and validity rules. A planned-yield view is useful because it keeps system loss visible, while valid-opportunity accuracy avoids assigning system failures to the learner. Both are needed; neither replaces the other.

Fictional library maker-activity example

In this fictional review, a client chose a library maker activity that involved identifying tool symbols. Two matched but nonidentical target sets were assigned descriptively to a concentrated schedule and a distributed schedule. Each schedule planned 18 acquisition opportunities. The concentrated schedule placed them in a shorter block; the distributed schedule spread them across several visits.

Two concentrated-schedule opportunities were invalid: the symbol set was incomplete once, and the communication display was unavailable once. Sixteen valid opportunities remained, with 11 initial correct responses. One distributed opportunity was invalid when the planned activity changed, leaving 17 valid opportunities and 13 initial correct responses.

  • concentrated valid coverage: 16 / 18 × 100 = 88.9%;
  • concentrated initial accuracy: 11 / 16 × 100 = 68.8%;
  • concentrated planned yield: 11 / 18 × 100 = 61.1%;
  • distributed valid coverage: 17 / 18 × 100 = 94.4%;
  • distributed initial accuracy: 13 / 17 × 100 = 76.5%; and
  • distributed planned yield: 13 / 18 × 100 = 72.2%.

At the same stated follow-up interval, each target set had eight valid unprompted retention probes. The concentrated set had six initial correct responses, 6 / 8 × 100 = 75.0%. The distributed set had seven, 7 / 8 × 100 = 87.5%.

These descriptive results do not prove that distribution caused better acquisition or retention. Targets, invalidity, history, and outside exposure could differ. The record supports another question, not a declaration of superiority.

Implementation was reviewed separately in 12 observations across eight defined components, creating 96 opportunities. Ninety-one were correct and five were mismatches, which reconciles to 96. Integrity was 91 / 96 × 100 = 94.8%.

Evidence layerConcentratedDistributedBounded readingPlanned dose1818Equal planned counts onlyValid exposure1617Different system loss remains visibleInitial acquisition11/1613/17Describes these target sets and periodPlanned yield11/1813/18Includes missed or invalid exposureRetention6/87/8Small denominators at one intervalActive/calendar timerecord separatelyrecord separatelyEfficiency depends on the selected unitLearner experiencedirect record requireddirect record requiredAccuracy cannot establish preferenceIntegrity91/96 combined91/96 combinedFive provider-side mismatches need review

Begin with the three invalid opportunities, five fidelity mismatches, target-set comparability, actual gaps, active time, calendar burden, prompt use, access, and the learner's feedback. The arithmetic does not choose retain, revise, compare, or stop.

Implementation-integrity record

ObservationCorrect version availableTarget and schedule identifiedValidity checkedSpacing and dose accuratePrompt/correction accurateConsequence accurateAAC/access preservedHelp/refusal honoredCorrect componentsOpportunitiesNotes

A mismatch may reflect unclear schedule notation, material preparation, timer drift, inconsistent partners, missing access supports, or version confusion. It should not be recoded as a learner variable. If fidelity differs systematically across schedules, qualify any outcome comparison.

Review and decision record

FieldReview entryVersions, targets, settings, partners, and dates reviewedMeaningful purpose and continued authorizationPlanned versus delivered dose and actual spacingInitial, prompted, corrected, and latency resultsActive teaching time and elapsed calendar timeRetention/generalization probe timing and resultsInvalidity, outside exposure, and fidelity limitsLearner choice, assent/dissent, fatigue, burden, and unwanted effectsHold / revise / compare / retain / fade / stop questionQualified decision owner, rationale, and next review date

The Ethics Code for Behavior Analysts places client rights, involvement, consent, assent when applicable, competence, confidentiality, documentation, and responsible data use within professional duties. The CASP ABA Practice Guidelines access page provides current context for individualized treatment planning and evaluation. Neither endorses this worksheet or a specific practice schedule.

Several findings should reopen the arrangement: a changed purpose, repeated exposure failures, growing fatigue or distress, unequal communication access, excessive schedule burden, poorly matched target sets, unstable fidelity, or weak follow-up evidence. The responsible team may decide to concentrate, distribute, embed, simplify, pause, or stop practice. That decision requires more than a label and a percentage.

Related resources

Sources

The studies differ in participants, targets, schedules, time scales, consequences, measures, and designs, and their findings are not uniform. They support explicit timing, dose, access, direct measurement, follow-up, learner feedback, and cautious interpretation. They do not provide a universal trial count, spacing interval, calendar, prompt, consequence, mastery rule, probe interval, or expected effect.