To define response classes and variability dimensions, first identify what Luis's responses accomplish and which forms belong together. Then select the feature whose differences matter to his goal, such as tool category, sequence, partner, material, timing, or message. Record accessible equivalents, accepted repetitions, unsafe forms, prompts, and ambiguous cases. The scoring rule should reflect meaningful diversity rather than visible difference alone.

Start with functional equivalence

Group responses only when they share the relevant effect for Luis's decision. Similar movement can serve different functions, and different forms can communicate the same request. Preserve that distinction in the data dictionary.

Choose the dimension before scoring

Topography, sequence, temporal pattern, location, material, partner, message content, and outcome can each vary. Select one or a small set with a defensible link to the goal.

Write inclusion and boundary examples

List clear members, nonmembers, accessible equivalents, repeated but valid forms, unsafe variants, prompt products, and ambiguous cases. Train observers on the same examples and sample agreement.

Separate diversity from quality

A response can be different and ineffective, familiar and excellent, or novel and unsafe. Track accuracy, usefulness, choice, latency, effort, and partner response separately from the diversity measure.

Use this sequence to define classes and dimensions

Clarify Luis's goal, define the function, choose the dimension, list boundaries, test observers, preserve access, score raw sequences, and revise only through a dated definition change.

Build Luis's response-class and dimension map

Create one versioned response-class and dimension map for the maker-space tool checkout. Include Luis's priority, response class, scored dimension, eligible opportunity, response window, raw sequence, distinct and repeated forms, prompts, comparison history, contingency, integrity, context, access, health, direct experience, partner response, invalidity, missingness, withdrawal, evidence limits, qualified decision owner, correction, and reassessment trigger. Store only decision-relevant information with role-limited access.

Validate Luis's variability evidence

Reproduce 18 turns, 17 valid requests, 12 distinct tool categories, four repeated categories, one word-order variant, and one invalid screen failure. Verify response function, category boundary, modality equivalence, prompt code, partner response, access, ambiguity, and Luis's priorities.

Connect Luis's evidence to a decision

Luis's map supports separate decisions about functional communication, category breadth, partner recognition, and system reliability. If the goal concerns access to needed supplies, the same message through another modality remains equivalent. If the goal concerns choice among projects, tool category may be the meaningful dimension.

Work through Luis's example

Luis completes 18 checkout requests using speech, typing, and symbol-based AAC. Twelve request a distinct tool or supply category, four repeat a still-needed category, one changes only word order, and one is unclear after the screen freezes. Tool-category diversity is 12 distinct categories across 17 valid requests. Modality remains an access field and the frozen-screen turn remains invalid. Preserve every planned and valid opportunity, response in order, category code, prompt, repetition reason, consequence, delay, partner response, access and health state, withdrawal, invalid event, correction, and unresolved item. This fictional example demonstrates one measurement control. It supplies no diagnosis, creativity judgment, universal variability target, treatment effect, coverage decision, payment promise, or outcome guarantee for Luis.

Address Luis's main interpretation risk

Treating speech, typing, and symbols as three different response classes can reward modality switching instead of meaningful choice. Treating word order as a new tool request inflates the distinct count. Excluding repeated requests can hide that the first request was missed. Review the response property, opportunity mix, history rule, reinforcement, prompts, partner behavior, access, health, direct experience, comparison, and design strength separately. A higher or lower diversity score gains meaning only through the person's goal and the conditions that produced it.

Keep scope and authority clear for Luis

For Luis's response-class and dimension map, the CASP public summary supplies high-level ABA behavioral-health-treatment scope for autistic people, while licensed detail sits outside this page. The current BACB Ethics Code addresses competence, client and stakeholder involvement, consent and assent when applicable, assessment, medical needs, risk, data, documentation, and evaluation for covered people. The BACB outline is examination content. These sources supply no individualized variability protocol or authority for Luis.

Treat variability as a measurement claim for Luis

Barba's conceptual analysis examines lag and threshold procedures and argues that common response-distribution measures can differ from the property used to arrange reinforcement. The critical review likewise evaluates the evidence and procedure details behind variability-as-operant claims. These papers show active conceptual disagreement. For Luis, define the scored property, programmed consequence, obtained consequence, and behavior measure on compatible units before making a process claim.

Use the applied lag evidence within its limits for Luis

Olin, Sonsky, and Howard used a changing-criterion design with one eleven-year-old participant, several questions, prompts, visual aids, error correction, generalization probes, and one maintenance probe. The category-naming evaluation compared progressive and terminal lag arrangements with two children and used several variability measures. These small studies illustrate procedures and measurement choices. They provide no universal lag value, dose, outcome, or creative-skill standard for Luis.

Separate communication diversity from access for Luis

The mand-variability paper reviews several dimensions of mand variability and the predominant use of lag schedules in applied research. A communication form can be familiar, repeated, and fully effective. ASHA's AAC portal says AAC users should always have access to their tools or devices. Keep Luis's modality, access, partner recognition, response effort, and functional result separate from the diversity score.

Preserve assent and direct experience for Luis

Breaux and Smith propose practice procedures for individualized assent and withdrawal responses while describing limits in the evidence base. Their paper is practice guidance rather than a separate BACB mandate. Follow governing consent and assent sources, give Luis an accessible way to accept, pause, decline, or change the activity, and treat distress or withdrawal as a response requiring action rather than a variability error.

Choose Luis's next bounded action

Luis keeps all reliable modalities available and asks staff to score the requested category plus whether the partner responded. A repeated request stays valid when the need remains unresolved. Record the qualified owner, source evidence, effective date, current task and definition version, ordinary supports, access and health state, implementation check, accessible explanation, disagreement route, and reassessment trigger. Preserve the earlier record when responses, people, settings, systems, schedules, health, or priorities change. A repaired arrangement creates a dated phase rather than an error in Luis's prior performance.

Close Luis's variability review

Review the response-class and dimension map with Luis, the qualified behavior analyst, relevant partners, access owners, and specialists named in the manifest. Confirm that meaningful diversity remains separate from novelty, recency-rule performance, generalization, creativity, preference, error, access failure, and treatment effect; every denominator is reproducible; AAC and basic access remain protected; urgent needs received action; and conclusions stay bounded to sampled conditions. Keep this page draft and noindex until every required review is complete.

Related resources

Sources