To separate initial response, correction, retry, and probe evidence, assign Wei one mutually exclusive first-response code after the natural cue. Preserve the correction sequence, prompted practice, immediate retry, later teaching opportunity, and unprompted probe as separate events with their own timestamps and outcomes. Retain self-correction, help, withdrawal, invalid trials, prompts, consequences, and accessible response forms so later success never overwrites the initial response.

Code the initial response once

Use the first eligible response after the natural cue and within the declared window. Preserve correct, incorrect, help, self-correction, no response, withdrawal, and invalidity without later recoding.

Open a linked correction event

When correction is due, create a separate record for feedback, prompt, practice, duration, repetitions, stop, and completion. Link it to the initial opportunity while retaining distinct status.

Treat the retry as its own condition

State whether the retry is prompted, immediate, delayed, or presented with changed materials. Report correct and incorrect responses only among retries actually delivered.

Reserve independent probes

Define natural cue, ordinary supports, response window, prompt and feedback exclusions, opportunity selection, and timing. Keep the probe outside correction and teaching denominators.

Use this sequence to separate initial response correction retry and probe evidence

Code Wei's first response, link correction without overwriting it, record the retry condition, schedule independent probes, retain every planned state, and report each numerator with its own eligible denominator.

Build Wei's initial-response and correction event record

Create one versioned initial-response and correction event record for the community-transit route review. Include Wei's priority, natural cue, accessible response, first-response states, independent window, correction due event, written sequence, prompt, feedback, retry, consequence, latency, repetitions, stop, return to instruction, ordinary supports, access, health, exposure, integrity, agreement, invalidity, missingness, withdrawal, direct experience, burden, later independent probes, generalization, maintenance, result, uncertainty, decision, owner, correction, and review trigger. Store only decision-relevant information with role-limited access.

Validate Wei's correction evidence

Reproduce Wei's 42 planned, two invalid, 19 + 9 + 5 + 4 + 3 = 40 initial states, 8/9 correction coverage, 6/8 retry performance, and 8/12 probes. Verify cue, window, first response, self-correction, help, prompt, correction onset, retry, consequence, probe conditions, access, withdrawal, and missingness. Confirm that no later event overwrites the source record.

Keep Wei's evidence and experience separate

Measure planned coverage, initial responses, correction fidelity, retry performance, later probes, access, generalization, and maintenance with their own units. Ask Wei separately about clarity, effort, comfort, usefulness, burden, and whether assessment should continue. Accurate correction delivery can coexist with poor fit. Better later performance can also occur without evidence that the correction procedure alone produced the change.

Work through Wei's example

Wei has 42 planned route-choice opportunities. Two are invalid because the map fails to load. Across 40 valid initial responses, 19 are correct, nine incorrect, five requested help, four self-corrections, and three no response. Eight of the nine errors receive correction; six immediate retries are correct. Twelve later unprompted probes yield 8/12 correct. Report every denominator separately. Preserve every planned and valid unit, raw count, denominator, response code, correction version, prompt, retry, consequence, latency, repetition, access state, integrity result, probe, withdrawal, direct report, system failure, correction, and unresolved item. This fictional example demonstrates one assessment control. It offers no universal correction sequence, prompt hierarchy, reinforcement rule, independence, mastery, generalization, diagnosis, trait, treatment effect, legal conclusion, coverage decision, payment promise, or outcome guarantee.

Address Wei's main assessment risk

Recoding six successful retries as correct initial responses creates 25/40 and hides the nine errors. Pooling requested help with errors changes a functional communication response into failure. Combining 6/8 retries with 8/12 probes also mixes coached and independent conditions. Review goal relevance, cue and response definitions, opportunity quality, ordinary supports, access, health, effort, correction timing, prompts, reinforcement, partner behavior, context, client experience, authority, and causal scope separately. A high corrected-response percentage cannot repair an inaccessible, burdensome, unwanted, or ethically unsound arrangement.

Keep error-correction concepts and authority clear for Wei

For Wei's initial-response and correction event record, the ABAI Basic Principles page supplies broad higher-education context. The BACB Test Content Outline includes measurement, prompting, error correction, integrity, generalization, maintenance, and evaluation as examination content. The current BACB Ethics Code addresses competence, client involvement, consent and assent when applicable, medical needs, positive reinforcement, risk, data, documentation, and evaluation for covered people. These sources create no individualized correction procedure or practice authority.

Read the procedure comparison cautiously for Wei

In Wei's review, Leaf and colleagues offer one narrow comparison involving 28 autistic participants learning tact relations. Both defined procedures were effective and efficient under the study conditions, with different independent-correct and independent-error patterns. The result supplies no universal correction sequence, prompt, retry rule, or expected outcome for Wei.

Keep post-correction reinforcement individualized for Wei

For the community-transit route review, Cariveau and colleagues studied reinforcement following correction with four autistic children. The small, task-specific work shows why consequence details deserve measurement. It does not establish one consequence rule, magnitude, timing, or duration for Wei. Verify current choice, obtained delivery, later performance, and burden directly.

Use prompting literature within scope for Wei

The prompting decision tool organizes evidence-informed variables for selecting prompting strategies, including skill, task, error history, prompt type, fading, efficiency, and context. Apply those questions to Wei's correction prompt without converting the tool into a universal hierarchy, legal authority, or outcome guarantee.

Preserve treatment scope, AAC, assent, and access for Wei

The CASP public summary supplies high-level ABA behavioral-health-treatment scope for autistic people, with licensed detail outside this page. ASHA's AAC portal says AAC users should always have access to their communication tools or devices. Breaux and Smith provide assent-focused practice guidance in an evolving evidence base. Follow governing sources and preserve communication, health care, food, water, bathroom use, mobility, rest, relationships, safety, and emergency help for Wei.

Choose Wei's next bounded action

Wei chooses the help symbol and spoken question as equivalent routes and asks partners to preserve self-correction time before beginning the written correction. Record the qualified owner, supporting evidence, effective date, current procedure version, ordinary supports, health and access state, implementation check, accessible explanation, disagreement route, and reassessment trigger. Preserve the original assessment when cues, responses, people, systems, tasks, health, or priorities change. A new arrangement creates a dated version rather than an error in Wei's earlier performance.

Close Wei's correction review

Review the initial-response and correction event record with Wei, the qualified behavior analyst, relevant partners, access owners, and specialists named in the manifest. Confirm that initial responses remain separate from correction, retry, and probe evidence; every denominator is reproducible; AAC and basic access remain protected; health and motor questions stay with qualified professionals; and conclusions remain bounded to the sampled response, task, people, and context. Keep this page draft and noindex until every required review is complete.

Related resources

Sources