Task analysis and step-by-step teaching

• Identity and version: Name the component or programme, its edition or protocol where relevant, and the purpose being evaluated.
• Source role: Separate direct comparative evidence, descriptive implementation findings, guidance, theoretical support and firsthand experience.
• Population and setting: Retain the studied age, communication/access characteristics and setting. State when an intended application lies outside them.
• Intended and actual delivery: Record the planned components and dose separately from what was received, by whom, and with what support or interruptions.
• Comparison and other support: Name the actual comparison and concurrent changes. If there is no comparison group or it is unreported, say so rather than supply an assumed usual-care condition.
• Outcome and reporter: Record the outcome, time point and reporter. Keep the young person’s account separate from proxy judgement, and access or participation separate from task performance.
• Fidelity and adaptation: Record whether the core components were delivered and describe adaptations; distinguish an adapted application from the protocol tested.
• Burden and unwanted effects: Record time, effort, practical access, cost or workload where known, alongside refusal, fatigue, distress, restriction and other unwanted consequences. Unreported burden is not evidence of no burden.

A019 separates analysis of a meaningful task from the methods used to teach it. The first describes steps, decisions and barriers; the second specifies explanations, models, practice, cues and feedback. Chaining is one way to order instruction, and prompting is one source of assistance. None should be supplied by implication from a family label.

The complete earlier article and its bounded earlier evidence section are preserved exactly with conversation identifiers and hashes. Both sit inside messages that reached the capture limit, but their own sections end before a subsequent named entry. This establishes bounded completeness, not recovery of the current master. The historical evidence section used broad strength labels and pooled-literature claims. This revision replaces those labels with source-specific conclusions and leaves unverified aggregate leads visible in the ledger.

The retained corpus comprises two professional/review sources, three fully inspected original small experimental studies and one abstract-limited original study. It deliberately includes mixed comparisons, implementation measures and a direct procedure-choice assessment. It is a targeted reconstruction, not an exhaustive account of every task-analytic teaching study.

Autism Focused Intervention Resources and Modules (AFIRM)'s older packet identifies six qualifying single-case studies under the 2014 review. National Clearinghouse on Autism Evidence and Practice (NCAEP)'s later report lists thirteen articles across the recategorised 1990–2017 corpus. Dates and classification changes matter; these are not two estimates of the same fixed population. Neither count supplies a standardized task-analysis dose or a percentage of children who benefit.¹²

The NCAEP process required positive effects on at least one outcome for the evidence-based-practice inventory. It also excluded some equal-result alternating-treatment comparisons in its overall screening. That method serves its classification purpose, but the resulting inventory should not be read as a balanced pooled estimate of benefits, null effects and unwanted consequences. Small studies of particular packages can demonstrate learning under controlled conditions without showing that task analysis was the active ingredient.²

This explains the place of the comparison group studies here. A method does not become ineffective because it fails to outperform another active method. Conversely, two methods producing acquisition with a nonsignificant comparison have not necessarily been shown equivalent. Exact learner, task, exposure and consequence conditions matter.³⁴

Slocum and Tiger held many features of experimental motor tasks constant. That strengthens a narrow teaching-order comparison while limiting its application to self-care, flexible practical action or learning in a busy classroom. Physical guidance, models and rewards were shared ingredients. Choosing a procedure because it won a brief trial was not reliably predictive of performance on longer sequences in this small sample.³

McKay and colleagues addressed a different component question: whether to fade the level of manual guidance as well as using a delay. Participants had already learned with physical prompting. The study did not compare physical guidance with a method without it, and its adult participant must not be silently counted as a child. The absence of a significant difference across the nested comparisons does not justify a universal manual-prompt policy.⁴

The shopping study supplies a more natural task, but it combined in-store teaching and simulation. Six pupils reached criterion; partial progress or incomplete instruction in others matters. The reported generalisation concerns the same grocery items in a new shop and maintenance is limited to the reported period. It is not evidence of general shopping judgement, financial decision-making or independent community participation.⁵

Browder and colleagues also demonstrate why the unit of intervention needs naming. Teachers learned a lesson template, systematic prompting and self-monitoring; pupils then received more opportunities to respond. The study primarily staggered teacher training. Its pupil data do not isolate improvement in knowledge from expanded opportunities to demonstrate it, and the authors identify the need for a clearer pupil-level experiment. Sustaining lesson responses across books is not the same as mastering the books' abstract themes.⁶

Direct choice was an unusual strength of the Slocum study, but it was a choice among arranged alternatives. The no-teaching condition provided neither the same interaction nor the same access to leisure rewards. Forced-exposure trials also occurred, although only free choices counted toward the preference measure. The study therefore offers evidence about behaviour under those conditions, not blanket endorsement of chaining or physical contact.³

Other sources rely largely on performance or adult judgements. McKay's discussion of apparently non-aversive guidance is an author interpretation in a selected group; Browder measured teacher satisfaction, not pupils' experience. These gaps should remain visible rather than being filled by the word “independence”.⁴⁶

For Atlas, dignity, meaningful goals, accessible refusal and proportionate support are editorial requirements. They are not claims that a particular study demonstrated coercion or caused harm. A local review should distinguish what the person chose, what help occurred, what the person could do and how they experienced it. Continued use of a useful tool can be a valid outcome; uneventful completion can still conceal substantial effort or unwanted direction.

High observation agreement does not establish treatment fidelity. In the Slocum study, agreement concerns recording responses; it is not an independently reported fidelity percentage. In Browder's study, the fidelity score for lesson steps did not include consistency of pupil prompting. McKay reported a more specific sampled procedural-integrity assessment. These different measures should not be merged.³⁴⁶

No common optimal dose, universally preferred chaining direction, smallest correct step size or compulsory fading schedule is established. The article's practice illustration is constructed, with no observed outcome. Its five planning components describe the scope of this working family, not a tested new protocol.

The shopping paper's full methods, broader child-led experience, longer-term everyday transfer and application beyond the inspected populations remain follow-up gaps. Historical large instructional-review counts have not been used to inflate this family’s evidence. All six source records must retain their access levels when reused. Separate AI editorial review is recorded in the batch support records. Human specialist and recipient review, current-master/source-version selection, adoption and publication remain pending. No professional consultation is claimed.

A019-S01 — Task Analysis: evidence-based practice (EBP) Brief Packet — Sam and AFIRM Team

Access: Full public packet opened; identity, evidence and implementation sections read. Locator: Packet pp.2–8 and references.

• Identity and version: AFIRM task-analysis guidance, based on the 2014 review.
• Source role: Professional implementation resource, not an additional primary trial.
• Population and setting: Autism; six cited single-case studies, ages 6–14 in this older packet.
• Intended and actual delivery: Analyse the skill, select chaining and separate teaching practices, prepare materials, monitor. No universal dose.
• Comparison and other support: No assigned comparison group; underlying interventions heterogeneous.
• Outcome and reporter: Practice-level summary of targeted outcomes; no pooled family effect.
• Fidelity and adaptation: Implementation checklist and individual adaptation described.
• Burden and unwanted effects: No systematic harms, child-experience or comparative cost analysis supplied.

Source limits: Six here is an older classification, not a contradiction of the 2020 recategorised 13-article inventory. Provider guidance and claimed low resource cost are not treated as independent effectiveness or economic evidence. ¹

A019-S02 — Evidence-Based Practices for Children, Youth, and Young Adults with Autism — Steinbrenner and colleagues

Access: Full public report opened; methods, classification and task-analysis appendix inspected. Locator: Printed pp.22–30 and pp.127–128.

• Identity and version: NCAEP classification covering 1990–2017 literature.
• Source role: Original systematic evidence classification; not a pooled component-effect estimate.
• Population and setting: Autistic children/young people, mixed developmental profiles; 13 task-analysis articles, aggregate task-analysis participant n not verified.
• Intended and actual delivery: Task-analysis packages commonly include other teaching practices; no common dose.
• Comparison and other support: Study-specific comparisons; equal-result alternating-treatment studies were among overall exclusions.
• Outcome and reporter: Positive effects on at least one outcome determine EBP inventory; 13 is not a universal success rate.
• Fidelity and adaptation: Study quality screened; multiple components could be assigned to several categories.
• Burden and unwanted effects: No task-analysis-specific pooled harms/acceptability estimate.

Source limits: Classification is selective for positive effects and cannot supply a balanced pooled benefit–harm estimate. Not an Atlas certainty grade, not a guarantee across special educational needs and disabilities (SEND), not 13 independent replications of an identical protocol. ²

A019-S03 — An Assessment of the Efficiency of and Child Preference for Forward and Backward Chaining — Slocum and Tiger

Access: Full original article PDF opened; methods, results and discussion read. Locator: JABA 44 pp.795–804.

• Identity and version: Forward/backward chaining of experimental motor sequences.
• Source role: Adapted alternating-treatments comparisons and concurrent-chain preference assessment.
• Population and setting: Four US schoolchildren aged 10–12; attention deficit hyperactivity disorder (ADHD), speech or learning difficulties; not an autism-only sample.
• Intended and actual delivery: 3/6/9/18-step sequences; prompts, physical guidance and rewards; 1–6 ten-trial sessions in a 45-minute block, usually four days/week.
• Comparison and other support: Forward versus backward; preference baseline lacked prompting, interaction and reward access.
• Outcome and reporter: All acquired; 16 comparisons favoured forward 8 / backward 6 / tied 2. No consistent winner; short-task predictions weak. Choices were observed, not satisfaction interviews.
• Fidelity and adaptation: Counterbalanced tasks/order; response-observation agreement high, separate treatment-fidelity estimate unreported.
• Burden and unwanted effects: Daily assent recorded; physical guidance and rewards used; direct choices contextual, no systematic harms outcome.

Source limits: Small contrived-task study; no everyday-transfer or durable-effect estimate. No statistical equivalence established. Control and reward contingencies limit interpretation of preference; authors question one apparent preference. ³

A019-S04 — Comparison of Prompting Hierarchies on the Acquisition of Leisure and Vocational Skills — McKay and colleagues

Access: Full author copy opened; methods/results/limitations read; PMC access blocked. Locator: Participants; Experimental design; Procedural integrity; Results and Discussion.

• Identity and version: Two physical-prompt hierarchies with constant two-second delay.
• Source role: Alternating-treatments study; three comparisons per participant.
• Population and setting: Three adolescents aged 14–15 and one adult aged 37, intellectual disability; diagnoses included Angelman syndrome and pervasive developmental disorder (PDD)/autism spectrum disorder (ASD). All had prior physical-prompt teaching.
• Intended and actual delivery: Forward chaining of 12-step construction tasks; about three sessions/day, three days/week; edibles and praise.
• Comparison and other support: Gradually faded manual guidance versus full guidance after delay; neither arm omitted physical guidance.
• Outcome and reporter: All acquired/generalised to another setting/instructor; trial-to-mastery difference nonsignificant (p=.175).
• Fidelity and adaptation: Integrity mean 99%, sampled 33% of sessions; controlled tasks and prior experience.
• Burden and unwanted effects: No formal learner acceptability/harms measure; authors judged guidance non-aversive in this selected sample.

Source limits: Abstract ASD wording is broader than the diagnosis detail in Methods. Twelve comparisons are nested within four people; nonsignificance is not equivalence. Instructor ease and no observed aversion do not establish recipient preference or justify routine physical guidance. ⁴

A019-S05 — Teaching Elementary Students with Moderate Intellectual Disabilities How to Shop for Groceries — Morse and Schuster

Access: Publisher abstract and reference record opened; full methods not accessible. Locator: Abstract.

• Identity and version: Grocery-shopping package combining real and simulated practice.
• Source role: Original multiple-probe-across-participants study; abstract-limited.
• Population and setting: Ten elementary-aged pupils with moderate intellectual disability; exact ages/recruitment details unavailable in abstract.
• Intended and actual delivery: In-store constant-time-delay teaching plus pictorial storyboard simulation; received dose unavailable.
• Comparison and other support: Baseline/probe phases, not an active matched teaching comparison group.
• Outcome and reporter: Six reached criterion; two improved but school year ended before criterion. Six-week retention/new-store same-item transfer reported for achievers.
• Fidelity and adaptation: Detailed procedural integrity and adaptations unavailable.
• Burden and unwanted effects: Time/transport and recipient experience not quantified; harms not reported in abstract.

Source limits: Not all ten achieved criterion; outcomes for remaining pupils cannot be reconstructed from this abstract. No isolated task-analysis effect or broad community independence claim. ⁵

A019-S06 — Training Teachers to Follow a Task Analysis to Engage Middle School Students With Moderate and Severe Developmental Disabilities in Grade-Appropriate Literature — Browder, Trela and Jimenez

Access: Full permitted institutional manuscript opened; methods/results/limits read; not version of record. Locator: PDF pp.5–14,16,19–21; teacher and student data, limitations.

• Identity and version: Teacher 25-step lesson template plus prompting and self-monitoring.
• Source role: Multiple-probe across three teachers; pupil responses secondary.
• Population and setting: Six pupils aged 12–14, intelligence quotient (IQ) eligibility <55, significant intellectual/developmental disability; some augmentative and alternative communication (AAC). US special-education classrooms.
• Intended and actual delivery: General workshop/books then staggered individual modelling, role play and self-monitoring; ongoing lessons. Exact minutes/dose not reported.
• Comparison and other support: Baseline already included general training and adapted books.
• Outcome and reporter: Teacher steps and independent pupil responses increased; opportunity-to-respond changes partly determine pupil scores. Broad comprehension not established.
• Fidelity and adaptation: Trainer fidelity 100%; teacher scoring omitted consistency of pupil prompting. Observer agreement sampled 25%.
• Burden and unwanted effects: Teacher satisfaction positive; no direct learner experience measure. Adaptation workload and little collaboration time; no mainstream inclusion achieved.

Source limits: Teacher-level intervention, not a clean test of student task analysis alone. Small sample, changing response opportunities and books; comprehension/transfer distinction unresolved. No specialist or learner approval inferred from a published instructional method. ⁶

1. 1.

   Task Analysis: EBP Brief Packet — Sam and AFIRM Team. 2015 citation; packet includes 2016 page footers. https://files.eric.ed.gov/fulltext/ED595409.pdf. Accessed 2026-09-12; Full public packet opened; identity, evidence and implementation sections read. Locator: Packet pp.2–8 and references.

   ↩1 ↩2
2. 2.

   Evidence-Based Practices for Children, Youth, and Young Adults with Autism — Steinbrenner and colleagues. 2020. https://files.eric.ed.gov/fulltext/ED609029.pdf. Accessed 2026-09-12; Full public report opened; methods, classification and task-analysis appendix inspected. Locator: Printed pp.22–30 and pp.127–128.

   ↩1 ↩2 ↩3
3. 3.

   An Assessment of the Efficiency of and Child Preference for Forward and Backward Chaining — Slocum and Tiger. Winter 2011; publisher webpage later-posting date differs. https://files.eric.ed.gov/fulltext/EJ964433.pdf. Accessed 2026-09-12; Full original article PDF opened; methods, results and discussion read. Locator: JABA 44 pp.795–804.

   ↩1 ↩2 ↩3 ↩4 ↩5
4. 4.

   Comparison of Prompting Hierarchies on the Acquisition of Leisure and Vocational Skills — McKay and colleagues. 2014. https://www.researchgate.net/publication/266673397_Comparison_of_Prompting_Hierarchies_on_the_Acquisition_of_Leisure_and_Vocational_Skills. Accessed 2026-09-12; Full author copy opened; methods/results/limitations read; PMC access blocked. Locator: Participants; Experimental design; Procedural integrity; Results and Discussion.

   ↩1 ↩2 ↩3 ↩4 ↩5
5. 5.

   Teaching Elementary Students with Moderate Intellectual Disabilities How to Shop for Groceries — Morse and Schuster. January 2000. https://journals.sagepub.com/doi/10.1177/001440290006600210. Accessed 2026-09-12; Publisher abstract and reference record opened; full methods not accessible. Locator: Abstract.

   ↩1 ↩2
6. 6.

   Training Teachers to Follow a Task Analysis to Engage Middle School Students With Moderate and Severe Developmental Disabilities in Grade-Appropriate Literature — Browder, Trela and Jimenez. Winter 2007. https://libres.uncg.edu/ir/uncg/f/B_Jimenez_Training_2007.pdf. Accessed 2026-09-12; Full permitted institutional manuscript opened; methods/results/limits read; not version of record. Locator: PDF pp.5–14,16,19–21; teacher and student data, limitations.

   ↩1 ↩2 ↩3 ↩4