Professional learning unit · Secondary · New York State

Specially Designed InstructionFrom IEP to Daily Instruction

A working unit for the people who actually deliver it: eight modules that turn the language of an IEP into decisions you make on a Tuesday, in a Regents-level course, without lowering what you ask of the student.

8 modules · 12–16 hours Asynchronous or facilitated Aligned to HLPs, 2nd ed. 2024
  1. Step 1Grade-level standard
  2. Step 2Learning target
  3. Step 3Disability-related barrier
  4. Step 4IEP need
  5. Step 5Specially designed instruction
  6. Step 6Student response & data
  7. Step 7Instructional adjustment
Section 01

Unit overview

What this unit is for, who it is for, and the seven students you will plan for throughout.

A tenth grader with an IEP sits in Living Environment on Monday. She has extended time, a word processor, and the same CER graphic organizer the rest of the class has. All three are being provided exactly as written. Her lab write-ups have not moved in eight weeks, and nobody in the building is teaching her to write a scientific argument.

That is the problem this unit is about — not whether those three supports are correctly labelled, but whether anyone is teaching. The labels turn out to matter anyway, because of what they let a teacher believe. She has extended time and an organizer, so her IEP is being implemented is a true sentence, and it can be true of a student no one is instructing. Telling SDI from its neighbours is worth the trouble for exactly one reason: the confusion has a cost, and the cost is always that student.

This is an instructional design problem, not a compliance problem. IDEA and New York regulation define SDI by its purpose: adapting the content, methodology, or delivery of instruction to address needs that result from the student's disability and to secure access to the general curriculum. Neither condition can be read off a strategy list. Extended time is not SDI. Neither is a graphic organizer, or small group, or a check-in. They can all be part of SDI, or part of something else entirely, and telling the difference requires knowing why you chose them and what you expect to change.

So this unit is built around a chain of reasoning rather than a catalogue of strategies. You will run the same seven-step sequence dozens of times, on seven different students, in seven different courses, until it becomes the thing you do automatically when you open a lesson plan next to an IEP.

Who this is for

Primary audience

  • Secondary special education teachers — consultant teacher, resource room, integrated co-teaching, and special class
  • General education content teachers who have students with IEPs, which in New York is nearly all of them
  • Teaching assistants and paraprofessionals who implement SDI under supervision

Also designed for

  • Related service providers — speech-language pathologists, OT, PT, school psychologists — particularly Modules 3, 4, and 6
  • Instructional coaches — the observation checklist in Section 16 is built for coaching cycles
  • Building and district administrators — Modules 1, 2, and 6 plus the facilitator guide

The unit assumes wildly uneven starting points. A first-year teaching assistant and a fifteen-year special educator can both work through it, because the anchor is not prior knowledge, it is the seven-step chain, and everybody starts that chain in the same place.

A commitment about rigor

Design stance

This unit never uses placement as a proxy for expectation. A student in a 15:1 special class in a Regents Living Environment section is working toward the Living Environment standards. The Blueprint's third principle is that teachers design, provide, and assess the effectiveness of specially designed instruction to provide students with disabilities with access to participate and progress in the general education curriculum — access and progress, not a parallel curriculum. Where this unit does discuss modified content, it says so explicitly and explains the IEP decision that authorizes it.

Learning objectives

By the end of the unit, you should be able to:

#ObjectiveModules
1Define specially designed instruction using IDEA and NYSED terminology1
2Explain the relationship among disability-related needs, IEP goals, grade-level standards, and SDI1, 3
3Distinguish SDI from accommodations, modifications, differentiation, UDL, intervention, scaffolding, and generally good teaching2
4Identify high-leverage practices that support effective SDI4
5Identify disability-related barriers affecting access to instruction3
6Select evidence-based instructional practices matched to student needs4, 5
7Design SDI that changes content, methodology, and/or delivery without unnecessarily lowering rigor5, 7
8Implement explicit instruction, scaffolding, corrective feedback, active engagement, and systematic instruction5
9Use progress-monitoring data to determine whether SDI is working6
10Intensify or modify instruction when a student does not respond adequately6
11Translate IEP information into concrete daily instructional decisions3, 7, 8
12Document and communicate SDI clearly enough that another educator could implement it7, 8

How the unit is built

ModTitleTimeWhat it does
1What Is Specially Designed Instruction?75 minEstablishes the legal and instructional definition and the two-condition test
2SDI Is Not an Accommodation90 minSeparates SDI from six neighbours it is routinely confused with; includes Is This Actually SDI?
3From Disability-Related Need to Instruction90 minTeaches barrier analysis and IEP reading as the front half of the chain
4High-Leverage Practices for SDI75 minMaps the 2024 HLP framework onto the planning sequence
5Designing and Delivering SDI110 minExplicit instruction, modeling, scaffolding, feedback, engagement, gradual release
6Data-Based Individualization100 minProgress monitoring, decision rules, and the taxonomy of intensification
7SDI in Secondary Classrooms100 minApplies everything to labs, CER writing, primary sources, multi-step math, long-term projects
8SDI Design Challenge90 minFull case analysis, plan construction, and peer critique

Total: roughly 12 hours of module work plus 2–3 hours for the assessments and performance task. Facilitated delivery pacing options are in Section 17.

The seven students

Every module returns to the same seven fictional students. They are not illustrations of disability categories; they are illustrations of the fact that a category tells you almost nothing about what to teach. Two of them are classified with a learning disability and need entirely different SDI. Two are classified with autism and need entirely different SDI. That is the point.

Marisol Reyes

Grade 10 · Living Environment (Regents) · Learning disability
Strengths
Strong biological reasoning. In lab discussion she predicts outcomes accurately and explains mechanisms in detail. Scored proficient on the oral component of a class-wide ecology assessment.
Interests
Works weekends at a veterinary clinic; wants to be a vet tech.
Grade-level expectation
Construct a written explanation using the claim–evidence–reasoning framework, citing lab data.
Present levels
Reads grade-level text with adequate accuracy and comprehension. On written CER tasks she produces a claim, then stops, or restates the claim as evidence. Written products average 41 words against a class median of 190. Same content, spoken aloud, averages 6 accurate reasoning statements.
Disability-related need
Written expression: she cannot hold and sequence a multi-part argument in writing. This is a production barrier, not a knowledge barrier.
IEP
Annual goal: given a lab data set and a strategy prompt, Marisol will write a CER response scoring 3 of 4 on the department rubric in 4 of 5 trials. Program modification: strategy instruction in written expression. Accommodation: extended time (1.5x), word processor.
Assessment data
CER rubric baseline 1.4/4 across five trials. Written expression subtest, 6th percentile. Reading comprehension, 44th percentile.
Observation
Begins writing immediately, writes one sentence, then erases and restarts. Asks “is this right?” four to six times per task.
Current supports
Extended time, a word processor, and the class CER graphic organizer used by all students.
Response to date
No change over eight weeks. Extended time produces longer periods of not writing.

Amara Okafor

Grade 9 · Algebra I (Regents) · Learning disability
Strengths
Fast and accurate with integer operations and fluent with the distributive property. Strong verbal reasoning; explains her thinking readily.
Interests
Marching band drumline; interested in audio engineering.
Grade-level expectation
Solve multi-step linear equations in one variable, including equations with variables on both sides.
Present levels
Solves one-step and two-step equations at 85% accuracy. On equations with variables on both sides, accuracy drops to 22%. Error analysis of 40 problems shows the failures are not computational: she applies inverse operations to one side only, or performs operations in a self-invented order, in 31 of 31 incorrect items.
Disability-related need
Mathematics problem solving: the concept of the equation as a balanced relationship has not been established, so procedures are applied as unconnected moves.
IEP
Annual goal: given multi-step linear equations, Amara will solve with 80% accuracy across 3 consecutive probes, showing a justification for each step. Program modification: instruction in mathematics using concrete and representational models.
Assessment data
Algebra readiness screener, 12th percentile. Math computation, 61st percentile. Math problem solving, 9th percentile.
Observation
Works quickly and confidently, including when wrong. Does not check answers by substitution.
Current supports
A formula reference sheet, extended time, and a worked-example handout given to the whole class.
Response to date
Accuracy on the weekly equation probe: 20%, 25%, 20%, 30%, 20% over five weeks. Flat.

Devon Whitaker

Grade 11 · U.S. History & Government · Other health impairment (ADHD)
Strengths
Extensive background knowledge in twentieth-century history, much of it self-taught. Contributes sophisticated points in discussion. Strong document analysis when the document is in front of him.
Interests
Military history, podcasts, coaches a middle-school basketball team.
Grade-level expectation
Synthesize information from lecture, text, and primary sources into an evidence-based argument essay.
Present levels
Document-based writing scores 4/6 when he has accurate notes. His own lecture notes capture 3–4 of roughly 15 key ideas per class, and the recorded items are frequently anecdotal detail rather than the argument. Comparison of his notes to a criterion set: 24% of essential content captured.
Disability-related need
Sustained attention and working memory during extended oral input; he loses the thread while writing and cannot recover the structure of what was said.
IEP
Annual goal: given a 20-minute lecture, Devon will produce notes capturing 80% of criterion content across 4 of 5 trials, using a taught note-taking routine. Program modification: instruction in note-taking and self-monitoring strategies. Accommodation: copy of teacher notes available after the lesson.
Assessment data
Working memory index, 8th percentile. Verbal comprehension, 78th percentile. Note-capture criterion probes: 24%, 19%, 31%.
Observation
Writes continuously for the first four minutes of a lecture and then stops writing entirely. Looks up and re-engages when a visual appears.
Current supports
Preferential seating and teacher-provided notes emailed after class.
Response to date
Essay scores unchanged. He does not open the emailed notes; log shows 2 opens in 11 weeks.

Tyler Brennan

Grade 12 · Environmental Science · Autism
Strengths
Exceptional retention of technical vocabulary and quantitative detail. Can recall specific figures from readings weeks later. Precise, careful lab technique.
Interests
Hydrology and watershed data; maintains his own spreadsheet of local stream gauge readings.
Grade-level expectation
Interpret and evaluate scientific graphs, including graph types not previously encountered, and use them as evidence in argument.
Present levels
On line graphs and bar graphs he has seen before, accuracy is 92%. On an unfamiliar graph type — a climograph, a box plot, a stacked area chart — accuracy falls to 30%, and he applies the procedure for a familiar graph type instead. He does not generate the question “what is this axis measuring?” unprompted.
Disability-related need
Generalization and cognitive flexibility: strategies learned in one form are not transferred to novel forms without explicit instruction to do so.
IEP
Annual goal: given an unfamiliar graph, Tyler will apply a taught graph-interrogation routine and answer interpretation questions with 80% accuracy across 4 of 5 novel graph types. Program modification: instruction in generalization of cognitive strategies across novel formats.
Assessment data
Novel-graph probes: 30%, 25%, 40%. Familiar-graph probes: 90%, 95%.
Observation
Begins answering immediately without reading axis labels. Becomes visibly frustrated when told the answer is wrong but cannot locate the error.
Current supports
Extended time, a printed vocabulary glossary, and a graph-reading anchor chart posted for the class.
Response to date
Familiar-graph performance excellent and stable. Novel-graph performance flat over the marking period.

Jaden Cole

Grade 11 · Chemistry (Regents) · Autism
Strengths
Strong stoichiometric reasoning and mental arithmetic. Writes clearly and concisely. Reads well above grade level.
Interests
Competitive chess; builds mechanical keyboards.
Grade-level expectation
Plan and carry out a multi-day investigation and produce a formal lab report with procedure, data, analysis, and conclusion.
Present levels
Single-session lab work is strong (88% average). Multi-day investigations are not submitted: 1 of 6 completed this year. He can state every step of the report if asked orally, but does not initiate the first step, and when a procedure changes mid-investigation he stops working for the remainder of the period.
Disability-related need
Executive function: task initiation, decomposition of multi-session work, and recovery after an unplanned change.
IEP
Annual goal: given a multi-day investigation, Jaden will independently decompose it into dated steps and complete 4 of 5 investigations on time using a taught planning routine. Program modification: instruction in task decomposition and self-monitoring. Accommodation: advance notice of changes to procedure.
Assessment data
Executive function rating scale, clinically significant for initiation and planning. Lab report completion: 17%.
Observation
Sits with the packet open and blank for 10–12 minutes at the start of a project day. Once the first step is underway, works productively for the remainder of the period.
Current supports
Extended time, a checklist appended to the project packet, teacher check-ins.
Response to date
Check-ins produce work in that moment; nothing carries to the next day.

Nia Ferguson

Grade 10 · English 10 · Speech or language impairment
Strengths
Decodes fluently and accurately at grade level. Excellent memory for narrative. Draws sharp inferences about character motivation in discussion of familiar texts.
Interests
Theatre tech crew; wants to study stage management.
Grade-level expectation
Analyze how an author's choices about sentence structure and text organization shape meaning in complex informational text; participate in text-based academic discussion.
Present levels
Comprehension of simple and compound sentences is at grade level. On sentences with embedded relative clauses, appositives, or fronted subordinate clauses, comprehension accuracy drops to 35%, and she typically identifies the nearest noun as the subject. In discussion she contributes when the move is “say what you think,” and is silent when the move is “respond to what your classmate said and cite the text.”
Disability-related need
Receptive and expressive syntax: complex sentence structures are not parsed reliably, in reading or in discussion.
IEP
Annual goal: given complex sentences from grade-level informational text, Nia will identify the main clause and paraphrase accurately in 8 of 10 items across 3 consecutive probes. Related service: speech-language therapy 2x30 direct, 1x30 indirect. Program modification: instruction in syntactic analysis embedded in ELA content.
Assessment data
Listening comprehension of complex syntax, 5th percentile. Oral reading fluency, 62nd percentile. Vocabulary, 40th percentile.
Observation
Rereads long sentences three or four times, then moves on. Marks a text as “confusing” at exactly the sentences with embedded clauses.
Current supports
Text-to-speech, a vocabulary list, sentence starters posted for the whole class.
Response to date
Text-to-speech has no effect on comprehension of complex sentences, which is consistent with the barrier being syntactic rather than a decoding problem.

Cassidy Nolan

Grade 11 · Anatomy & Physiology · Emotional disability
Strengths
Near grade level in all academic areas; above grade level in science content knowledge. Reads independently, retains detail, asks precise questions one-to-one.
Interests
Wants to be a paramedic; volunteers with a fire department Explorer post.
Grade-level expectation
Complete a dissection lab practical and defend identifications orally to a partner and to the teacher.
Present levels
Written work when submitted averages 87%. But she leaves the room or puts her head down at the start of any graded, observed, or timed performance. Antecedent data across 22 incidents: 20 occurred within 90 seconds of an evaluative task being announced. Work completion on non-evaluative tasks, 91%; on evaluative tasks, 34%.
Disability-related need
Self-regulation under evaluative demand. The behavior is escape-maintained and is a barrier to demonstrating knowledge she already has.
IEP
Annual goal: using a taught self-regulation routine, Cassidy will begin and remain engaged in evaluative tasks for the full period in 4 of 5 opportunities. Behavior intervention plan in place. Program modification: instruction in self-monitoring and coping strategies embedded in content classes.
Assessment data
FBA concluded escape from evaluative demand. Engagement in evaluative tasks: 34%.
Observation
The exit occurs before the task starts, not during it. When she stays past the first two minutes, she completes the task.
Current supports
Break pass, preferential seating, reduced number of items on tests.
Response to date
Break pass use is high and rising; engagement unchanged. Reducing the number of items has not changed initiation.
Read these two pairs together

Marisol and Amara share a classification. One needs strategy instruction in written argument; the other needs conceptual instruction in the meaning of an equation. Tyler and Jaden share a classification. One needs explicit instruction in generalizing a strategy to novel formats; the other needs explicit instruction in decomposing multi-day work. The classification predicted none of it. The present levels and the error patterns did.

Section 02

Research and evidence base

What this unit rests on, where the line falls between legal requirement and research recommendation, and where the evidence is thinner than the confident tone of the field suggests.

The legal floor

Two things are law. Everything else in this unit is professional judgment informed by research.

Federal · IDEA

34 CFR §300.39(a)(1): “Special education means specially designed instruction, at no cost to the parents, to meet the unique needs of a child with a disability…”

§300.39(b)(3): Specially designed instruction means “adapting, as appropriate to the needs of an eligible child under this part, the content, methodology, or delivery of instruction — (i) To address the unique needs of the child that result from the child's disability; and (ii) To ensure access of the child to the general curriculum, so that the child can meet the educational standards within the jurisdiction of the public agency that apply to all children.”

New York · 8 NYCRR

§200.1(vv): Specially designed instruction means “adapting, as appropriate to the needs of an eligible student under this Part, the content, methodology, or delivery of instruction to address the unique needs that result from the student's disability; and to ensure access of the student to the general curriculum, so that he or she can meet the educational standards that apply to all students.”

§200.1(ww) defines special education as “specially designed individualized or group instruction or special services or programs… provided at no cost to the parent.”

The New York and federal definitions are substantively the same. New York's phrasing is what you will hear in a CSE meeting in Albany or Buffalo, so it is the phrasing this unit uses. The important thing about both is the sentence structure: a single verb (adapting), three objects (content, methodology, or delivery), and two purpose clauses joined by and. Both purposes must hold. That is the whole test, and Module 1 is built on it.

The third legal anchor is the standard of progress. In Endrew F. v. Douglas County School District RE-1 (2017), the Supreme Court held unanimously that an IEP must be reasonably calculated to enable a child to make progress appropriate in light of the child's circumstances — rejecting the argument that merely more than trivial progress suffices. That case is why this unit refuses to treat “the student is in a special class” as an answer to “what should the student learn?”

New York's own framing

NYSED's Blueprint for Improved Results for Students with Disabilities sets out seven principles; three of them govern this unit directly.

#Blueprint principleWhere it lands here
1Students engage in self-advocacy and are involved in determining their own educational goals and plan.Modules 5 and 6 — self-monitoring and fading as design goals, not afterthoughts
3Teachers design, provide, and assess the effectiveness of specially designed instruction to provide students with disabilities with access to participate and progress in the general education curriculum.The spine of the entire unit
4Teachers provide research-based instructional teaching and learning strategies and supports for students with disabilities.Modules 4 and 5

Two structural facts about New York matter for how you will look for SDI in a real IEP.

  • There is no box on the New York IEP labelled “specially designed instruction.” SDI is distributed across the present levels of performance, the measurable annual goals, the recommended special education program, and the supplementary aids and services / program modifications and accommodations. Reading the IEP as the SDI plan is a skill; Module 3 teaches it.
  • New York's continuum is defined by service, not by label. Under §200.6, integrated co-teaching classes provide “specially designed instruction and academic instruction” to a group of students with and without disabilities, capped at 12 students with disabilities; consultant teacher services run a minimum of two hours weekly; resource room a minimum of three, with a maximum instructional group of five. Those numbers describe the container. They do not describe the instruction, and a compliant service delivery schedule with no designed instruction inside it is the failure mode this unit exists to prevent.

The instructional evidence base

SourceWhat this unit takes from itCurrency
High-Leverage Practices for Students with Disabilities, 2nd ed. (CEC & CEEDAR)The four-domain structure and six pillar practices; verbatim HLP descriptions used in Module 4Mar 2024
WVDE Specially Designed Instruction Guidance DocumentThe content / methodology / delivery breakdown with planning questions; myth-busting; the guiding-questions set adapted in Module 2Aug 2025
NCII Data-Based Individualization and the Taxonomy of Intervention IntensityThe five-step DBI cycle and the seven intensification dimensions in Module 6Ongoing
IRIS Center modules and case studiesExplicit instruction, error analysis, self-regulation, DBI — assigned as pre-reading in Modules 5 and 6Ongoing
NYSED Part 200, Blueprint, standards-based IEP guidanceAll regulatory statements, terminology, and the standard-to-goal process in Module 3Current
Riccomini, Morano & Hughes (2017), TEACHING Exceptional ChildrenThe argument that SDI, HLPs, explicit instruction and intensive instruction are one connected system rather than four initiatives2017

Where the evidence is strong, and where it is not

Well established

  • Explicit instruction for students with disabilities in reading and mathematics. Decades of converging evidence; one of the most consistently supported findings in the field.
  • Progress monitoring with validated measures to detect non-response, and the practice of changing instruction when the data flatten.
  • Cognitive and metacognitive strategy instruction for written expression and reading comprehension at secondary level.
  • Corrective feedback delivered immediately and tied to a goal.

Emerging, contested, or overclaimed

  • Universal Design for Learning as an outcome intervention. UDL is a coherent design framework with a strong theoretical base; the evidence that implementing UDL raises achievement for students with disabilities is thinner than its adoption rate suggests. This unit treats UDL as good first design, not as SDI, and says so on evidentiary as well as legal grounds.
  • Co-teaching. Effects on student outcomes are inconsistent across studies. What predicts benefit appears to be what happens instructionally inside the co-taught room, not the staffing arrangement. Module 7 plans for the instruction, not the model.
  • Learning styles matching. Not supported. If a strategy in a plan is justified by “she's a visual learner,” that is not an evidence-based rationale, and Module 3 replaces it with barrier analysis.
  • Effect sizes quoted in PD materials. Frequently drawn from meta-analyses with heterogeneous designs. Use them to rank plausibility, not to predict what a particular student will gain.
A note on state guidance

This unit draws heavily on West Virginia's SDI guidance and refers to Kentucky's and Maryland's, because their published materials are unusually clear. None of it is New York requirement. Where another state's framing conflicts with NYSED terminology or Part 200, New York governs, and the text flags the difference rather than smoothing it over. WVDE's language that SDI is “the total sum of the IEP,” for example, is a useful teaching frame but is not how New York's regulation is worded, and this unit uses the New York wording.

Section 03

Pre-assessment

Fifteen items, mostly scenarios. Take it before Module 1. Do not look at the key. The point is to find out what you will be watching for, not to score well.

15 items~25 minutesKey in Section 15

Items 1–9 are multiple choice. Items 10–12 are matching or sorting. Items 13–15 are short response; write two to four sentences each. The post-assessment in Section 14 is parallel in structure and difficulty, so the two are directly comparable.

Part A · Definition and distinctions

1. Under both IDEA and New York regulation, specially designed instruction is defined as adapting the content, methodology, or delivery of instruction for two stated purposes. Which pair states them correctly?

  1. To remediate skill deficits, and to prepare the student for the next grade level
  2. To address needs that result from the student's disability, and to ensure access to the general curriculum so the student can meet the standards that apply to all students
  3. To provide accommodations listed on the IEP, and to document service minutes
  4. To individualize the pace of the curriculum, and to reduce frustration

2. A tenth-grade global history teacher gives every student in the class a blank Cornell notes template at the start of each unit and reviews how to use it in September. For a student with an IEP in that class, this practice is:

  1. SDI, because the student has an IEP and is receiving a support
  2. SDI, because note-taking is a documented area of need for many students with disabilities
  3. Not SDI as described, because it is provided identically to all students and is not tied to an identified disability-related need for that student
  4. Not SDI, because graphic organizers can never be SDI

3. Which statement about accommodations is accurate?

  1. An accommodation changes what the student is expected to learn
  2. An accommodation changes how a student accesses or demonstrates learning without changing the standard or the expectation
  3. Providing an accommodation is by itself the delivery of specially designed instruction
  4. Accommodations are only permitted on state assessments

4. A ninth grader with a learning disability in mathematics is failing multi-step equations. Error analysis shows she applies inverse operations to only one side of the equation. Which instructional decision is best supported by that specific evidence?

  1. Reduce her problem set from 20 items to 10
  2. Provide a formula reference sheet and extended time
  3. Explicitly teach the equation as a balance using a concrete model, then representational, then abstract, with a self-questioning step at each move
  4. Move her to a lower-level mathematics course

5. In the 2024 second edition of the High-Leverage Practices, which four domains organize the 22 practices?

  1. Assessment; Instruction; Social/Emotional/Behavioral; Collaboration
  2. Collaboration; Data-Driven Planning; Instruction in Behavior and Academics; Intensify and Intervene as Needed
  3. Screening; Diagnosis; Intervention; Evaluation
  4. Tier 1; Tier 2; Tier 3; Special Education

6. A student's progress-monitoring graph shows four consecutive data points below the goal line, with a flat trend over six weeks. The intervention has been delivered with fidelity. The most defensible next step is to:

  1. Continue as planned, because six weeks is too short to judge
  2. Lower the goal so the trend line meets it
  3. Conduct diagnostic assessment to identify why the student is not responding, then adapt the intervention on a specific dimension
  4. Add a second adult to the classroom

7. A twelfth grader with autism scores 92% on graph interpretation questions using graph types he has practiced and 30% on graph types he has not seen. His IEP identifies difficulty generalizing strategies to novel formats. Which is SDI?

  1. Giving him only the graph types he has already mastered
  2. Providing extended time on graph interpretation items
  3. Explicitly teaching a graph-interrogation routine, then deliberately practicing it across deliberately unfamiliar graph types with feedback until he applies it unprompted
  4. Seating him near the projector

8. Which statement most accurately describes the relationship between Universal Design for Learning and specially designed instruction?

  1. UDL is SDI delivered to a whole class
  2. If a teacher uses UDL well, SDI is not necessary
  3. UDL reduces predictable barriers in the design of instruction for everyone; SDI addresses a particular student's disability-related need and is individually determined. Good UDL can reduce how much SDI is needed but does not replace it
  4. UDL applies to general education and SDI applies to special education, so the two never appear in the same lesson

9. In New York State, where is specially designed instruction documented on the IEP?

  1. In a dedicated section labelled “Specially Designed Instruction”
  2. Only in the testing accommodations section
  3. Across multiple sections — present levels of performance, measurable annual goals, the recommended special education program and services, and supplementary aids and services / program modifications
  4. It is not documented on the IEP; it is documented in the teacher's plan book only

Part B · Sorting and matching

10. Sort each practice into the column where it belongs as described. Some descriptions are deliberately incomplete; if a description does not give you enough to decide, mark it Insufficient information and write what you would need to know.

 As described in the classroomSDI / Accommodation / Modification / Not enough information
aThe whole class gets a word bank on the unit test.
bA student receives 1.5x time on all assessments per her IEP.
cA teacher teaches one student a four-step routine for interrogating an unfamiliar graph, models it, practices it with feedback across novel graph types, and charts weekly accuracy against his IEP goal.
dA student answers 10 of the 25 test questions.
eA student's assigned reading is replaced with a text written three grade levels below, per an IEP team decision.
fA teacher pulls four students to a back table during independent work to review the same worksheet the class is doing.

11. Match each 2024 high-leverage practice to the planning question it answers.

HLPPlanning question
HLP 6 — Use student assessment data, analyze instructional practices, and make necessary adjustments___ What is interfering with access?
HLP 11 — Identify and prioritize long- and short-term learning goals___ Is it working?
HLP 16 — Use explicit instruction___ What should the student learn?
HLP 20 — Provide intensive instruction for academics and behavior___ What will I do differently?
HLP 4 — Use multiple sources of information to develop a comprehensive understanding of a student's strengths and needs___ What happens if the student still does not respond?

12. Put the seven planning steps in order by writing 1–7. Instructional adjustment · Disability-related barrier · Grade-level standard · Specially designed instruction · Student response and data · Learning target · IEP need

Part C · Short response

13. An eleventh grader with ADHD captures 24% of criterion content in his lecture notes. His current support is a copy of the teacher's notes emailed after class; the log shows he has opened them twice in eleven weeks. Explain in two to four sentences why the current support is not SDI, and describe what SDI for this need would involve.

14. A colleague says: “All of my students benefit from explicit vocabulary instruction, so I'm already providing SDI to the students with IEPs in my class.” Respond. Identify what is right in the statement and what is wrong with it.

15. A student in a 15:1 special class is enrolled in Regents Living Environment. Her teacher has been giving her a simplified summary instead of the assigned text because “she's in a special class.” Using the definition of SDI and what you know about grade-level access, explain what is wrong with this reasoning and what should happen instead.

Before you start Module 1

Write down the two items you were least sure about. At the end of the unit you will take a parallel post-assessment; come back to this note first.

Module 01

What Is Specially Designed Instruction?

The definition is one sentence long and almost nobody uses it correctly. This module takes it apart.

75 minutesObjectives 1, 2HLP 11, 12
Essential question

If two teachers were doing exactly the same thing in two different classrooms, how could one of them be delivering SDI and the other one not?

Learning objectives
  • State the IDEA and NYSED definitions of SDI and identify the two purpose conditions that both contain.
  • Classify an instructional adaptation as a change to content, methodology, or delivery.
  • Explain why SDI is a property of a decision rather than a property of a strategy.
  • Locate where SDI lives in a New York IEP.
Key vocabulary
TermWorking definition
Specially designed instructionAdapting the content, methodology, or delivery of instruction to address needs resulting from the disability and to secure access to the general curriculum.
ContentWhat is taught — the knowledge and skills. Includes prerequisite skills taught alongside grade-level content.
MethodologyHow it is taught — the instructional approach, sequence, strategy, or program.
DeliveryBy whom, where, when, and in what grouping instruction is provided.
Disability-related needThe specific access or performance barrier arising from the disability, stated in observable instructional terms — not the classification.
General curriculumThe same curriculum, based on the same learning standards, that applies to all students in the district.
Present levels of performanceThe IEP section describing current academic achievement and functional performance; the baseline that every SDI decision is measured against.
Before you go on

The definition, taken apart

Here is the New York text with the load-bearing parts marked:

Adapting, as appropriate to the needs of an eligible student under this Part, the content, methodology, or delivery of instruction to address the unique needs that result from the student's disability; and to ensure access of the student to the general curriculum, so that he or she can meet the educational standards that apply to all students. 8 NYCRR §200.1(vv)

Three things follow from that sentence, and each one overturns something teachers commonly believe.

1. SDI is a verb applied to instruction, not a category of student support

The subject of the sentence is adapting. Not providing, not listing, not offering. Something has to change about the instruction itself. A support that sits beside instruction — a device, a seat, a clock — is not what this sentence describes. It may be necessary, it may be required by the IEP, and withholding it may be a violation. It is still not, by itself, specially designed instruction.

This is the distinction that makes teachers uncomfortable, so be precise about what is being claimed. Nobody is saying accommodations are unimportant. The claim is narrower: an accommodation removes an obstacle so the student can do the thing. SDI teaches the student to do the thing. Extended time gives Marisol more minutes in which to not write. Explicit strategy instruction in how to structure a written argument teaches her to write.

2. There are exactly three things you can adapt

Content, methodology, delivery. Everything else is one of those three wearing different clothes. Getting fluent at naming which one you are changing is the single most useful habit in this unit, because it forces you to say what is actually different about Tuesday.

ElementThe question it answersWhat changing it looks like at secondary level
ContentWhat is taught?Teaching the prerequisite concept of equality alongside the grade-level equation-solving standard. Pre-teaching the six terms that carry the argument in a primary source. Adding a strategy — a note-taking routine, a graph-interrogation routine — as content in its own right, taught to mastery.
MethodologyHow is it taught?Concrete–representational–abstract sequence instead of a worked example. Explicit modeling with think-aloud instead of discovery. Self-regulated strategy development with memorized mnemonic instead of a handout. Systematic prompt fading on a planned schedule.
DeliveryBy whom, where, when, how grouped?Ten minutes of pre-teaching with the special educator before the lab. Practice distributed across three settings so the strategy generalizes. A group of three with the same error pattern rather than a group of ten with the same schedule. The SLP delivering syntactic instruction inside the ELA text rather than in a pull-out room.

Structure adapted from the WVDE Specially Designed Instruction Guidance Document (Aug 2025, p. 2), which credits the PROGRESS Center at AIR.

3. Both purpose clauses have to hold at once

This is the part that gets dropped. The regulation is not “adapting instruction to address disability-related needs.” It is that, and to ensure access to the general curriculum so the student can meet the standards that apply to all students. An adaptation that addresses a disability-related need while removing the student from grade-level standards has satisfied half the definition and violated the other half.

The two-condition test

For any instructional decision, ask both questions. Both must be yes.

  1. Is this adaptation targeted at a need that results from this student's disability, identified in the present levels, and would you not be doing it for this student if that need were not there?
  2. Does this adaptation move the student toward the grade-level standard that applies to all students, rather than away from it?

Why the same practice can be SDI in one room and not the next

Return to the essential question. Two chemistry teachers each hand a student a partially completed lab report scaffold. Same paper. In one room it is SDI and in the other it is not, and the difference is invisible if you are only looking at the paper.

SDI

The scaffold was built for Jaden after his teacher analyzed six unsubmitted multi-day investigations and found that failure occurs at initiation, not execution. It contains dated decomposition steps, not content hints. His teacher explicitly taught the decomposition routine over four lessons, modeled it aloud, faded from “I fill in the dates” to “you fill in the dates” on a written schedule, and charts on-time step completion weekly against the IEP goal. In eight weeks the scaffold will be a blank index card.

Not SDI

The scaffold was distributed to the whole class because the last set of lab reports was disappointing. It is a good scaffold and it will help several students. Nothing about it was determined by Jaden's needs, nobody taught him how to use it, no data are tied to it, and there is no plan for removing it. This is competent general instruction. It is not specially designed.

Notice that the classroom observation would look nearly identical. The difference lives in the reasoning behind the paper and in the data behind the reasoning. That is why this unit teaches a chain rather than a checklist — and why documentation matters, which is Module 7.

Where SDI lives in a New York IEP

There is no SDI field. Reading a New York IEP for SDI means assembling it from four places.

  1. Present levels of academic achievement and functional performanceThe barrier is described here, or it is nowhere. If the present levels say “Marisol struggles with writing,” you cannot design anything. If they say “produces a claim and stops; 41 words against a class median of 190; the same content spoken aloud contains six accurate reasoning statements,” the instruction almost designs itself.
  2. Measurable annual goalsThe goal names what the SDI is for. A goal that specifies a taught routine (“using a taught note-taking routine”) is telling you methodology is being adapted. A goal aligned to the grade-level standard is telling you the second purpose clause is satisfied.
  3. Recommended special education program and servicesConsultant teacher, resource room, integrated co-teaching, special class, related services — this is the delivery element: by whom, where, when, how often, in what group size. Under §200.6 these carry specific minimums and caps.
  4. Supplementary aids and services, program modifications, accommodations, supports for school personnelAccommodations live here. So do program modifications, which are where you find authorized changes to content. And “supports for school personnel” is the sleeper field: it is where the training a paraprofessional needs to implement the SDI correctly gets written down.
New York detail

Under §200.4(d)(2)(ix), transition planning in New York begins with the first IEP in effect when the student is age 15, a year earlier than the federal floor of 16. For a secondary teacher this is not paperwork trivia. It means that from ninth or tenth grade onward, every SDI decision should be answering a postsecondary goal too: the routine you teach Devon for taking notes is a routine he will use in a community college lecture hall where nobody emails him anything.

Interactive activity · Element sort with justification

Time: 20 minutes. Format: individual, then pairs.

For each decision below, do three things: (1) name the element being adapted — content, methodology, or delivery, and it may be more than one; (2) apply the two-condition test; (3) write the one piece of information you would need from the IEP or the data to be confident in your answer.

 DecisionElement(s)Both conditions?What I'd need to know
aBefore the Living Environment unit on homeostasis, the special educator teaches Marisol the six terms that carry the argument in the assigned reading, using examples and non-examples, for ten minutes twice a week.
bThe chemistry teacher posts the lab safety rules in large print at the front of the room.
cAmara's teacher replaces the worked-example handout with a concrete balance-scale model, then drawings, then symbols, over three lessons, with Amara stating the justification for each move.
dNia's speech-language pathologist delivers her syntax instruction inside the English 10 text the class is reading that week rather than on decontextualized sentence cards.
eCassidy is given a pass to leave the room whenever she needs a break.
fTyler's teacher deliberately assigns three graph types he has never seen and coaches him through the interrogation routine on each, then removes the coaching on the fourth.

Debrief prompt: Which of these could flip from “not SDI” to “SDI” with a change in how, not what? Item (e) is the one to argue about.

Knowledge check

Click a question to reveal the answer and explanation.

1.1The federal and New York definitions of SDI both contain two purpose clauses joined by “and.” What goes wrong instructionally when only the first one is honored?
Short response — look for

The student's disability-related need gets addressed while the student drifts away from the grade-level standard. In practice this looks like a decade of instruction on prerequisite skills that never reconnects to the curriculum: the student gets better at something, and further behind at everything the diploma requires. Both clauses are conjunctive, so an adaptation that serves the need but abandons access does not meet the definition — and under Endrew F. a program has to be reasonably calculated to enable progress appropriate in light of the student's circumstances, which for a student pursuing a Regents diploma means progress in the general curriculum.

1.2Which of the following is the clearest example of adapting delivery?
  1. Teaching a self-questioning routine using explicit modeling
  2. Adding instruction in the concept of equality to a lesson on multi-step equations
  3. Providing ten minutes of pre-teaching with the special educator immediately before the general education lesson, in a group of three students with the same error pattern
  4. Allowing a student to type instead of handwrite
Answer · C

Delivery answers by whom, where, when, and in what grouping. Option C changes the person, the timing, and the group. A is methodology (how it is taught). B is content (what is taught). D is an accommodation — it changes the mode of response without changing the instruction, and the expectation is unchanged.

1.3Scenario. A U.S. History teacher emails his lecture notes to Devon after every class, as the IEP accommodation specifies. Devon has opened them twice in eleven weeks. Is the requirement being met? Is SDI happening?
Answer

The accommodation is being provided — the teacher is doing what the IEP says, and that obligation is met. SDI is not happening for this need. Nothing about the instruction has been adapted to address Devon's difficulty sustaining attention and holding information during extended oral input. The notes substitute for a skill rather than build one, and the usage data show they are not even doing that. Two conclusions follow: the team should keep the accommodation if it has value as a safety net, and should design instruction — explicit teaching of a note-taking routine with cued pauses and self-monitoring — that targets the actual need. The usage log is also data, and it should go to the CSE.

1.4True or false, with justification: because Amara is in a 15:1 special class for Algebra I, her instruction should be based on a lower-level curriculum.
False

Placement is a delivery decision made under §200.6 based on the intensity of need and the least restrictive environment; it says nothing about which standards apply. Amara is enrolled in Algebra I, so the Algebra I standards apply. What her placement does tell you is that she needs instruction more intensive than what a larger group can deliver — which is an argument for concentrated, explicit, well-monitored instruction on the grade-level content plus the missing prerequisite, not for a different curriculum. Conflating “restrictive setting” with “lower expectations” is the single most consequential error in secondary special education.

1.5Matching. Connect each IEP section to the SDI element it most directly documents.

IEP sections: (i) Present levels of performance  (ii) Measurable annual goals  (iii) Recommended special education program and services  (iv) Program modifications

SDI functions: (w) Authorized changes to what is taught  (x) The baseline the adaptation is measured against  (y) By whom, where, when, and in what group  (z) What the adapted instruction is aiming at

Answer

(i)–(x), (ii)–(z), (iii)–(y), (iv)–(w).

The takeaway is that no single section holds SDI. If you read only the accommodations page — which is what most content teachers are handed — you have read the part of the IEP that contains the least information about how to teach the student.

Reflection and application

Pick one student on your caseload or in your section. Open the IEP and, without writing any plan yet, answer two questions in writing: What does the present levels section tell me that changes what I would do on Tuesday? and If it tells me nothing, what would I need it to say? Bring the second answer to your next CSE meeting.

Takeaway

SDI is not a strategy you can point at. It is an adaptation of content, methodology, or delivery that is caused by a specific disability-related need and aimed at a grade-level standard. Change either end and the same strategy stops being SDI.

Module 02

SDI Is Not an Accommodation

Six neighbours that get mistaken for SDI, and the sixteen practices that could be either.

90 minutesObjective 3HLP 13, 15
Essential question

If a practice helps a student, why isn't that enough to call it specially designed instruction?

Learning objectives
  • Distinguish SDI from accommodations, modifications, differentiation, UDL, intervention, and scaffolding using purpose and origin rather than appearance.
  • Explain why the same practice can be SDI in one case and not in another.
  • Apply a decision procedure to sixteen common secondary practices.
Key vocabulary
TermWorking definitionWho decides
AccommodationA change in how a student accesses instruction or demonstrates learning. The standard and the expectation are unchanged.IEP team
ModificationA change in what a student is expected to learn or demonstrate. The expectation itself changes.IEP team
Differentiated instructionAdjusting content, process, or product across a class in response to readiness, interest, or learning profile.Classroom teacher
Universal Design for LearningDesigning instruction from the outset with multiple means of engagement, representation, and action/expression to reduce predictable barriers for everyone.Designer of the lesson
InterventionSupplemental instruction targeting a specific skill deficit, usually delivered through a tiered system, available to any student who meets criteria.School team / data rules
ScaffoldTemporary support that lets a student do now what they cannot yet do alone, calibrated and then removed.Whoever is teaching
Specially designed instructionAdaptation of content, methodology, or delivery, caused by a disability-related need, aimed at grade-level access.IEP team; delivered by many
Before you go on
  • WVDE Additional SDI Resources — the Comparison of SDI, Differentiated Instruction, and Universal Design for Learning one-pager, and the EBP Spotlight: Scaffolding SDI.
  • WVDE SDI Guidance, pp. 6–7: “Difference Between SDI, Accommodations, and Modifications” and the guiding questions, and p. 21, FAQ 15 on small-group instruction.
  • IRIS Center: Universal Design for Learning — the Perspectives & Resources pages on the three principles.

The six neighbours

Each of these is a real, defensible practice. None of them is SDI by definition, and each is confused with SDI for a slightly different reason. Sorting them by who decides, on what basis, for whom, and what changes makes the boundaries visible.

 Triggered byWho receives itWhat changesWhy it gets confused with SDI
SDIA need resulting from a disability, documented in present levelsOne student, individually determinedContent, methodology, or delivery of instruction—
AccommodationSame disability-related needOne studentAccess or response conditions. Instruction unchanged.It is on the IEP, so it feels like special education. It is legally required and instructionally passive.
ModificationTeam decision that the expectation itself must changeOne studentWhat is expected — the standard or criterionIt changes content, and SDI can change content. The difference is that SDI changes content to reach the standard; a modification changes the standard.
DifferentiationRange of readiness, interest, or profile in the roomGroups of students, fluidlyContent, process, or product across a classIt uses the same three words. It is planned for a group and not caused by a disability.
UDLPredictable variability in any learner populationEveryone, by design, in advanceThe design of the lesson before students arriveGood UDL looks like a room full of supports. Nothing in it is individually determined.
InterventionScreening or performance data meeting a thresholdAny student who qualifies, with or without an IEPSupplemental instruction on a targeted skillIt is intensive, explicit, and data-driven — it looks exactly like well-run SDI. The difference is origin: eligibility criteria versus an IEP need.
ScaffoldThe gap between the task and current performanceAnyone, situationallyTemporary support inside a taskScaffolding is a component of SDI. Providing a scaffold is not, on its own, SDI.
The relationship, stated plainly

These are not competing options. A well-run secondary classroom has all of them at once: a UDL-designed lesson, differentiated within it, with scaffolds provided as needed, accommodations honored for students who have them, interventions running for students who qualify, and SDI delivered to specific students for specific documented needs. The error is not using them. The error is counting one of them as SDI and concluding the obligation is met.

The one that needs extra care: intervention

Intervention and SDI overlap more than any other pair, and the distinction is genuinely subtle. An eleventh-grade Tier 3 reading intervention and eleventh-grade SDI in reading can be the same program, the same length, the same explicitness, delivered by the same person.

What separates them is origin and obligation. Intervention is available to students who meet criteria and can be discontinued when they stop meeting them. SDI flows from an IEP need, and its provision is an entitlement that does not depend on continued qualification for a program. In practice, three things follow: SDI must be individually determined rather than assigned by protocol; SDI must be documented and monitored against the IEP goal; and SDI cannot be discontinued because a program's cohort ended.

WVDE's guidance makes a related point worth carrying: in a multi-tiered system, SDI is not synonymous with Tier 3. It is delivered in whatever tier the student's instruction occurs in.


Is This Actually SDI?

This is the centre of the unit. Sixteen practices that show up in every secondary building, each shown twice: once where it is part of SDI and once where it is not. Read the pairs. Do not read the practice names as a list of answers — every one of them appears on both sides.

SDI · Extended time

Rarely. Extended time changes the conditions of a task, not the instruction. It becomes part of SDI only when it is the delivery change that makes designed instruction possible — for example, Nia's schedule builds in the extra ten minutes she needs because her SLP-delivered syntactic instruction is embedded inside the same reading the class is doing, and the reading takes her longer because instruction is happening during it.

Not SDI · Extended time

The ordinary case. Marisol gets 1.5x on everything. She uses the additional time to sit in front of a blank document for longer. Nothing about the instruction changed; an obstacle was loosened. Keep it — and stop calling it her program.

SDI · Preferential seating

Almost never on its own. It can be a genuine delivery adaptation when the seat is instrumentally part of a taught routine: Devon sits within arm's reach of the document camera because his note-taking routine depends on a visual cue he has been taught to use as a self-monitoring prompt, and the cue is being systematically faded.

Not SDI · Preferential seating

The ordinary case, and the most over-listed accommodation in secondary education. “Seat near teacher” on an IEP with no instructional change behind it is a seating chart, not a program.

SDI · Graphic organizer

Marisol's CER organizer is not the class organizer. It has one box for the claim, three numbered evidence rows tied to specific data table columns, and a reasoning row with a taught sentence frame linking evidence to a biological principle. Her teacher modeled its use aloud on three lab data sets, practiced it with immediate corrective feedback, and is fading it: full organizer → headings only → blank paper with the routine recited from memory. Rubric scores are charted weekly against the IEP goal.

Not SDI · Graphic organizer

The class CER organizer, handed to Marisol along with everyone else, unmentioned and unexplained. It has been on her desk for eight weeks and her rubric average has not moved off 1.4. A tool nobody taught her to use is not instruction.

SDI · Small-group instruction

Amara works with two other students who share the same error pattern — operating on one side of the equation — on a concrete-to-abstract sequence built for that error, for fifteen minutes, three times a week, with weekly probes. The group was formed by error analysis and will dissolve when the error resolves.

Not SDI · Small-group instruction

Four students go to the back table during independent work and do the same worksheet as the class with an adult nearby. WVDE's guidance is direct about this: if the small-group lesson is not planned and not focused on an evidence-based strategy addressing an identified need, it does not meet the definition. Proximity is not methodology.

SDI · Breaking work into smaller pieces

Jaden is taught to decompose a multi-day investigation himself: a modeled routine, a written procedure, guided practice on three projects, then independent decomposition with the teacher checking only the product. The skill transfers to his English research paper. The chunking is the curriculum, not the accommodation.

Not SDI · Breaking work into smaller pieces

The teacher hands Jaden a packet already broken into steps with dates filled in. He completes it. Next project, same problem, because nobody taught him to do the breaking. A year of this produces a senior who cannot start anything alone, which is the opposite of what his IEP is for.

SDI · Teacher-provided notes

Devon receives a criterion note set after he has taken his own notes using the taught routine, and the two are compared as a self-monitoring step: he highlights what he captured, counts it, and graphs the percentage himself. The teacher notes are an instructional feedback device, and the comparison step is faded as capture rises.

Not SDI · Teacher-provided notes

Notes emailed after class, opened twice in eleven weeks. It is a legitimate accommodation and a complete instructional non-event. Worse, if it were used, it would remove the practice opportunity that his goal depends on.

SDI · Text-to-speech

A student whose disability-related need is decoding, not comprehension, is taught to use TTS strategically — when to slow the rate, when to re-listen to a sentence, how to pair listening with annotation — so that grade-level informational text becomes usable and the student can work on the grade-level analysis standard rather than on getting through the words.

Not SDI · Text-to-speech

Nia has TTS. Her barrier is syntactic, not decoding. Hearing an embedded relative clause read aloud does not parse it for her. The tool is provided, the data are flat, and the flat data are the diagnostic: when a support does not change performance, the barrier was misidentified.

SDI · Reduced number of questions

Only when the reduction is instructionally reasoned and time-limited: during the acquisition phase of Amara's balance routine she does six equations with full written justification instead of twenty bare problems, because the justification is the skill being taught. Item count rises as she moves from acquisition to fluency, and the assessment expectation does not change.

Not SDI · Reduced number of questions

Cassidy gets ten items instead of twenty-five on every test as a standing practice. Her barrier is initiation under evaluative demand, not stamina, and the data confirm it — she leaves in the first ninety seconds. Fewer items has not moved engagement at all. This is a modification of expectation dressed as a support, and it is aimed at the wrong thing.

SDI · Explicit vocabulary instruction

Six terms selected because they carry the argument in the primary source Devon must analyze, taught with examples and non-examples, with retrieval practice distributed across the week, because his present levels identify a working-memory barrier that prevents new terms from being available when he needs them mid-analysis. Retention is probed and charted.

Not SDI · Explicit vocabulary instruction

The class does a vocabulary preview every Monday. It is excellent teaching, it is well supported by evidence, and it is what every student in the room gets. Universal good practice does not become SDI by having a student with an IEP in the room.

SDI · Modeling

The teacher thinks aloud through the graph-interrogation routine on a climograph Tyler has never seen, deliberately narrating the step he skips — “before I look at the shape, I read what each axis measures” — because his error data show that exact omission. The model is repeated on novel graph types and then withdrawn on a planned schedule.

Not SDI · Modeling

The teacher works an example on the board before independent practice, as she does every day for everyone. Modeling is a component of explicit instruction and one of the best things a teacher does. It is not individually determined and it is not aimed at a documented need.

SDI · Repeated practice

Distributed retrieval practice scheduled specifically for Devon across three settings and two weeks, because his working memory profile means massed practice produces performance that disappears by Thursday. The schedule is a designed feature and the retention interval is measured.

Not SDI · Repeated practice

“More reps.” Giving a student additional problems of the same kind that produced the errors, with no change to how they are taught. Practice does not fix a misconception; it rehearses it. Amara's five flat weekly probes are what this looks like on a graph.

SDI · Sentence starters

Nia is taught a small set of discussion moves — “Building on what ___ said, the text shows ___, which suggests ___” — to mastery, with the syntactic structure explicitly analyzed, rehearsed, and then faded to a written prompt, then to nothing. The stems are the object of instruction, not a crutch beside it.

Not SDI · Sentence starters

Discussion stems are posted on the wall for the class. Nia has been looking at them for a semester and has not used one. A poster is not instruction, and a stem a student cannot construct is not a support.

SDI · Visual schedule

Jaden co-constructs a dated project timeline as part of a taught decomposition routine, and the construction — not the schedule — is what is being taught. He fills more of it each time; the teacher fills less. By the fourth project the schedule is his.

Not SDI · Visual schedule

A finished timeline is taped to the front of the packet by the teacher. It organizes the work. It does not teach organization, and at grade 11, with a postsecondary goal of independent enrollment, teaching organization is the point.

SDI · Guided notes

Devon's guided notes are structured around the argument of the lecture rather than its details, with deliberate blanks at exactly the points his criterion analysis shows he loses the thread, plus two scheduled cued pauses. Blanks are reduced by 25% every two weeks against his capture data.

Not SDI · Guided notes

The teacher distributes fill-in-the-blank notes to the whole class. Good practice for a lecture-heavy course. Universal, undesigned for anyone, and permanent.

SDI · One-to-one instruction

Fifteen minutes of individual instruction is the delivery element required for the acquisition phase of an explicitly taught strategy, and the plan specifies the criterion at which it moves to a pair, then to the whole-class setting. The one-to-one is a phase, with an exit condition.

Not SDI · One-to-one instruction

An adult sits beside a student and helps with whatever comes up. This is the most expensive thing a school does and among the least likely to change performance. Without designed content it produces prompt dependence, which the WVDE guidance names directly as the risk of over-prompting.

SDI · Frequent check-ins

Check-ins occur at a fixed, faded schedule as the reinforcement component of Cassidy's taught self-regulation routine — at 2 minutes, then 5, then 12 — and each one is a scheduled opportunity for her to self-rate and receive feedback on the routine, not on the content. The schedule thins against engagement data.

Not SDI · Frequent check-ins

“I check in with her a lot.” Responsive, kind, and unplanned. It generates work in the moment and nothing that survives to the next day — which is precisely what Cassidy's data show.

The pattern across all sixteen

Look back at the sixteen pairs. Every single “SDI” side has four features and every “not SDI” side is missing at least one: (1) it was chosen because of a specific documented need, (2) somebody explicitly taught the student to do it, (3) data are being collected against a goal, and (4) there is a plan for fading it. Provided-but-not-taught is the single most common failure, and it is what turns a good idea into a permanent prosthetic.

Interactive activity · Flip the card

Time: 25 minutes. Format: table groups of three, then whole group.

Each group takes four practices from the list of sixteen. For each one, do the reverse of what you just read:

  1. Write a two-sentence classroom description in which the practice is not SDI. Make it a plausible, well-intentioned description — not a straw man.
  2. Then change the minimum necessary to make it SDI for one of the seven students in Section 1. You may not change the practice itself.
  3. Name which of the four features you added, and name the element you adapted: content, methodology, or delivery.

Whole-group debrief: collect the changes on chart paper sorted by feature. In most groups, “somebody explicitly taught the student to use it” is the change that appears most often. Ask why that one is so easy to skip.

If you are working alone: do this for the four practices you are most likely to have used this month. Write them down before you decide which side they were on.

Knowledge check

2.1Which of these best captures the difference between an accommodation and specially designed instruction?
  1. Accommodations are for state tests; SDI is for the classroom
  2. Accommodations are provided by general educators; SDI is provided by special educators
  3. An accommodation changes access or response conditions while the instruction and the expectation stay the same; SDI changes the instruction itself in response to a disability-related need
  4. Accommodations are optional; SDI is required
Answer · C

B is a common and consequential misconception: under New York's continuum, general education teachers deliver SDI routinely, most obviously in integrated co-teaching, and paraprofessionals implement it under supervision. D is wrong in both halves — accommodations on an IEP are legally required, and the word “optional” has no place in either.

2.2Scenario. An English teacher designs a unit with three text formats, choice of three response products, and vocabulary supports built into every text, because she knows the range in her classes. Two students in the section have IEPs. Is this SDI?
Answer · No, this is UDL, and that is a compliment

The design was made in advance for predictable variability across the whole population, not in response to either student's documented need. It is likely to reduce how much additional adaptation those two students require, which is exactly what good universal design is supposed to do. It does not discharge the obligation to design instruction for their specific IEP needs, and the giveaway is that the plan would be identical if neither student were enrolled.

2.3Sort: SDI, accommodation, modification, or not enough information?
  1. A student uses speech-to-text for all extended writing.
  2. The IEP team determines a student will be assessed on identifying two characters rather than comparing and contrasting them.
  3. A teacher explicitly teaches a paraphrasing routine to two students with reading comprehension goals, practices it in social studies and science, and probes retention monthly.
  4. A student sits in the front row.
  5. A teacher gives a student a partially completed outline.
Answer

(a) Accommodation. Response mode changes; the writing standard does not. (b) Modification. The expectation itself is changed by team decision. (c) SDI. Need-driven, explicitly taught, cross-setting for generalization, monitored. (d) Accommodation at best, and probably nothing — there is no instructional change. (e) Not enough information, and this is the item that matters. You need to know: was it built for this student's identified need, was the student taught to complete it, is it tied to a goal, and is it being faded? Same paper, two verdicts.

2.4A colleague argues that a Tier 3 reading intervention delivered to a student with an IEP is automatically that student's SDI. Where is the argument wrong?
Short response — look for

It confuses form with origin. The intervention might be an excellent vehicle for SDI — and often is. But a student qualifies for intervention by meeting screening criteria, while SDI flows from an individually determined IEP need. Three practical consequences: the intervention must actually match this student's documented need rather than the average profile of the cohort; progress must be monitored against the IEP goal, not just the program's internal assessments; and the student's entitlement does not end when the program cycle ends. Also worth naming: SDI is not confined to Tier 3. It is delivered in whatever tier the student's instruction is happening in.

2.5Error analysis. A ninth-grade IEP lists these as the student's specially designed instruction. Identify what is actually in the list and what is missing.

Extended time (1.5x) · Preferential seating · Copy of notes · Graphic organizers as needed · Breaks as needed · Reduced homework · Check for understanding

Answer

Items 1, 2, 3, and 5 are accommodations. Item 6 is a modification of expectation, and it is worth asking whether the team actually decided that or whether it drifted in. Items 4 and 7 are undefined practices — “as needed” and “check for understanding” tell a substitute teacher nothing they could implement.

What is missing is all of it: no statement of what will be taught, no statement of how it will be taught, no statement of who teaches it, where, or how often, no connection to a goal, and no data plan. This list is not unusual. It is roughly what most content teachers in most secondary buildings receive, and it is why Module 7 spends its time on documentation.

Reflection and application

Take the accommodations page for one student you teach. For each item, write one sentence answering: what instruction, if any, sits behind this? Where the answer is “none,” you have found either a support that is fine as a support, or a need that nobody is teaching to. Decide which, and be honest about the ratio.

Takeaway

Nothing on the list of sixteen is SDI or not-SDI on its face. What decides it is whether the practice was chosen for a documented need, taught rather than merely provided, measured against a goal, and built to be removed.

Module 03

From Disability-Related Need to Instruction

Barrier analysis: the skill that makes the rest of the chain possible, and the one most often skipped.

90 minutesObjectives 2, 5, 11HLP 4, 5, 6, 11
Essential question

The student can't do the task. What exactly is stopping them — and how would I know if I were wrong?

Learning objectives
  • Move from a grade-level standard to a specific, observable learning target.
  • State a disability-related barrier in instructional terms rather than diagnostic ones.
  • Use error patterns and performance discrepancies to distinguish competing barrier hypotheses.
  • Read a New York IEP for the information that constrains instructional design.
Key vocabulary
TermWorking definition
Learning targetThe observable thing a student will do in this lesson or unit, derived from the standard. Standards are for the year; targets are for Tuesday.
Disability-related barrierThe specific mechanism by which the disability interferes with performing the target. Stated as a verb phrase about behavior, never as a classification.
Barrier hypothesisA testable claim about that mechanism, with a prediction attached: if this is the barrier, then performance should look like this and not that.
Performance discrepancyA gap between two conditions that isolates the barrier — oral versus written, familiar versus novel, evaluative versus non-evaluative.
Error patternThe systematic, repeated form of a student's mistakes. Random errors mean something different from consistent ones.
Point of accessNYSED's term for where a student's current skills sit in relation to the grade-level standard — the place instruction has to start from to get to the standard.
Before you go on

Step 1 and 2: standard, then target

Start at the standard, always. Not at the student's level, not at the IEP goal, not at last year's materials. NYSED's standards-based IEP process opens with “Know the grade-level content standards” for a reason: everything downstream is a comparison against that anchor, and if you start the comparison somewhere else, you get a plan for a curriculum nobody is enrolled in.

The standard is annual and abstract. The learning target is what a student does in front of you this week. Making that move well is a skill, and most vague SDI plans are downstream of a vague target.

Grade-level standardWeak targetUsable target
Construct a written explanation from evidence (Living Environment lab)“Improve science writing”Given a data table from the enzyme lab, write a CER paragraph with one claim, three data-based evidence statements, and reasoning linking evidence to enzyme function
Solve multi-step linear equations with variables on both sides“Work on algebra”Solve 3x + 7 = 5x − 9 and state a justification for each transformation
Analyze how sentence structure shapes meaning in informational text“Improve reading comprehension”Given a sentence with an embedded relative clause, identify the main clause and paraphrase the sentence accurately
Interpret and evaluate scientific graphs as evidence“Understand graphs”Given an unfamiliar graph type, state what each axis measures, describe the relationship, and identify one claim the graph does not support
A test for your target

If two teachers watched the same student for one period, could they independently agree on whether the target was met? If not, the target is not yet a target, and no SDI written against it will be implementable by anyone but you.

Step 3: the barrier

Here is where most plans go wrong, and the failure is almost always the same. The barrier is written as a restatement of the classification (“has a learning disability”), a restatement of the outcome (“struggles with writing”), or a folk theory (“is a visual learner,” “is unmotivated,” “has low processing”). None of these tells you what to do on Tuesday.

A usable barrier statement has three properties: it names a mechanism, it is observable, and it predicts where performance will and will not break down. That last property is what makes it testable, and testability is what saves you from a semester of teaching the wrong thing.

Not a barrier statementA barrier statementWhat it predicts
“Marisol has a writing disability”Cannot hold and sequence a multi-part argument in written production; the content is available orallyOral performance strong, written performance weak on the same content. If oral were also weak, this hypothesis is wrong.
“Amara is bad at algebra”Has not established the equation as a balanced relation, so inverse operations are applied as isolated movesOne-step problems fine, both-sides problems fail, and the errors are one-sided operations rather than arithmetic slips.
“Devon doesn't pay attention”Cannot sustain attention and hold structure across extended oral input; loses the argument while transcribing detailStrong when the source is in front of him, weak when it is spoken; note capture collapses after the first few minutes.
“Tyler is rigid”Does not transfer a learned interrogation strategy to novel formats without explicit generalization instructionNear-ceiling on familiar formats, floor on novel ones, with familiar-format procedures misapplied.
“Cassidy is avoidant”Escape from evaluative demand, occurring at task announcement rather than during the taskExits cluster in the first ninety seconds; completion is high on non-evaluative work. If exits occurred mid-task, the hypothesis would be fatigue, not evaluation.

Three tools for finding the mechanism

  1. Look for the discrepancyFind two conditions where the same student performs very differently. Oral versus written. Familiar versus novel. Untimed versus timed. Observed versus unobserved. With the text present versus from memory. The variable that changes across the two conditions is a candidate mechanism. Marisol's 41 words versus six accurate oral reasoning statements is the whole diagnosis in one comparison.
  2. Read the errors, not the scoreA score tells you how much. An error pattern tells you what. Thirty-one of Amara's thirty-one wrong answers share a form; that is not a fluency problem and no amount of extra practice will touch it. Collect 15–20 errors before you conclude anything, and sort them into factual, procedural, and conceptual.
  3. Check the antecedentsFor behavior that interferes with learning, when the behavior occurs is the diagnosis. Cassidy's twenty of twenty-two incidents within ninety seconds of an evaluative announcement rules out several hypotheses at once and rules in one.
Watch for this

A support that has been in place for weeks and produced no change is evidence about the barrier, not just evidence about the support. Nia's text-to-speech has not moved her comprehension of complex sentences. That is not a failure of the tool; it is a finding. Whatever her barrier is, it is not access to the words. Treat every flat line as diagnostic information rather than as a reason to add another support on top.

Step 4: what the IEP tells you

The IEP is not the enemy of good instruction and it is not a formality. It is a constraint document, and reading it well is a professional skill. Four passes, in this order.

PassWhat you are looking forThe question it answers
1. Present levelsSpecific performance descriptions, baselines, conditions under which performance changesWhat is the barrier, and what is the baseline I will measure against?
2. Annual goalsThe condition clause, the behavior, the criterion, and any named routine or strategyWhat is the adapted instruction aiming at, and what counts as success?
3. Program and servicesProgram type, frequency, duration, location, group size; related services including indirect timeWho delivers this, where, when, and how often — the delivery element
4. Program modifications, supplementary aids, accommodations, supports for personnelAuthorized content changes; access supports; and what training or consultation is provided to adultsWhat am I allowed to change, what must I provide, and what support do I get?
If the IEP does not tell you

Present levels that say “struggles with reading comprehension” do not permit good design. You have two obligations and both are real: teach the student now using the best barrier hypothesis you can form from your own data, and bring your data to the team so the next IEP says something usable. Do not wait for the annual review to start teaching, and do not let a vague present-levels section survive another year because it was easier not to argue about it. The present levels section is the highest-leverage page in the document.

Worked example · Running the front half of the chain

Nia Ferguson, English 10

StepDecision
1. StandardAnalyze how an author's choices about sentence structure and text organization shape meaning in complex informational text.
2. Learning targetGiven a sentence containing an embedded relative clause from this week's text, identify the main clause and paraphrase the sentence accurately.
3. BarrierComplex syntax is not parsed reliably: with embedded or fronted clauses she assigns the nearest noun as subject. Accuracy 35% on complex sentences, at grade level on simple and compound ones. Decoding fluency is 62nd percentile, so this is not an access-to-words problem — confirmed by the flat response to text-to-speech.
4. IEP needGoal: identify the main clause and paraphrase accurately, 8 of 10 items, 3 consecutive probes. Related service: SLP 2x30 direct, 1x30 indirect. Program modification: instruction in syntactic analysis embedded in ELA content.

What the front half already rules out. Text-to-speech, extended time, and a vocabulary list are all reasonable supports that do not address a syntactic parsing barrier. Simplified text would address it by removing the standard, which fails the second purpose clause. What is left is instruction: explicitly teaching sentence deconstruction, on the class text, with the SLP's indirect time used to plan it jointly rather than to duplicate it elsewhere. That is Module 5.

Interactive activity · Competing hypotheses

Time: 30 minutes. Format: pairs.

For each student below you are given the observed failure and three plausible barrier hypotheses. For each: (a) choose the hypothesis best supported by the data in Section 1; (b) name the single piece of evidence that rules the other two out; (c) name one piece of evidence you would collect this week to confirm you are right; and (d) state what you would teach differently if hypothesis 2 turned out to be correct.

StudentObserved failureHypothesis 1Hypothesis 2Hypothesis 3
MarisolWritten CER responses average 41 words and score 1.4/4Does not understand the biology contentCannot organize and sequence a multi-part argument in written productionInsufficient time to write
DevonCaptures 24% of criterion lecture contentHandwriting speed is too slowCannot sustain attention and hold argument structure across extended oral inputLacks background knowledge in the content
Tyler30% accuracy on unfamiliar graph typesHas not learned graph vocabularyDoes not transfer a learned strategy to novel formatsReads too quickly and is careless
Cassidy34% engagement on evaluative tasksContent is too difficultEscape from evaluative demand at task onsetAssessments are too long

The hard part is (d), and it is the point of the activity. If you cannot say how instruction would change under a different hypothesis, you have not made a diagnosis — you have made a label.

Knowledge check

3.1Which is a usable barrier statement?
  1. Student has an other health impairment
  2. Student struggles with note-taking
  3. Student captures 24% of criterion lecture content; capture collapses after roughly four minutes and does not recover; performance with the source document present is strong
  4. Student is not motivated to take notes
Answer · C

A is a classification. B is a restatement of the outcome. D is an attribution with no mechanism and no test. C names a mechanism, quantifies it, and predicts a discrepancy you could go check tomorrow.

3.2A student's error analysis on 40 algebra items shows 31 of 31 incorrect responses involve operating on only one side of the equation. What does this rule out?
Answer

It rules out a computation or fluency explanation, because the arithmetic within the errors is correct. It rules out a random-error or carelessness explanation, because the errors are systematic. It rules out “needs more practice,” because more repetitions of a consistently misapplied procedure rehearse the misconception. What it rules in is conceptual: the equation is not being treated as a balanced relation. That is a content and methodology problem, and it points directly at a concrete-to-abstract sequence with justification at each step.

3.3Scenario. A student's present levels read: “Jordan has difficulty with reading comprehension and requires support.” You teach him tenth-grade global history. What do you do?
Short response — look for

Two things, in parallel. Now: form your own barrier hypothesis from evidence you can collect this week — can he answer the same questions when the passage is read to him? Do errors cluster on inference or on literal recall? Does he lose meaning at particular sentence structures? Then teach against your best hypothesis and monitor. Next: document what you find and bring it to the team so the IEP is amended or the next annual review produces present levels a colleague could act on. A strong answer names the vague present levels as a fixable problem rather than an excuse, and does not wait for the CSE to start teaching.

3.4Match each student to the SDI element that must change first.

Students: (i) Amara  (ii) Nia  (iii) Jaden

Elements: (w) Content — a prerequisite concept must be taught alongside the grade-level skill  (x) Methodology — the strategy itself must become the object of instruction  (y) Delivery — the instruction must move inside the general education text and be co-planned with the related service provider

Answer

(i)–(w): the concept of equality is missing content that has to be taught, not practiced around. (iii)–(x): Jaden knows every step of the report; what he cannot do is decompose and initiate, so the decomposition routine becomes the thing being taught. (ii)–(y): Nia's need is best served by syntactic instruction delivered inside the class text with the SLP co-planning, rather than on decontextualized cards in a separate room. Reasonable people can argue the second element for each student, and the argument is the useful part — most real cases change more than one.

3.5Why does this unit insist on starting at the grade-level standard even when a student is performing several years below it?
Short response — look for

Because the second purpose clause of the SDI definition requires access to the general curriculum so the student can meet the standards that apply to all students, and because Endrew F. requires a program reasonably calculated to enable appropriate progress. Practically: starting at the standard and identifying the point of access lets you teach the prerequisite and the grade-level content together. Starting at the student's current level and working forward almost always produces a parallel curriculum that never rejoins the standard, and the gap compounds every year. Naming the prerequisite is essential; substituting it for the standard is not the same thing.

Reflection and application

Collect fifteen errors from one student on one type of task this week. Do not score them — sort them. Then write one barrier statement with a prediction attached, and one sentence describing what you would see if you were wrong.

Takeaway

A barrier statement that does not predict where performance will break down is a label, not a diagnosis. Find the discrepancy, read the errors, and write the mechanism — then the instruction follows almost on its own.

Module 04

High-Leverage Practices for SDI

The 2024 framework, mapped onto the planning sequence you are already using.

75 minutesObjectives 4, 6All 22 HLPs
Essential question

If I only had time to get good at six things, which six would make the most difference to my students with disabilities?

Learning objectives
  • Name the four domains and six pillar practices of the 2024 HLP framework.
  • Map each step of the planning sequence to the HLPs that operationalize it.
  • Use HLP language to make an SDI plan legible to a colleague who was not in the room.
Make sure you have the right edition

The 2017 booklet High-Leverage Practices in Special Education was replaced in March 2024 by a second edition, High-Leverage Practices for Students with Disabilities (CEC and CEEDAR, free PDF). Still 22 practices with the same numbers, but the domains changed, six practices were designated pillars, and some wording changed — HLP 7 is now “responsive” rather than “respectful,” and HLP 20 gained “for academics and behavior.” The rename from “in Special Education” to “for Students with Disabilities” is deliberate: these are practices for all educators, not job duties of the special educator. If your building's materials still show the old Social/Emotional/Behavioral domain, they are the previous edition.

Key vocabulary
TermWorking definition
High-leverage practiceA practice used frequently across settings and content areas, supported by evidence, and central enough to teaching that improving it improves many other things.
Pillar practiceOne of six HLPs designated in the 2024 edition as most essential to master — HLPs 1, 3, 6, 7, 16, and 20.
Embedded practiceOne of the remaining sixteen, described as living inside and supporting the pillars rather than standing separately.
Culturally inclusive pedagogies and practicesAttention to the wholeness of students' identities and experiences, woven through every practice in the 2024 edition rather than added as a separate chapter.
Before you go on

The four domains and six pillars

DomainPillar practice(s)Embedded practices
CollaborationHLP 1 Collaborate with professionals to increase student success
HLP 3 Collaborate with families to support student learning and secure needed services
HLP 2 Organize and facilitate effective meetings with professionals and families
Data-Driven PlanningHLP 6 Use student assessment data, analyze instructional practices, and make necessary adjustments that improve student outcomesHLP 4 Use multiple sources of information to develop a comprehensive understanding of a student's strengths and needs
HLP 5 Interpret and communicate assessment information to collaboratively design and implement educational programs
HLP 11 Identify and prioritize long- and short-term learning goals
HLP 12 Systematically design instruction toward a specific learning goal
Instruction in Behavior and AcademicsHLP 16 Use explicit instruction
HLP 7 Establish a consistent, organized, and responsive learning environment
What to teach: HLP 9 Teach social behaviors · HLP 14 Teach cognitive and metacognitive strategies to support learning and independence · HLP 21 Teach students to maintain and generalize new learning across time and settings
How to teach: HLP 13 Adapt curriculum tasks and materials for specific learning goals · HLP 15 Provide scaffolded supports · HLP 17 Use flexible grouping · HLP 18 Use strategies to promote active student engagement · HLP 19 Use assistive and instructional technologies · HLP 8 & 22 Provide positive and constructive feedback to guide students' learning (22) and behavior (8)
Intensify and Intervene as NeededHLP 20 Provide intensive intervention for academics and behaviorHLP 10 Conduct functional behavioral assessments to develop individual student behavior support plans

The HLPs are the machinery of the chain

The planning sequence tells you what question to answer next. The HLPs tell you what skilled practice answering it looks like. Mapping one onto the other is the point of this module.

Planning stepHLPs that do the workWhat that means concretely
STEP 1–2
Standard → learning target
HLP 11, HLP 12Prioritize what matters most from the grade-level standard, identify the essential prerequisites, and sequence lessons so each one builds on the last. HLP 11 is explicit that goals come from grade-level standards, assessment data, learning progressions, and IEP goals together.
STEP 3
Barrier
HLP 4, HLP 10Use multiple sources — error analysis, work samples, observation, formal assessment, the student's own account — rather than one score. For interfering behavior, a functional behavioral assessment is how you find the antecedent.
STEP 4
IEP need
HLP 5, HLP 1, HLP 3, HLP 2Interpret assessment information so the team can act on it, and collaborate with colleagues and family to build the program. HLP 2 sits underneath: you cannot collaborate if meetings are run badly.
STEP 5
SDI
HLP 16 · 14 · 13 · 15 · 17 · 18 · 19 · 21 · 8/22Explicit instruction is the pillar; the rest are how it is carried out. Note where the elements land: HLP 13 changes content, HLP 16 and 14 change methodology, HLP 17 and 19 change delivery.
STEP 6
Response and data
HLP 6The pillar of Data-Driven Planning, and note its full wording: use the data, analyze your own instructional practices, and make adjustments. The practice includes examining what you did, not only what the student did.
STEP 7
Adjustment
HLP 20, back to HLP 6Intensify along a specific dimension when response is inadequate, then re-enter the loop. Module 6 makes this concrete.
Two things not to miss

HLP 21 is the one everybody skips. “Teach students to maintain and generalize new learning across time and settings” is a separate practice because generalization does not happen on its own. Tyler's entire case is HLP 21. If you teach a routine and never plan for its transfer, you have taught a classroom behavior, not a skill.

HLP 7 is not classroom management. Its 2024 wording is “consistent, organized, and responsive” — changed from “respectful.” It is a pillar because none of the instructional practices work in a room where the conditions are unpredictable, and because responsiveness is a design property, not a personality trait.

Worked example · The same case in HLP language

Devon Whitaker, U.S. History & Government

The plan, described twice. Both describe the same Tuesday.

Without the shared language

“I'm going to work with Devon on his note-taking. I'll show him a system and check how it's going. He gets my notes too, and I'll keep an eye on his essays.”

A colleague covering the class cannot implement this. Neither can the CSE evaluate it.

With it

HLP 4: criterion note-capture probes plus essay scores plus the accommodation usage log establish the barrier. HLP 11: priority goal is capture of argument structure, not detail. HLP 16: the two-column routine is taught explicitly — modeled with think-aloud, guided practice on recorded 5-minute segments, independent practice in live lecture. HLP 18: two scheduled cued pauses per lecture with a partner comparison. HLP 22: immediate corrective feedback against a criterion set, then self-scored. HLP 14: self-monitoring — he graphs his own capture percentage. HLP 21: the routine is practiced in Environmental Science and in an assembly. HLP 6: capture percentage charted weekly; decision rule at four points below the goal line.

Interactive activity · HLP audit of a real lesson

Time: 25 minutes. Format: individual with a partner debrief. Bring: a lesson plan you have actually taught in the last month.

  1. Using the pillar chart, tag every HLP visible in the lesson as you taught it. Be strict — tag only what an observer could have seen.
  2. Now pick one student with an IEP who was in that lesson. Re-tag: which of those HLPs were operating for that student's documented need, as opposed to for the room?
  3. The gap between the two lists is the honest measure of how much SDI happened.
  4. Choose one HLP to add for that student next time. Write the specific move, not the practice name. “HLP 15” is not a plan; “I will provide a partially completed causal chain for the first two paragraphs and remove it on the third” is.

Debrief question: Most teachers find the first list long and the second list short. Is that a problem with the lesson, or a correct description of a lesson that was well designed universally and not yet designed individually?

Knowledge check

4.1Which set correctly names the four domains of the 2024 HLP framework?
  1. Collaboration; Assessment; Social/Emotional/Behavioral; Instruction
  2. Collaboration; Data-Driven Planning; Instruction in Behavior and Academics; Intensify and Intervene as Needed
  3. Planning; Teaching; Assessing; Reflecting
  4. Universal; Targeted; Intensive; Individualized
Answer · B

A is the 2017 edition. If your district materials show A, they predate March 2024. The Social/Emotional/Behavioral domain was dissolved and its practices redistributed — HLP 9 into Instruction, HLP 10 into Intensify and Intervene.

4.2Name the six pillar practices and say why designating pillars was a useful editorial decision.
Answer

HLP 1 collaborate with professionals; HLP 3 collaborate with families; HLP 6 use assessment data and adjust; HLP 7 establish a consistent, organized, responsive environment; HLP 16 use explicit instruction; HLP 20 provide intensive intervention for academics and behavior.

Twenty-two practices is more than any teacher can develop at once, and an undifferentiated list invites the response of doing a little of everything badly. Naming six — the authors call them the MVPs — gives buildings a defensible focus and makes the other sixteen legible as the components of those six rather than as competing initiatives.

4.3Scenario. A teacher has taught Tyler a graph-interrogation routine. He uses it perfectly on the graphs from instruction and not at all on a new one. Which HLP is missing?
  1. HLP 16 — Use explicit instruction
  2. HLP 21 — Teach students to maintain and generalize new learning across time and settings
  3. HLP 17 — Use flexible grouping
  4. HLP 19 — Use assistive and instructional technologies
Answer · B

The explicit instruction worked — he performs the routine correctly when cued by a familiar format. What was never taught is the transfer, and for Tyler that is the whole disability-related need. Generalization is planned instruction: deliberately varied formats, deliberately faded cues, and practice in settings other than the one where the routine was learned.

4.4HLP 6 says to use student assessment data, analyze instructional practices, and make necessary adjustments. Why does that middle clause matter?
Short response — look for

Because it locates the first adjustment in the teaching rather than the student. When a student is not responding, the practice directs you to examine what you did — was it delivered as designed, at the intended frequency, with the intended explicitness — before concluding the student needs something more intensive. Skipping that step produces intensification of an intervention that was never actually implemented, which is both futile and expensive. It also protects against the drift toward explanations located inside the student.

4.5Matching. Connect each HLP to the SDI element it most directly changes.

HLPs: (i) HLP 13 Adapt curriculum tasks and materials  (ii) HLP 16 Use explicit instruction  (iii) HLP 17 Use flexible grouping  (iv) HLP 14 Teach cognitive and metacognitive strategies

Elements: (w) Content  (x) Methodology  (y) Delivery

Answer

(i)–(w), (ii)–(x), (iii)–(y), (iv)–(w) and (x) both.

HLP 14 is deliberately the interesting one. A cognitive strategy is content — it is a thing the student learns and must be taught to mastery — and teaching it explicitly is methodology. That double status is why strategy instruction is such a productive form of SDI at secondary level: it changes what the student knows and how instruction is delivered at the same time, and the thing learned is portable to every other course.

Reflection and application

Of the six pillars, which one is strongest in your practice and which is weakest? For the weakest, name one student and one specific move you will make next week. Then find the corresponding IRIS module and schedule 30 minutes for it.

Takeaway

The HLPs are not a separate initiative sitting beside SDI. They are the vocabulary for describing what skilled SDI actually consists of, and using them makes your plan implementable by somebody who was not in the room when you thought of it.

Module 05

Designing and Delivering SDI

Explicit instruction, modeling, scaffolding, guided practice, corrective feedback, active engagement, and the part everybody forgets: the plan to stop.

110 minutesObjectives 6, 7, 8HLP 16, 15, 18, 22, 14, 21
Essential question

What does the student actually see, hear, and do differently — minute by minute — because this instruction was specially designed?

Learning objectives
  • Build an explicit instruction sequence for a specific secondary learning target.
  • Distinguish modeling from demonstration, and scaffolding from doing it for the student.
  • Deliver corrective feedback that is immediate, specific, and tied to a goal.
  • Write a fading plan with criteria at the same time you write the support.
  • Maintain grade-level rigor while changing content, methodology, or delivery.
Key vocabulary
TermWorking definition
Explicit instructionHighlighting essential content, anticipating misconceptions, choosing examples and non-examples deliberately, modeling, scaffolded practice, and independent practice with feedback.
Systematic instructionContent sequenced so each lesson builds on the last; complex skills broken into steps and taught in a planned order.
Modeling / think-aloudMaking the invisible thinking audible. Not showing the answer — showing the decisions that produced it, including the ones the student currently skips.
ScaffoldTemporary support calibrated to current performance, used flexibly, evaluated, and removed once no longer needed.
Corrective feedbackGoal-directed information about the task, the process, or self-regulation, delivered close in time to the performance.
Opportunities to respondThe count of times a student actively produces an observable response in a lesson. A measurable variable, and the most underused lever in secondary teaching.
FadingThe planned, criterion-based withdrawal of support. Written before the support is first used.
Prompt dependenceWhat over-prompting produces: performance that occurs only in the presence of an adult cue.
Before you go on

Explicit instruction, described accurately

The 2024 HLP wording is worth reading closely because it is more demanding than the “I do, we do, you do” shorthand suggests:

Educators use explicit instruction to make learning new content, skills, and strategies accessible to students. When using explicit instruction, educators explain concepts by highlighting essential content, anticipating common misconceptions, and strategically choosing examples, non-examples, and language to facilitate understanding. They model and scaffold processes… Educators provide students opportunities for supported and independent practice with feedback to learn, maintain, and generalize newly learned knowledge and skills to other relevant settings and situations. HLP 16, High-Leverage Practices for Students with Disabilities, 2nd ed., 2024

Three of those clauses are routinely dropped in practice, and dropping them is what turns explicit instruction into lecture.

  • Anticipating common misconceptions. Not correcting them when they appear — planning for them in advance and building the non-example that prevents them.
  • Strategically choosing non-examples. A concept is defined as much by what it excludes as by what it includes. For Amara, 3x + 7 = 5x − 9 solved with a one-sided operation, presented deliberately as a non-example to analyze, teaches more than three more correct worked examples.
  • Maintain and generalize. Built into the definition, not an optional extension.

The sequence, with what each phase is for

  1. Set the target and activate what's relevantState what the student will be able to do and why it matters in this course. Retrieve only the prior knowledge the new skill depends on — not a general warm-up. For Nia: “Today you'll find the main clause in sentences that have a clause tucked inside them. Yesterday you found subjects and verbs in short sentences. Same job, harder sentence.”
  2. Model with a think-aloudNarrate the decisions, not the answer. Deliberately narrate the step the student's error data show they skip. For Tyler: “Before I look at the shape of this, I'm reading the vertical axis. It says millimetres. Millimetres of what? I need the caption.” Do it three times before expecting anything.
  3. Guided practice with scaffolds and high response ratesThe student does it with support. This is where opportunities to respond are won or lost: aim for a response every 30–60 seconds, not a question every five minutes to whoever raises a hand. Use whiteboards, partner turns, written responses everyone produces.
  4. Immediate corrective feedbackCorrect at the point of error, not on the returned paper next week. Name the step, not the person: “You subtracted 7 from the left side. What has to happen on the right for this to stay balanced?”
  5. Independent practice at high successAim for 85–90% success. Below that, the student is practicing errors and learning that the task is aversive. Above 95%, you have stopped teaching anything.
  6. Plan maintenance and generalizationSchedule distributed retrieval. Practice the skill in a second setting and a second format on purpose. This is HLP 21 and it is not optional.
  7. Fade on a written criterionDecide now what performance will trigger removing the next layer of support, and write it down. Support with no exit criterion becomes permanent by default.

Scaffolding versus doing it for them

HLP 15 defines scaffolded supports as assistance that lets a student complete tasks they cannot yet do independently, calibrated to performance, used flexibly, evaluated, and gradually removed once no longer needed. Every clause in that sentence is a test a real scaffold has to pass.

TestScaffoldSubstitution
CalibratedSet just above current independent performance, so the student does the hard partSet wherever it makes the work go smoothly
FlexibleAdjusted within the lesson based on what the student doesHanded out identically every time
EvaluatedYou know whether it changed performance, because you measuredAssumed to be helping because it feels helpful
RemovedWithdrawn on a written criterion, in layersStill in place in June because nobody revisited it
Who does the thinkingThe student does the cognitive work; the scaffold reduces load elsewhereThe scaffold does the cognitive work; the student does the transcription
The prompt-dependence trap

The WVDE guidance names over-prompting directly: it “leads to learned helplessness,” and systematic prompting requires a plan to fade adult prompts as quickly as possible. At secondary level the stakes are concrete. A twelfth grader who can only start a task when an adult tells him to has an SDI plan that has been quietly training him for a life with an adult permanently beside him. Every prompt you introduce should be introduced together with the sentence that removes it.

Corrective feedback that does something

HLP 8/22 calls feedback “one of the most important instructional practices,” effective when it is goal-directed, focused on task, process, or self-regulation, and delivered “timely, contingent, genuine, meaningful, age appropriate, and at rates commensurate with task and phase of learning.” That last phrase is the one that changes practice at secondary level: feedback rate should drop as the student moves from acquisition to fluency to maintenance. Dense feedback during acquisition is correct. The same density during maintenance prevents independence.

Useful

“Your claim is clear. Evidence 2 is a restatement of the claim, not data — look at column three of your table and find a number that supports it.” (Task-level, specific, actionable, points to where.)

“You skipped step one of the routine. What's step one?” (Process-level, and it makes the student retrieve rather than receive.)

“You caught your own error on number four without me. That's the self-check working.” (Self-regulation level, and it names the transferable behavior.)

Not useful

“Good job!” (Person-level, non-contingent, tells the student nothing.)

A rubric score returned nine days later. (Not timely; the performance it refers to is gone.)

“This needs more detail.” (Names a deficiency without naming the fix.)

Correcting every error in a first draft. (Volume without a priority; the student cannot act on twenty things.)

Active engagement: count it

HLP 18 is about student engagement as an engineered property of the lesson. The measurable version is opportunities to respond — the number of times a student produces an observable response. In a typical secondary lecture-and-discussion period, an individual student's OTR count is often in the single digits. For a student like Devon, whose barrier is sustaining attention across extended oral input, that count is the intervention.

  • Choral and written responses instead of hand-raising, so every student responds every time.
  • Partner turns at fixed points — and for Devon, at the two cued pauses that are already part of his note-taking routine.
  • Response cards, whiteboards, one-sentence written stops. Low technology, high count.
  • Count it once. Have a colleague tally one student's responses for twenty minutes. The number is usually a surprise and it changes planning more than any reading will.

Rigor is preserved by changing the route, not the destination

The most common way SDI goes wrong at secondary level is that the adaptation quietly lowers the target. Here is the same lesson adapted three ways for three students, with the standard held fixed.

StudentGrade-level target (unchanged)ElementThe adaptation
Marisol
Living Environment
Write a CER explanation from enzyme lab dataMethodology + contentExplicit strategy instruction in a written-argument routine using self-regulated strategy development: memorize the routine, model it, collaboratively write two, then independent. The organizer is faded in three layers. She writes the same paragraph the class writes.
Amara
Algebra I
Solve equations with variables on both sidesContent + methodologyThe prerequisite concept of equality taught concretely (balance), then representationally, then abstractly, with a spoken justification required at every transformation. Same equations as the class, six at a time instead of twenty during acquisition.
Nia
English 10
Analyze how sentence structure shapes meaningMethodology + deliveryExplicit sentence-deconstruction routine taught on the class text: box the main clause, bracket the embedded clause, paraphrase each, then recombine. Co-planned with the SLP, who uses indirect service time to build the week's sentence set from the actual reading.
Kept the destination

Marisol writes a CER paragraph on the enzyme lab. So does everyone else. What changed is that she was taught a routine for constructing one, with modeling, practice, and feedback, and the support she uses will be gone by the fourth lab.

Moved the destination

Marisol answers three short-answer questions instead of writing a paragraph, because writing is hard for her. The task is now easier and she is no longer working on the standard. Nobody decided this at a CSE meeting; it happened on a Tuesday because it was reasonable in the moment. This is how a modification arrives without anyone choosing it.

Interactive activity · Lesson redesign

Time: 35 minutes. Format: content-alike pairs. Bring: an upcoming lesson.

Working from your own lesson and one student from Section 1 (or one of your own), produce a one-page delivery plan containing exactly these things:

  1. Target, written so two observers would agree whether it was met.
  2. The model. Write the actual think-aloud script for the step the student's error data show is skipped. Write the words. This is the part people skip in planning and improvise badly in the moment.
  3. Two non-examples you will deliberately present, and what misconception each one prevents.
  4. Guided practice with the OTR count you are targeting for this student, and how you will get it.
  5. Feedback — one sentence of the exact corrective language for the most likely error.
  6. Fading criterion — the performance level at which the next layer of support comes off, and what comes off first.
  7. Generalization — the second setting or format where this will be practiced, and when.

Peer critique: swap plans. Your partner's only job is to answer one question — could I teach this lesson tomorrow from this page alone? Where the answer is no, that is the part that was still in your head.

Knowledge check

5.1Which is modeling, and which is demonstration?
  1. The teacher solves the equation on the board and the students copy the steps.
  2. The teacher solves the equation aloud, saying “I subtracted 7 from the left, so I have to subtract 7 from the right, or it isn't balanced any more — that's the step I see people skip.”
Answer

(a) is demonstration — it shows the product. (b) is modeling — it makes the decision process audible, and it targets the specific error the student's data show. For Amara, (a) is what the whole-class worked-example handout already did for five flat weeks. The difference between the two is the difference between showing the answer and teaching the thinking.

5.2A student completes independent practice at 100% accuracy every day for two weeks. What should change?
  1. Nothing — this is the goal
  2. Add more problems of the same type
  3. Fade a layer of support and raise task difficulty toward the grade-level demand; check whether performance holds
  4. Move the student to a less restrictive setting
Answer · C

Sustained 100% means the task is below the student's current capacity, usually because a scaffold is doing work the student could now do alone. The target band for independent practice is roughly 85–90%: high enough that the student is not rehearsing errors, low enough that learning is still occurring. Persistent ceiling performance is a fading signal, and if the fading criterion had been written when the support was introduced, this would have triggered automatically instead of requiring somebody to notice.

5.3Scenario. A teaching assistant sits beside Jaden every project day and tells him each next step. He completes the work. Evaluate.
Short response — look for

The work is getting done and the skill is not being learned; performance is under the control of the adult prompt rather than Jaden's own routine, which is prompt dependence, and it is directly counter to his IEP goal of independently decomposing an investigation. What should happen instead: the TA delivers a defined role within the taught routine — a single prompt at a set time, then silence — on a written fading schedule (verbal prompt → gesture toward his own checklist → nothing), with the TA recording latency to initiation each day. Two things to name as well: paraprofessionals can and do implement SDI, and they can only do it well when the special educator has specified the procedure, trained them on it, and supervises it. If nobody has done that, the failure is the plan's, not the TA's.

5.4Which feedback statement is most likely to improve future performance, and why?
  1. “Nice work on this lab report.”
  2. “This is missing analysis.”
  3. “Your claim and data are solid. Your reasoning sentence restates the data instead of connecting it to enzyme function — go back to the frame and use the ‘because’ clause.”
  4. “See rubric.”
Answer · C

It is task and process level, it is specific about what is wrong and where, and it points to the taught routine rather than to a general quality. A is person-level praise with no information. B names a deficiency without a fix. D transfers the diagnostic work to the student, which is exactly the work the student cannot yet do. Note also that C is only possible because a routine with a named component was explicitly taught — feedback quality is downstream of instructional design.

5.5You are introducing a partially completed graphic organizer for a student. What must you write down at the same moment?
Answer

The fading plan: what performance level triggers removing the next layer, which layer comes off first, and how you will know. For example: at 3/4 on the rubric across three consecutive labs, drop from full organizer to headings only; at 3/4 across three more, drop to blank paper with the routine recited from memory. Also worth writing: the generalization setting, and the criterion at which the student runs the routine without being told to.

The reason for insisting on “at the same moment” is empirical rather than bureaucratic. Supports introduced without exit criteria are almost never removed, because removing them requires someone to decide to, and nobody is scheduled to decide.

Reflection and application

Look at every support currently in place for one student. For each, write the fading criterion. If you cannot write one, you have found either a permanent accommodation, which is fine and should be named as such, or a scaffold that has quietly become permanent, which is not.

Takeaway

Explicit instruction is not lecturing clearly. It is anticipating the misconception, modeling the decision the student skips, engineering a high rate of response, correcting at the point of error, and writing the plan to stop before you start.

Module 06

Data-Based Individualization

How to know whether it is working, and what to do on the day the answer is no.

100 minutesObjectives 9, 10HLP 6, 20, 4, 5
Essential question

Four weeks in, the line is flat. What exactly do I change — and how do I avoid changing five things at once?

Learning objectives
  • Choose a progress measure sensitive enough to detect change within weeks.
  • Apply a decision rule rather than an impression.
  • Check implementation fidelity before concluding a student did not respond.
  • Intensify along one named dimension of the taxonomy of intervention intensity.
  • Decide among maintain, fade, modify, and intensify.
Key vocabulary
TermWorking definition
Data-based individualizationNCII's research-based process for individualizing and intensifying instruction through systematic use of assessment data, validated interventions, and adaptation strategies.
Progress monitoringBrief, repeated, equivalent measures of the same skill, collected frequently enough to show a trend.
Goal lineThe line from baseline to the goal criterion by the goal date. The reference against which the trend is judged.
Trend lineThe line of best fit through the collected points. The trend, not the last point, is the finding.
Decision ruleThe rule agreed in advance for when to change instruction — commonly four consecutive points below the goal line, or a trend flatter than the goal line over six to eight data points.
Fidelity of implementationWhether the instruction was delivered as designed, at the intended frequency and explicitness.
Taxonomy of intervention intensityNCII's seven dimensions along which instruction can be made more intensive.
Diagnostic assessmentAssessment conducted after non-response to identify why, so the adaptation is targeted rather than generic.
Before you go on

The five-step cycle

NCII's DBI process is deliberately boring, and its value is that it prevents the two most common failure modes: changing nothing for a year, and changing everything at once so that nobody can tell what worked.

  1. Implement a validated intervention with fidelityStart with something with evidence behind it, delivered as designed. Most “the student didn't respond” conclusions are actually “the plan wasn't delivered,” and the difference matters enormously.
  2. Monitor progressBrief, repeated, equivalent probes on the skill in the goal. Weekly for most secondary academic goals; more often for behavior.
  3. If response is inadequate, assess diagnosticallyThis is the step that gets skipped. When the line is flat, the question is not “what else could we try” but “why is this not working.” Error analysis, a brief skills inventory, an FBA for behavior.
  4. Adapt the intervention along a named dimensionChange one thing, deliberately, chosen because the diagnostic pointed at it.
  5. Monitor again, and repeatThe cycle continues. Adaptation is normal, not a sign that the original plan was wrong.

Choosing a measure you can actually use

At secondary level the measurement problem is real: unit tests are too infrequent, and standardized measures are too insensitive to show four weeks of change. What works is a short, repeatable probe built from the goal itself.

StudentMeasureFrequencyWhy this one
MarisolCER paragraph on a novel data set, scored on the 4-point department rubricWeeklyDirectly samples the goal behavior. Same rubric the class is scored on, which keeps rigor visible.
Amara5 equations with variables on both sides; percent correct and percent with a valid written justificationWeeklyTwo scores separate the procedure from the concept. The justification score will move first if the conceptual instruction is working.
DevonPercent of criterion content captured in notes from a 20-minute segmentWeeklyA criterion set makes an otherwise subjective product countable.
Tyler4 interpretation items on a graph type he has never seenWeeklyThe measure must use novel formats, because familiar-format performance is already at ceiling and would show nothing.
JadenLatency to initiation in minutes; percent of dated steps completed on timeEach project dayLatency is the sensitive measure. Completion alone hides the ten minutes of not starting.
Nia10 complex sentences from the week's text: identify main clause, paraphraseWeeklyDrawn from the actual curriculum, so improvement means improvement on the standard.
CassidyPercent of evaluative tasks begun within 2 minutes; percent completedEach occurrenceTwo measures matching the two-part behavior: initiation and persistence.
Three rules for the measure
  1. It must sample the goal behavior. If the goal is written CER and the measure is a multiple-choice quiz, you are monitoring something else.
  2. It must be short enough to actually happen. A perfect measure administered twice a semester is worse than a rough one administered weekly.
  3. Forms must be equivalent. If this week's probe is harder than last week's, the line means nothing. This is the most frequently violated rule in teacher-made progress monitoring.

Reading the graph: four patterns

Trend meets or exceeds the goal line

Decision: maintain. Keep going. Begin planning the fade — and check that the fading criterion you wrote in Module 5 has not already been met while you weren't looking.

Performance at ceiling, several points running

Decision: fade, and raise the goal. The task is now below capacity. Remove the next layer of support and increase difficulty toward the grade-level demand. A ceiling is not a success to be preserved; it is a signal that the instruction has stopped teaching.

Four consecutive points below the goal line, or a flat trend

Decision: diagnose, then intensify. Do not add supports on top. Check fidelity first, then find out why, then change one dimension.

Highly variable, no discernible trend

Decision: look at conditions before instruction. Variability usually means something is changing that is not the instruction — who delivered it, which day, the probe difficulty, attendance, whether the period was interrupted. Fix the measurement and the consistency before concluding anything about the student.

Before you intensify: fidelity

HLP 6 says to use the data, analyze instructional practices, and adjust. The middle clause comes first. Four questions, honestly answered:

  • Did it happen? Not “was it scheduled” — did it occur, on how many of the planned occasions? Count.
  • Was it delivered as designed? Was the strategy explicitly modeled, or handed over? Was feedback immediate, or on returned work?
  • Was it the same across people? If a special educator, a general educator, and a paraprofessional all deliver parts of it, are they delivering the same thing?
  • Was the dosage what was planned? Three times a week means three, and “when we had time” is a fidelity finding, not a scheduling detail.
Why this order matters

Intensifying an intervention that was never delivered produces a more intensive version of nothing, consumes scarce staffing, and generates a record suggesting the student needs a more restrictive placement. The fidelity check protects the student from a placement decision built on an implementation failure.

The seven dimensions of intensification

When fidelity is confirmed and the diagnostic is done, NCII's taxonomy gives you named dimensions to change. Change one, deliberately, and keep monitoring.

DimensionWhat it meansWhat changing it looks like at secondary level
StrengthHow well the program works for students with intensive needsReplace a locally assembled routine with one that has evidence behind it — e.g. move Marisol from a homemade organizer to self-regulated strategy development for argumentative writing.
DosageOpportunities to respond and receive corrective feedbackNot “more minutes.” More responses per minute. Amara does 12 justified transformations in fifteen minutes rather than 6 problems in thirty.
AlignmentHow well the intervention targets the student's actual skill needsThe diagnostic found the barrier is conceptual, so instruction moves from procedural practice to the concept of equality. Misalignment is the single most common reason for flat lines.
Attention to transferExplicitly teaching the student to connect the skill to other contextsTyler's entire adaptation. Deliberately novel formats, explicit “this is the same routine” statements, practice in a second course.
ComprehensivenessHow fully the intervention uses explicit instruction principlesAdd the missing components: were misconceptions anticipated? Were non-examples used? Was there a model, or only a handout?
Behavioral or academic supportWhether an academic intervention includes behavioral/self-regulation components, and vice versaCassidy's case reversed: her behavior plan gains an academic component. Jaden's academic plan gains self-monitoring.
IndividualizationOngoing use of progress and diagnostic data to adapt for this studentMoving from a standard protocol to a plan revised every four weeks against his own data. This is the dimension that turns intervention into SDI.

Worked example · A flat line, diagnosed and fixed

Amara Okafor — weeks 1 through 12

Week% correct% w/ justificationWhat happened
1–520, 25, 20, 30, 200, 0, 0, 0, 0Baseline condition: reference sheet, extended time, whole-class worked examples. Flat.
5Decision pointFour consecutive points below the goal line. Fidelity check: the worked-example handout was distributed every day; no explicit modeling of the balance concept ever occurred. Finding — the designed instruction was never delivered.
5Diagnostic40-item error analysis: 31 of 31 errors are one-sided operations. Arithmetic is correct throughout. Conclusion: alignment. The plan targeted procedure; the barrier is conceptual.
6–935, 45, 55, 6040, 60, 70, 80Intensified on alignment: concrete balance model → representational → abstract, with a spoken justification required at each transformation. Note the justification score moves first — that is the concept arriving before the accuracy does.
10–1270, 80, 8585, 90, 90Intensified on dosage: 12 justified transformations per session instead of 6. Trend now exceeds the goal line. Next decision is fading the manipulatives, not adding anything.

What made this work: one dimension changed at a time, so the effect is attributable; the diagnostic came before the change; and the second measure — justification — gave an early signal four weeks before the accuracy line moved. Without that second measure, week 6's 35% looks like another flat week and the plan might have been abandoned exactly when it started working.

Interactive activity · Decision-making simulation

Time: 30 minutes. Format: table groups. Facilitator note: reveal one round at a time.

Your group is the team for one student. At each round you get four weeks of data and must make one decision — maintain, fade, modify, or intensify — state your reason, and if intensifying, name the single dimension. You may not change more than one thing.

RoundDataYour decisionDimension & reason
1Devon, note capture: 24%, 22%, 26%, 25%. Goal 80% by June; goal line requires ~4 points per week. The routine was taught in one 40-minute lesson in week 1 and has not been modeled since.
2After your round-1 change: 30%, 41%, 48%, 52%. Devon says he “forgets to use it in the second half of class.”
3After your round-2 change: 68%, 74%, 79%, 81%. He is now using the routine unprompted in U.S. History. In Environmental Science he is not using it at all.
4After your round-3 change: 84%, 88%, 86%, 90% in both courses, with the guided-note template still in use.

Suggested reasoning: Round 1 — fidelity failure, not student non-response; the fix is comprehensiveness, restore modeling and guided practice, do not intensify anything yet. Round 2 — add self-monitoring; the second-half decay points at behavioral/self-regulation support. Round 3 — attention to transfer; teach the routine explicitly in the second course rather than assuming it travels. Round 4 — maintain and begin fading the template on a written criterion; the goal is now nearly met and the support is the thing left to remove.

Knowledge check

6.1A progress-monitoring graph shows four consecutive points below the goal line. What is the correct first move?
  1. Intensify the intervention
  2. Lower the goal
  3. Check whether the instruction was delivered as designed, then assess diagnostically to identify why
  4. Refer for a more restrictive placement
Answer · C

Fidelity, then diagnosis, then adaptation. Intensifying first risks scaling up something that never happened. Lowering the goal changes the measurement rather than the instruction. Referring for a more restrictive placement on the basis of undiagnosed non-response is how students end up in restrictive settings for reasons that were fixable in the general education classroom.

6.2Match each situation to the taxonomy dimension you would change.

Situations: (i) The student performs the routine well in the resource room and not in biology. (ii) Diagnostic assessment shows the intervention targets fluency but the barrier is conceptual. (iii) The student gets 6 practice opportunities in a 30-minute session. (iv) The academic plan is sound but the student stops working when frustrated.

Dimensions: (w) Alignment  (x) Dosage  (y) Attention to transfer  (z) Behavioral support

Answer

(i)–(y), (ii)–(w), (iii)–(x), (iv)–(z).

On (iii): note that dosage is defined as opportunities to respond and receive feedback, not minutes. Thirty minutes with six responses is low dosage. Fifteen minutes with twenty responses is higher dosage in less time, which is usually also more feasible to schedule.

6.3Scenario. A student has scored 100% on her weekly probe for five consecutive weeks, well above her goal line. Her teacher is pleased and continues. Evaluate.
Short response — look for

A sustained ceiling is a signal, not a success. Either the probe is too easy to detect further growth, or a support is doing work the student could now do independently. The response is to fade a layer of support and raise the task toward the grade-level demand, then watch whether performance holds. Continuing unchanged wastes instructional time and, worse, builds dependence on a support that is no longer needed — and for a student whose goal is independence, that is an active harm rather than a neutral one.

6.4Why change only one dimension at a time?
Answer

Attribution. If you add small-group delivery, a new strategy, more minutes, and a self-monitoring sheet in the same week and the line moves, you have learned that something worked and you will now carry all four forever, including the three that did nothing. Changing one dimension keeps the plan explainable to the next teacher, keeps it defensible at a CSE meeting, and keeps the student's time from being spent on components that are not earning it. The practical exception is a crisis, where you stabilize first and analyze after — but say so out loud when you do it.

6.5Which is the better progress measure for Tyler, and why?
  1. Weekly quiz on the graphs used in that week's instruction
  2. Four interpretation items on a graph type he has never seen
Answer · (b)

His familiar-format performance is already 90–95%, so (a) would produce a flat line at the ceiling and show nothing at all. The goal is generalization to novel formats, so the measure has to sample novel formats. General principle: the measure must sample the goal behavior under the conditions the goal specifies. A measure that cannot go down cannot go up either.

Reflection and application

Pick one student and one goal. Design a probe you could actually give every week in under five minutes, and write the decision rule you will use. Then put the first four collection dates in your calendar now — the calendar entry is the intervention that makes the rest of it happen.

Takeaway

A flat line is information, not a verdict on the student. Check that the instruction happened, find out why it didn't work, change one named dimension, and watch again.

Module 07

SDI in Secondary Classrooms

Seven real high-school tasks, seven barriers, and what changes on Tuesday. Plus the documentation problem.

100 minutesObjectives 7, 11, 12HLP 1, 13, 21
Essential question

Could a colleague who has never met this student pick up my plan and teach it correctly on Monday?

Learning objectives
  • Design SDI for the specific task demands of secondary coursework.
  • Distinguish a task's surface demand from its underlying cognitive demand.
  • Plan SDI within the constraints of a 42-minute period, a co-taught section, and a shared caseload.
  • Document SDI so another educator can implement it without a conversation.
Key vocabulary
TermWorking definition
Task demand analysisListing everything a task actually requires, including the demands nobody teaches because they are assumed.
Surface vs. cognitive demandWhat the task looks like (write a lab report) versus what it requires (sequence a multi-day process, sustain a claim across paragraphs, quantify uncertainty).
Embedded instructionDelivering SDI inside the general education activity rather than beside it.
Co-planningJoint decisions about content, methodology, and delivery before the lesson. Distinct from co-teaching, and the part more often missing.
Before you go on
  • WVDE SDI Guidance, Section 2 (pp. 8–10): roles in planning and delivering SDI, co-teaching, and the role of instructional support staff.
  • TIES Center 5-15-45 Tool — collaborative planning structured by how much time you actually have. Useful if you have no common planning period.
  • 8 NYCRR §200.6 — know the parameters of the service you are delivering.

Analyze the task before you adapt it

Secondary tasks are compound. “Write a lab report” is not one demand, it is eight, and a student can fail the task while being fully capable of seven of them. Adapting the whole task when one component is broken is how rigor gets lost.

TaskWhat it actually demands
Interpret a scientific graphDecode axis labels and units · identify the variables · recognize the graph type's conventions · describe a relationship · distinguish what the data show from what they do not · connect to content knowledge
Read complex informational textDecode · parse sentence-level syntax · hold referents across sentences · identify structure · integrate with prior knowledge · monitor comprehension and repair
Write an evidence-based explanation (CER)Identify a claim · select relevant data · quantify or quote precisely · articulate a causal link · sequence the argument · sustain it across sentences · transcribe
Conduct a lab investigationRead and hold a procedure · sequence steps under time pressure · use tools accurately · record data as you go · recover from an unexpected result · coordinate with a partner
Solve a multi-step equationRepresent the relation · choose an operation · apply it to both sides · track state across steps · check by substitution
Analyze a primary sourceDecode archaic or dense syntax · source the document · contextualize · infer purpose and audience · corroborate · hold multiple documents in mind
Take notes from a lectureSustain attention across 20+ minutes · hold information while writing · distinguish argument from example · abbreviate · recover after losing the thread
Complete a long-term projectDecompose into steps · estimate and schedule · initiate without a cue · track progress across days · re-plan when something changes
Participate in academic discussionFollow rapid speech · hold a peer's claim in mind · formulate a response with an evidentiary move · produce it in real time · take turns
The productive question

Not “is this task too hard for the student?” but “which of these demands is the one they cannot meet, and can I teach it?” The first question leads to substituting the task. The second leads to SDI.

Seven students, seven Tuesdays

Environmental Science · Tyler · interpreting an unfamiliar graph

Standard & targetStep 1–2
Interpret and evaluate scientific graphs as evidence. Target: given a climograph, state what each axis measures, describe the relationship, and identify one claim the graph does not support.
BarrierStep 3
Does not transfer a learned interrogation strategy to novel formats; applies a familiar-graph procedure instead. 92% familiar, 30% novel.
IEP needStep 4
Goal: apply a taught graph-interrogation routine to unfamiliar graphs, 80% accuracy, 4 of 5 novel types. Program modification: instruction in generalizing cognitive strategies across formats.
SDIStep 5 — methodology + content
Teach a four-step routine as content: axes → units → relationship → limits. Model with think-aloud on a graph type he has never seen, narrating the axis step he skips. Then guided practice on three deliberately unfamiliar types (box plot, climograph, stacked area) with immediate correction. Then independent on a fourth. Explicitly state the transfer: “This is the same routine. The graph changed; the routine doesn't.” Practice it once in his Living Environment class.
EvidenceStep 6
Weekly: 4 items on a novel graph type. Target 80%, 4 of 5 types.
If no responseStep 7
Intensify attention to transfer: increase the variety and rate of novel formats, add a self-check where he names the routine aloud before starting.

English 10 · Nia · complex informational text and academic discussion

Target: identify the main clause and paraphrase sentences containing embedded relative clauses; contribute a response that builds on a peer and cites the text.

SDI (methodology + delivery). Sentence deconstruction taught explicitly on the class text: box the main clause, bracket the embedded clause, paraphrase each part, recombine. Modeled on three sentences from the assigned reading, guided on three, independent on four. For discussion, three moves are taught to mastery — not posted, taught: “Building on what ___ said…” / “The text says ___, which suggests…” / “I'd add ___, because…” Rehearsed one-to-one, then in a pair, then in the four-person seminar. The SLP's 30 minutes of indirect service each week is spent selecting next week's sentence set from the actual reading and building the discussion rehearsal, so therapy and ELA are the same instruction rather than two.

What was rejected and why. Simplified text — removes the standard. Text-to-speech alone — already in place, already flat, wrong barrier. Excusing her from discussion — removes the practice opportunity her goal depends on.

Chemistry · Jaden · a multi-day investigation

Target: plan and carry out a multi-day investigation and produce a formal report.

SDI (methodology + content). Decomposition is taught as a routine, not supplied as a packet. Four steps: read the end product → list what must exist for it → put a date on each → name the first physical action. That last step is the one his data demand: his failure is at initiation, so the routine has to terminate in something he can start doing within thirty seconds. Modeled on one investigation, guided on two, independent on the fourth. Advance notice of any procedural change is provided as an accommodation, and separately he is taught a two-step recovery routine for when something changes anyway, because his postsecondary goal does not include a world that warns him.

Measure: latency to initiation in minutes, each project day, plus percent of dated steps completed on time. Fading: teacher writes 4 of 4 dates → 2 of 4 → 0. Criterion for each step down is two consecutive projects at or above 80% on-time steps.

U.S. History · Devon · notes and primary sources

Target: synthesize lecture, text, and primary sources into an evidence-based argument.

SDI (methodology + delivery). A two-column routine taught explicitly — argument on the left, supporting detail on the right — because his capture data show he records anecdote and loses structure. Two cued pauses per lecture at planned points, each with a 30-second partner comparison. Immediately after, he compares his notes to a criterion set, highlights matches, counts, and graphs the percentage himself. That self-graphing is the self-monitoring component and it is faded last. For primary sources, six terms that carry the argument are pre-taught with retrieval practice distributed across the week rather than massed on Monday, because his working memory profile predicts the Monday version disappears by Thursday.

Generalization: the routine is explicitly re-taught, not assumed, in Environmental Science. HLP 21 is a lesson, not a hope.

Anatomy & Physiology · Cassidy · a lab practical

Target: complete a dissection practical and defend identifications orally.

SDI (methodology + delivery). Her academic skills are not the problem; initiation under evaluative demand is. A self-regulation routine is taught explicitly outside the moment of demand: name the feeling → state the first action → set a two-minute timer → rate afterward. Practice starts on low-stakes tasks and moves up a graduated ladder of evaluative pressure — ungraded self-check, ungraded partner check, graded partner check, graded teacher observation — with the ladder written down in advance so she can see the end of it. Check-ins occur on a fixed thinning schedule (2 min, then 5, then 12), and each one is feedback on the routine rather than on the biology. The break pass stays, and its use is now data.

What is explicitly rejected: reducing the number of items. Her data show she leaves before the first item, so fewer items is aimed at a demand that was never the problem, and it lowers the expectation for a student performing at 87% on the work she does submit.

Living Environment · Marisol · the CER paragraph

Target: write a CER explanation from enzyme lab data.

SDI (methodology). Self-regulated strategy development for written argument: she memorizes the routine before she uses it, the teacher models it aloud on two data sets including a deliberate false start and repair, they write two collaboratively, then she writes independently. Corrective feedback at the point of error, tied to the named component (“that's a restatement, not reasoning”). The organizer fades in three layers: full boxes → headings only → blank paper with the routine recited. Because her oral performance is strong, oral rehearsal precedes writing during acquisition and is faded first — it is a bridge from a strength, not a permanent substitute for writing.

Note on the accommodation: extended time and a word processor stay. They are not the plan; they are the conditions the plan runs in.

Algebra I · Amara · equations with variables on both sides

Target: solve 3x + 7 = 5x − 9 and justify each transformation.

SDI (content + methodology). Covered in Modules 5 and 6: the prerequisite concept of equality taught concretely then representationally then abstractly, with a spoken justification at every move, and deliberate non-examples of one-sided operations presented for analysis. Same equations as the class; six with justification during acquisition rather than twenty bare.


Working inside real constraints

Forty-two minutes

SDI does not require a separate block. Most of what is described above occupies five to twelve minutes and happens inside the lesson.

  • Pre-teach in the margins. Ten minutes before the lab, during a study hall, in the first five minutes while others settle.
  • Embed the model in the whole-class model. When you think aloud for everyone, deliberately narrate the step your student skips. Everyone benefits; one student was the reason.
  • Use guided practice as the SDI window. While the class works, that is the block where a group of three with the same error pattern gets fifteen minutes of designed instruction.
  • Make the exit ticket the probe. Progress monitoring you have to schedule separately usually does not happen.

Co-taught sections

Under §200.6, integrated co-teaching provides specially designed instruction and academic instruction to a group including students with and without disabilities, capped at 12 students with disabilities. What makes it work is not the staffing; it is the planning. The WVDE guidance is blunt about the failure modes: splitting subjects, one teacher circulating while the other teaches, one adult hovering over two students, or one person's ideas determining everything.

The practical minimum is a standing agreement about three questions before each unit: What is the target for all students? Which students need which designed instruction inside it? Who delivers which part? If you have no common planning time, the TIES 5-15-45 tool structures the conversation for whatever you do have — five minutes in a hallway is enough for the third question.

Teaching assistants and paraprofessionals

Instructional support staff can and do implement SDI, under the guidance and supervision of a certified special educator or related service provider. WVDE's conditions are the right ones: the special educator designs it, the TA is trained on the specific procedure and given the materials, communication is ongoing, and the TA collects data. What a TA cannot do is invent the plan in the moment, and asking them to is a design failure by the adult who wrote the plan, not a performance failure by the person implementing it.


Documentation: the test is a substitute teacher

Objective 12 is that another educator could implement your SDI from your documentation. This is not a compliance concern. It is the difference between a plan that survives your absence, your transfer, and June, and one that lives in your head.

Not implementable

Provide graphic organizers and check for understanding. Use small group as needed. Provide strategy instruction for writing.

Three sentences, zero decisions transferred. Nobody else can do this.

Implementable

CER strategy instruction, 3x/week during guided practice, 12 min, group of 2 (special educator).

Routine: Claim → 3 Evidence → Reasoning “because” clause. She recites it before writing.

Model 1 per session with think-aloud incl. one repair. Then collaborative write, then independent.

Feedback: name the component, not the paper. Standard correction for the common error — “that's a restatement, not reasoning; go to the because clause.”

Support level: currently Layer 2 (headings only). Fade to Layer 3 (blank) at 3/4 rubric x3 consecutive labs.

Probe: Friday, novel data set, department rubric. Chart in shared folder.

Materials: SRSD card, layer-2 template, rubric — in the blue folder, top drawer.

The five things a plan must carry
  1. What is taught — the routine or concept, named specifically enough to reproduce
  2. How it is taught — model, guided practice, feedback language for the predictable error
  3. Who, where, when, how often, how grouped
  4. Current support level and the criterion for the next fade
  5. The probe, its schedule, and where the data live

Interactive activity · The substitute test

Time: 30 minutes. Format: pairs, cross-content if possible.

  1. Write your current SDI documentation for one student, exactly as it exists right now. Do not improve it.
  2. Hand it to a partner who does not teach that student and, ideally, does not teach your subject.
  3. Your partner writes what they would actually do in the lesson, using only your document. No questions allowed.
  4. Compare their version to what you intended. Every gap is a decision you never wrote down.
  5. Rewrite against the five-item list. Re-test with a different partner.

Facilitator note: this activity is uncomfortable and it is the highest-yield thirty minutes in the unit. Say up front that everyone's first draft fails, because everyone's does.

Knowledge check

7.1A student cannot complete a lab report. Which question leads toward SDI?
  1. Is this assignment appropriate for this student?
  2. Which of the demands in this task can the student not meet, and can I teach it?
  3. Can I give a shorter version?
  4. Should the student be in a different class?
Answer · B

A, C, and D all resolve by changing the task or the placement. B is the only one that treats the task as fixed and the instruction as the variable, which is what the second purpose clause of the SDI definition requires. Note that B might still lead to a modification — but as a team decision reached after analysis, not as a Tuesday improvisation.

7.2Scenario. In a co-taught Living Environment section, the special educator circulates and helps students who raise their hands while the general educator teaches. Evaluate against §200.6.
Short response — look for

Integrated co-teaching is defined as providing specially designed instruction and academic instruction to the group. A circulating adult providing responsive help to whoever asks is delivering neither — it is undifferentiated assistance, and it is not tied to any student's documented need, taught to anyone, or monitored. What it should look like: co-planned decisions about which students need which designed instruction inside this unit, a named instructional role for both teachers during the lesson, and a defined window — typically guided practice — where designed instruction is actually delivered to identified students. Also fair to name: this failure is usually a planning-time problem, not an unwillingness problem, and the fix is a standing three-question agreement rather than exhortation.

7.3Which documentation would let a substitute implement the plan?
  1. “Provide scaffolding and check in frequently.”
  2. “Sentence deconstruction routine, 10 min at start of independent reading, group of 3. Box main clause, bracket embedded clause, paraphrase each, recombine. Model 2 sentences from today's text, guide 2, independent 4. Currently at Layer 2 (teacher boxes, student brackets). Friday probe: 10 sentences from the week's text. Materials in the green folder.”
Answer · (b)

It names what is taught, how, who and when, the current support level, the measure, and where the materials are. A substitute could run it. Item (a) is what most IEPs and most sub plans actually contain, which is why so much SDI stops the day the teacher is absent — and why continuity is a real equity issue rather than a paperwork one.

7.4A teaching assistant is assigned to support a student during a multi-day chemistry project. What should her role be?
Answer

A specified role within instruction designed and supervised by the special educator: deliver one defined prompt at a set time according to a written fading schedule, then withhold further prompting; record latency to initiation; report the data. She should have been trained on the routine and have the materials. What she should not be doing is deciding in the moment what help to give, which produces inconsistency across days and prompt dependence over time. If nobody has designed the procedure and trained her on it, the fix is upstream of her.

7.5Design task. A student in eleventh-grade Global History has an IEP goal for sourcing and contextualizing primary documents. Her barrier is holding multiple documents in mind while comparing. Sketch the SDI in three sentences.
Sample response — look for a named routine, explicit teaching, and a measure

Teach an external-memory routine explicitly: for each document, record source, date, purpose, and one-line claim on a numbered card before any comparison begins, so the comparison operates on the cards rather than on memory. Model it aloud on two documents, guided practice on a three-document set with immediate correction at the point of omission, then independent on a four-document set, with the card template faded to a blank index card by the third set. Measure weekly with a two-document comparison scored on a sourcing rubric, and if the trend is flat, check first whether the routine was actually modeled each week before intensifying on comprehensiveness.

What makes this SDI rather than a good idea: the routine is content that is explicitly taught, the barrier it targets is documented, there is a probe, and there is a fade.

Reflection and application

Take the task your students find hardest in your course. Write out every demand it makes, including the ones you have never taught because you assumed them. Then pick one student and circle the demand that is actually stopping them. Most teachers find they have been adapting a different demand entirely.

Takeaway

Secondary tasks are bundles of demands. Find the one that is broken, teach it explicitly inside the grade-level task, and write it down well enough that someone else could do it on Monday.

Module 08

SDI Design Challenge

A new student, the whole chain, and a structured peer critique. This is the rehearsal for the performance task.

90 minutesAll objectivesRehearsal for Section 12
Essential question

Given everything, what would I actually do — and what would make me change my mind?

Learning objectives
  • Run the complete seven-step sequence on an unfamiliar case.
  • Justify each decision with evidence from the case rather than with preference.
  • Critique a colleague's plan against the same criteria you will be scored on.

The case

Rafael Ortiz

Grade 10 · Global History & Geography II (Regents) · Traumatic brain injury
Background
Sustained a TBI in a car accident in seventh grade. Prior to the injury he was an honors student. Returned to school in eighth grade; classified in ninth.
Strengths
Excellent reasoning in the moment. In class discussion he constructs sophisticated causal arguments and handles counterexamples well. Reads grade-level text accurately and comprehends it. Writes clearly. Warm and well liked; peers seek him out for group work.
Interests
Cooking — runs a food account with a real following. Wants to go to culinary school and eventually own a restaurant.
Grade-level expectations
Analyze causes and effects of global events across periods; write a thematic essay sustaining an argument with evidence from multiple eras; complete a Regents examination requiring recall across a full year of content.
Present levels
On end-of-lesson exit tickets, Rafael scores 85–95%. On the same content assessed one week later, he scores 30–40%. On the same content assessed at the end of the unit, 25–35%. Cumulative assessments spanning multiple units, 22%. In discussion he can reason expertly about material he cannot recall unaided the following week. He frequently says “I know I knew this.” Performance in the last twelve minutes of the period is measurably weaker across all task types.
Disability-related needs
(1) Consolidation and delayed retrieval — new learning is acquired normally and is not retained or retrieved without support. (2) Cognitive fatigue in the latter part of a period.
IEP information
Annual goal: using a taught retrieval and review routine, Rafael will score 75% or higher on delayed assessments of content taught 7–14 days earlier, in 4 of 5 trials. Program: consultant teacher services, direct, 3x weekly; resource room 5x weekly. Program modification: instruction in retrieval, review, and self-management routines. Accommodations: extended time, breaks as needed, use of a personal reference notebook on assessments.
Assessment data
Immediate recall, 40th percentile. Delayed recall (30 min), 4th percentile. Recognition memory substantially stronger than free recall. Reading comprehension, 55th percentile. Verbal reasoning, 75th percentile.
Classroom observation
Takes thorough, well-organized notes and never looks at them again. Studies the night before a test by rereading. Genuinely surprised by his test scores. In the last ten minutes of class he stops volunteering and his written work becomes noticeably shorter.
Current accommodations and strategies
Extended time, breaks as needed, a personal reference notebook on assessments, and a study guide distributed to the whole class before each unit test.
Student response data
Unit test scores across the marking period: 38%, 41%, 33%, 40%, 36%. Exit tickets over the same period average 89%. The gap is stable and has not narrowed.
What Rafael says
“I get it when we do it. I just can't get it back later. Rereading my notes makes me feel like I know it and then I don't.”
Note what the case rules out

Rafael's 89% exit-ticket average rules out an acquisition problem, a comprehension problem, an attention problem during instruction, and an effort problem. His accurate grade-level reading rules out text access. His strong reasoning in discussion rules out a content-difficulty explanation. Whatever the plan is, it has to target the interval between learning and retrieval, and almost every support currently in place targets something else.

The challenge

Time: 45 minutes to build, 30 to critique, 15 to revise. Format: pairs build, then swap with another pair.

Build a complete SDI plan using the planning template from Section 16. Answer all seven questions, plus these six:

  1. What is the learning target? Pick one real Global History task and write the target so two observers would agree whether it was met.
  2. What barrier is preventing access? State the mechanism, and name the evidence in the case that supports your statement over a competing one.
  3. What does the IEP tell us? Cite the goal, the program, and the program modification, and say what each constrains.
  4. What SDI should be provided? Name the element(s) — content, methodology, delivery — and describe what happens minute by minute.
  5. What evidence-based practice supports it? Name it and say why it fits this barrier.
  6. How will the teacher know whether it worked? Specify the probe, the schedule, and the decision rule.
  7. What should change if he does not respond? Name the taxonomy dimension you would change first and the one you would change second.

Constraints, so the plan is real: 42-minute periods. You have consultant teacher time three days a week and resource room five. The Regents examination in June requires recall across the full year. Do not lower the target.

Peer critique protocol

Swap plans with another pair. Fifteen minutes each way. Ask these questions in order and write the answers on the plan itself.

  1. Is the target observable?Could two people watch the same period and agree? If not, stop here — nothing downstream can be evaluated.
  2. Is the barrier a mechanism or a label?Does it predict where performance breaks down? Find the sentence in the case that would falsify it.
  3. Does the SDI target that barrier, specifically?The most common failure is a good practice aimed at a different problem. If the barrier is retrieval and the plan is a better study guide, say so.
  4. Is the student being taught, or supplied?Point to the sentence where someone explicitly teaches Rafael something. If there isn't one, the plan is a support plan.
  5. Is the target still grade level?Check the second purpose clause. Has anything been quietly reduced?
  6. Could you teach it Monday from this page?Name the first decision you would have to guess at.
  7. Is there a fade?What comes off first, and at what criterion?

One defensible answer

Read this after you build yours. It is one good plan, not the answer, and a differently reasoned plan can score just as well.

Rafael Ortiz — SDI plan

Global History & Geography II
What should he learn?Standard / target
Standard: analyze causes and effects of global events across periods and sustain an argument with evidence from multiple eras. Target for this unit: given a thematic prompt spanning the Industrial Revolution and imperialism, produce three accurate causal claims with specific supporting evidence, from memory, seven days after instruction.
What can he do now?Baseline
Immediate: 89% (exit tickets). Delayed 7 days: 30–40%. Cumulative: 22%. Reasoning and comprehension at or above grade level.
What is interfering?Barrier
Consolidation and delayed retrieval. Acquisition is intact; the information is not durably encoded or is not retrievable without a cue. Recognition > free recall on formal testing, which points at retrieval rather than storage. Secondary barrier: cognitive fatigue reduces performance in the final twelve minutes.
What does the IEP identify?Need / goal
Goal: 75% or higher on delayed assessment of content taught 7–14 days earlier, 4 of 5 trials, using a taught retrieval and review routine. Program modification authorizes instruction in retrieval, review, and self-management. Consultant teacher 3x, resource room 5x supplies the delivery capacity.
What changes?Element
Content — a retrieval routine becomes something he is taught, not something he is told to do. Methodology — spaced retrieval practice replaces rereading. Delivery — the spacing schedule is distributed across the consultant teacher and resource room periods, and demanding tasks move to the first half of the period.
What does the teacher do?SDI
1. Teach the routine explicitly. A three-question self-quiz card written by Rafael at the end of each lesson: what was the claim, what was the evidence, what did it cause? Modeled by the teacher for four lessons with a think-aloud, guided for four, independent thereafter. Writing the questions is the instruction; answering them is the practice.

2. Schedule the spacing and teach him to run it. Retrieval at day 1, day 3, day 7, day 21 — free recall first, card only to check. Weeks 1–4 the special educator runs the schedule; weeks 5–8 they build it together; from week 9 Rafael schedules it and the teacher audits.

3. Break the rereading illusion, explicitly. Once, deliberately: predict a score after rereading, take a free-recall probe, compare. Then repeat after retrieval practice. This is a taught metacognitive lesson, not a gotcha, and it is what makes him keep doing the harder thing.

4. Fatigue as a delivery decision. New content and retrieval practice in the first 30 minutes; consolidation, card-writing, and review in the last 12. Timing is scheduled, not left to how the period goes.

5. Make the reference notebook an instructional object. He is taught to build it as a retrieval-cue index — questions, not answers — and to practice retrieving from the cue before opening the page.
What does he do?Observable
Writes three self-quiz questions in the last five minutes of each lesson. Completes free-recall retrieval at four scheduled intervals. Records his own delayed-recall percentage on a chart. From week 9, sets his own retrieval schedule for the coming week.
How is progress measured?Data
Weekly: a five-item free-recall probe on content taught 7 days earlier. Monthly: a cumulative probe spanning prior units. Both charted against a goal line to 75% by June.
Is it working?Decision rule
Four consecutive weekly points below the goal line, or a trend flatter than the goal line over six points, triggers review. Check fidelity first: did the card get written every lesson, did the day-3 and day-7 retrievals actually occur?
What next?Adjustment
First: dosage — add a day-14 retrieval and increase items per session. Second: comprehensiveness — add elaborative interrogation (“why would that cause this?”) at each retrieval, since his reasoning strength is unused in the current routine. Third: alignment — if free recall remains near floor while recognition stays strong, shift the goal's response format toward cued recall and teach cue self-generation, and bring that to the team as a proposed goal amendment rather than deciding it alone.
What the weaker plans do

In field testing, the three most common weak responses to this case are: a better study guide (supplies retrieval cues instead of teaching retrieval, and does nothing for the Regents in June); more time and a quieter room (aimed at a barrier the data rule out); and reduce the content he is responsible for (fails the second purpose clause, and abandons a student with 75th-percentile verbal reasoning). All three are reasonable-sounding. All three would be defensible if the barrier were something else, which is exactly why Module 3 spends its time on barrier statements.

Knowledge check

8.1Which single piece of data in Rafael's case most strongly rules out an attention or acquisition explanation?
Answer

The 89% exit-ticket average. If he could not attend during instruction or could not acquire the material, immediate performance would be weak too. Immediate performance is strong and delayed performance is weak; that dissociation is the diagnosis, and it is the reason a plan aimed at attention during class would be aimed at the wrong interval.

8.2Rafael's teacher proposes giving him a completed study guide for every unit. Evaluate.
Short response — look for

It is a reasonable support and it is not SDI for this need. A completed guide supplies the cues rather than teaching him to generate and use them, so it produces no durable change and does not transfer to the June Regents, where no guide is available. It also risks reinforcing the rereading illusion he already has. What would make it instructional: teaching him to build the guide as a question-only retrieval index and to practice recalling before checking. Same artifact, different verdict — which is the Module 2 lesson showing up in a new case.

8.3Which taxonomy dimension does “add a day-14 retrieval and more items per session” represent?
  1. Alignment
  2. Dosage
  3. Attention to transfer
  4. Individualization
Answer · B

Dosage is opportunities to respond and receive feedback. Both changes increase the number of retrieval attempts. Note that they do not change what is being taught (alignment) or where it is practiced (transfer) — keeping those distinctions clean is what lets you attribute any change that follows.

8.4Rafael's plan moves demanding tasks to the first 30 minutes of the period. Which SDI element is that, and is it SDI or an accommodation?
Answer

Delivery — it changes when instruction is provided. Whether it counts as SDI or accommodation is a genuinely arguable call, and the argument is worth having: as a bare scheduling preference it is closer to an accommodation; as a designed feature of an instructional plan built on documented within-period performance decline, tied to a goal, and monitored, it is a delivery adaptation. The useful move is not to win the label but to make sure the plan does the thing that matters — it is written down, reasoned from data, and someone would notice if it stopped happening.

8.5Your peer's plan for Rafael includes: “Provide retrieval practice and review strategies.” Give the two most useful critique questions.
Answer

“Who teaches him the routine, and what does that lesson look like?” — the plan currently supplies a practice rather than teaching one, which is the most common failure in the whole unit. “What is the spacing schedule, and who runs it until he can?” — retrieval practice without specified intervals is a phrase, not a procedure, and the fading of who owns the schedule is the independence goal. A third, if you have room: “What is the probe and when do you look at it?”

Reflection and application

Which of the seven questions was hardest for you to answer for Rafael? That is the step where your own planning most often stops. Write it on a card and put it where you plan.

Takeaway

A defensible SDI plan is not the one with the most supports. It is the one where every element traces back to a barrier the data actually support, and where somebody is being taught something they will still be able to do when you are not there.

Section 12

Culminating performance task

One student, one upcoming lesson, one plan. This is the assessment that matters.

90–120 minutesIndividualScored on the Section 13 rubric

What you are producing

A complete SDI plan for Simone Alvarez, built on the seven-step sequence, using the planning template in Section 16. Two to four pages. It must be specific enough that a colleague who has never met her could teach Wednesday's lesson from it.

  1. Step 1Grade-level standard
  2. Step 2Learning target
  3. Step 3Disability-related barrier
  4. Step 4IEP need
  5. Step 5Specially designed instruction
  6. Step 6Student response & data
  7. Step 7Instructional adjustment

The student

Simone Alvarez

Grade 10 · Geometry (Regents) · Learning disability
Strengths
Exceptional spatial and deductive reasoning. Given a labelled diagram, she completes formal proofs at or above grade level and frequently finds a shorter path than the one taught. Constructs accurately with compass and straightedge. Fluent with algebraic manipulation.
Interests
Works in her uncle's cabinet shop on weekends; reads architectural plans easily and enjoys it. Wants to study drafting or construction management.
Grade-level expectations
Prove geometric theorems about lines, angles, triangles, and parallelograms. Apply geometric concepts in modeling situations presented in written contexts. The June Regents examination presents most items as dense verbal problem statements without an accompanying diagram.
Present levels of performance
When a problem is presented as a labelled diagram, Simone's accuracy is 88%. When the identical problem is presented as a paragraph of text with no diagram, her accuracy is 24%. Given the same text read aloud to her, accuracy is 31%. Given the text plus a diagram she did not draw, accuracy returns to 85%. Analysis of 30 incorrect items shows the errors occur before any geometry is attempted: she does not extract which objects exist, how they are related, or what is being asked. In 26 of 30, no diagram was drawn at all. Reading rate and decoding accuracy are within normal limits.
Disability-related need
Reading comprehension of dense, relation-heavy technical text: she does not build a mental or external representation of the relationships described in a written problem statement. This is a representation-building need, not a decoding or fluency need, and not a geometry need.
IEP information
Annual goal: Given a written geometry problem with no diagram, Simone will construct a correctly labelled diagram representing all stated relationships in 8 of 10 items across 3 consecutive probes, using a taught representation routine. Program: integrated co-teaching, Geometry, daily. Resource room 3x weekly. Program modification: instruction in strategies for representing information from technical text. Accommodations: extended time (1.5x); text-to-speech available; use of a formula reference sheet.
Assessment information
Reading comprehension (narrative), 48th percentile. Reading comprehension (technical/expository), 7th percentile. Decoding and fluency, 55th percentile. Nonverbal reasoning, 91st percentile. Mathematics computation, 70th percentile.
Classroom observations
Reads a word problem once, quickly, then writes an answer or writes nothing. Does not reread. Does not draw. When a classmate draws the figure, she immediately says the answer, often before the classmate has finished labelling. In the cabinet shop she reads dimensioned plans without difficulty — a diagram, not a paragraph.
Current accommodations and strategies
Extended time, text-to-speech, formula sheet, and a whole-class problem-solving poster reading “Read · Plan · Solve · Check.”
Student response data
Weekly quiz scores over eight weeks: 42%, 38%, 45%, 40%, 44%, 39%, 43%, 41%. The diagram-based portion of each quiz averages 86%; the word-problem portion averages 19%. Text-to-speech usage log shows she uses it, and her accuracy on read-aloud items is 31% — better than reading silently, and still far below her diagram performance.
What Simone says
“I can do the math. I just don't know what they're asking. By the time I get to the end of the sentence I don't remember what was at the beginning.”

The upcoming lesson

Wednesday, Geometry · 42 minutes · integrated co-teaching, 24 students, 9 with IEPs

Unit: Properties of parallelograms. Lesson objective: Apply properties of parallelograms to solve contextual problems presented in written form.

Planned lesson:

  • 0–5 min: warm-up, two diagram-based angle problems
  • 5–15 min: whole-class review of parallelogram properties; teacher works one contextual problem on the board
  • 15–33 min: independent/partner practice, six contextual word problems, no diagrams provided
  • 33–40 min: three problems reviewed as a class
  • 40–42 min: exit ticket, one contextual problem

Sample problem from the practice set: “A community garden is shaped like a parallelogram. The gardeners want to run a water line along one diagonal. The side running east–west measures 40 feet and the adjacent side measures 26 feet. The angle at the southwest corner measures 65 degrees. Determine the length of the diagonal connecting the southwest and northeast corners, to the nearest tenth of a foot.”

Staffing: You and a co-teacher. Both of you are available for the whole period. You also have Simone in resource room on Tuesday, Thursday, and Friday.

What to submit

Use the planning template in Section 16 and answer all ten planning questions. In addition, address these explicitly:

  1. The barrier. State the mechanism. Name the specific data in the case that support your statement and the data that rule out at least two competing explanations. Simone's case is deliberately constructed so that three plausible-sounding barriers are wrong.
  2. The element. Name whether you are changing content, methodology, delivery, or a combination, and say why.
  3. Wednesday, minute by minute. Map your SDI onto the 42-minute lesson as planned. You may propose changes to the lesson, but the target for all students stays the same and the period does not get longer. Say what you do, what your co-teacher does, and what Simone does.
  4. The evidence-based practice. Name it, cite a source, and explain why it fits this barrier rather than why it is generally good.
  5. The measure. Specify the probe, the schedule, the goal line, and the decision rule.
  6. The fade. What comes off first, at what criterion, and what does full independence look like on the June Regents?
  7. If she doesn't respond. Name the taxonomy dimension you would change first and second, and why in that order.
  8. Two rejected options. Name two reasonable-sounding supports you deliberately did not choose, and explain what each one would fail to address.
Ground rules
  • The Geometry standard does not change. Simone is sitting the Regents examination in June.
  • You may not remove word problems from her work. The word problems are the assessment format.
  • Every claim about Simone must point to a line in the case. If the case does not say it, do not assume it.
  • The plan must be implementable by your co-teacher without asking you a question.
A hint about the traps

Three supports will occur to almost everyone and all three are aimed at barriers Simone's data rule out. Before you commit to any support, find the number in the case that tells you whether it addresses her actual need. One of the three is already in place and already failing, and the case tells you by how much.

Section 13

Performance-task rubric

Eight criteria, 32 points. Also usable as a self-check before submitting, and as a peer critique instrument.

Criterion4 — Exemplary3 — Proficient2 — Developing1 — Beginning
1. Standard and target Grade-level standard correctly identified and unchanged. Target is observable, specific to Wednesday's content, and two observers would agree whether it was met. Standard correct and unchanged. Target is observable but somewhat general. Standard identified but the target is vague, or the target has been subtly reduced below the class expectation. No grade-level standard, or the target has been replaced with a lower one.
2. Barrier analysis States a mechanism in instructional terms, cites specific data supporting it, and explicitly rules out at least two competing explanations using the case data. States a mechanism and cites supporting data. Competing explanations not addressed, or addressed weakly. Names a general area of difficulty (“reading”) without a mechanism, or cites no data. Restates the classification, the outcome, or an attribution (“unmotivated,” “visual learner”).
3. Use of IEP information Goal, program, program modification, and accommodations all used correctly, with the plan explicitly aligned to the goal and constrained by the service delivery available. Goal and at least two other IEP elements used correctly and the plan aligns to the goal. IEP referenced generally; the plan is not clearly aligned to the goal. IEP not used, or accommodations are presented as the plan.
4. SDI design Element(s) named and justified. Instruction is explicit: what is modeled, guided, practiced independently, with the feedback language for the predictable error. The student is taught something, not given something. Element named. Instruction is explicit with modeling and practice, though one component is thin. An adaptation is described but it is provided rather than taught, or explicit instruction components are largely absent. A list of supports or accommodations presented as SDI.
5. Rigor maintained Route changes; destination does not. Grade-level demand is preserved throughout, including in the practice set and the exit ticket, with a stated path to independence on the Regents format. Grade-level demand preserved in the main task; a minor reduction appears somewhere. Expectations lowered in a way not authorized by the IEP, or the word-problem format is avoided. Task substituted or content substantially reduced.
6. Evidence base Practice named, sourced, and justified specifically against this barrier, with the fit argued rather than asserted. Practice named and sourced; the justification is general. A practice is named with no source or no reasoning about fit. No practice named, or one contradicted by evidence (e.g. learning-styles matching).
7. Measurement and decision rule Probe samples the goal behavior under goal conditions, schedule is feasible and specified, goal line and decision rule stated, and forms are equivalent. Probe and schedule specified; decision rule present but loose. Vague plan to “monitor progress” with no probe, schedule, or rule. No measurement plan.
8. Adjustment, fading, and communication Names the intensification dimensions in order with reasoning, states the fading criterion and what comes off first, defines independence on the Regents, and the plan is fully implementable by a colleague. Adjustment and fading both addressed; the plan is mostly implementable with one or two gaps. Adjustment or fading addressed but not both; a colleague would need to ask several questions. Neither addressed; the plan cannot be implemented by anyone else.

Scoring

Bands

PointsInterpretation
29–32Exemplary. Ready to lead this work with colleagues.
24–28Proficient. Independently designs sound SDI.
18–23Developing. Revise and resubmit with the specific criteria named.
Below 18Beginning. Revisit Modules 3 and 5 before resubmitting.

Two non-negotiables

A plan scoring 1 on criterion 5 (rigor maintained) cannot be scored proficient overall regardless of total points. Lowering the standard without team authorization is not a partial success.

A plan scoring 1 on criterion 4 (SDI design) cannot be scored proficient either. A list of accommodations is the specific error this unit exists to correct, and passing it would defeat the purpose.

For facilitators — the three traps

Simone's case is built so that three appealing supports are all wrong, and how a participant handles them is the fastest read on criterion 2.

  • Text-to-speech. Already in place. Read-aloud accuracy is 31% against 85–88% with a diagram. Decoding is 55th percentile. The barrier is not access to the words.
  • Extended time. Already in place, and the errors occur before any geometry is attempted — more minutes do not build a representation.
  • Providing the diagram. The most seductive one, because accuracy jumps to 85% when someone else draws it. It is a legitimate short-term bridge and it fails the June Regents, where nobody draws it for her. The distinction between supplying the representation and teaching her to construct it is the whole point of the task, and it is the same distinction as Module 2's sixteen pairs.

The strong response teaches an explicit text-to-diagram representation routine — identify the objects, mark the given relationships, label the unknown, then verify the diagram against the sentence — modeled with think-aloud on Wednesday's board problem, practiced with feedback in the co-taught period and in resource room, measured by a weekly probe scoring the diagram rather than the answer, and faded from co-constructed to independent by spring.

Section 14

Post-assessment

Fifteen items, parallel in structure and difficulty to the pre-assessment. Take it before you look at either key.

15 items~25 minutesKey in Section 15

Part A · Definition and distinctions

1. A teacher describes an adaptation as SDI. Which pair of conditions must both be satisfied for that to be accurate?

  1. It appears on the IEP, and it is provided consistently
  2. It addresses a need resulting from the student's disability, and it secures access to the general curriculum so the student can meet the standards that apply to all students
  3. It is delivered by a certified special education teacher, and it is documented
  4. It is evidence-based, and the student likes it

2. A biology teacher builds a partially completed causal-chain organizer specifically for one student after analyzing eight of her written responses, teaches her to use it over four lessons with modeling and feedback, charts her rubric scores weekly against her IEP goal, and plans to remove the middle boxes at 3/4 on three consecutive labs. This is:

  1. An accommodation, because a graphic organizer is a support
  2. Differentiation, because it responds to a student's readiness
  3. SDI, because it is need-driven, explicitly taught, measured against a goal, and has a fading criterion
  4. UDL, because organizers offer multiple means of representation

3. Which statement about modifications is accurate?

  1. A modification is any change to a lesson made for a student with a disability
  2. A modification changes what the student is expected to learn or demonstrate, and is a team decision
  3. Modifications and accommodations are interchangeable terms
  4. Modifications may only be delivered in a special class setting

4. A twelfth grader's lab reports are not submitted for multi-day investigations, though his single-session work averages 88%. He sits with the packet open and blank for 10–12 minutes at the start of each project day, then works productively once started. Which decision is best supported by that evidence?

  1. Give him a pre-broken packet with the dates already filled in
  2. Extend the due date for each investigation
  3. Explicitly teach a decomposition routine that terminates in a first physical action, with modeling, guided practice, and a written fading schedule; measure latency to initiation
  4. Assign a teaching assistant to prompt him through each step

5. In the 2024 High-Leverage Practices, which set correctly identifies the six pillar practices?

  1. HLPs 1, 3, 6, 7, 16, 20
  2. HLPs 1, 2, 3, 4, 5, 6
  3. HLPs 13, 15, 17, 18, 19, 22
  4. HLPs 4, 5, 11, 12, 14, 21

6. An intervention has produced a flat trend over seven data points. A fidelity check reveals it was delivered on 6 of 21 scheduled occasions. What should happen?

  1. Intensify on dosage, since the student clearly needs more
  2. Conclude the student requires a more restrictive placement
  3. Deliver the intervention as designed, then re-evaluate response — the student has not yet had the opportunity to respond to it
  4. Change to a different intervention

7. A tenth grader has 88% accuracy on geometry problems presented as labelled diagrams and 24% on the identical problems presented as text. Errors occur before any geometry is attempted. Which is SDI?

  1. Providing the diagram with each word problem
  2. Reading the word problems aloud to her
  3. Explicitly teaching a routine for converting a written problem statement into a labelled diagram, with modeling, practice, feedback, a weekly probe scoring the diagram, and a fading plan
  4. Extended time on all assessments

8. Which statement best describes the relationship between scaffolding and SDI?

  1. Scaffolding is a synonym for SDI
  2. Scaffolding is a component practice that appears within SDI; providing a scaffold is not by itself SDI
  3. Scaffolding belongs to general education and SDI to special education
  4. Scaffolds should remain in place permanently for students with IEPs

9. A general education chemistry teacher is asked whether she can deliver SDI. The accurate answer is:

  1. No — only certified special educators may deliver SDI
  2. Yes — general educators routinely deliver SDI, and in New York integrated co-teaching is built on it; the special educator coordinates the design and monitoring
  3. Only if she holds a dual certification
  4. Only during resource room

Part B · Sorting and matching

10. Sort each practice as described. Mark Insufficient information where the description does not let you decide, and state what you would need to know.

 As described in the classroomSDI / Accommodation / Modification / Not enough information
aA student writes three self-quiz questions at the end of each lesson, having been taught to generate them, and retrieves from them on a scheduled interval that he now sets himself.
bA student uses a personal reference notebook during assessments per her IEP.
cA teacher posts academic discussion stems on the classroom wall.
dAn IEP team determines a student will be assessed on identifying the main idea rather than analyzing author's craft.
eA student receives a partially completed outline before each essay.
fThree students with the same error pattern receive fifteen minutes of concrete-to-abstract instruction three times a week, with weekly probes charted against IEP goals.

11. Match each dimension of the Taxonomy of Intervention Intensity to the situation calling for it.

DimensionSituation
Alignment___ The student performs the routine in resource room and not in chemistry
Dosage___ Diagnostic assessment shows the plan targets fluency but the barrier is conceptual
Attention to transfer___ The academic plan is sound but the student disengages when frustrated
Comprehensiveness___ The student gets five practice opportunities in a 30-minute session
Behavioral support___ The strategy was named for the student but never modeled, and no non-examples were used

12. A student's plan includes each of the following. Label each as content, methodology, or delivery.

  1. Ten minutes of pre-teaching with the special educator before the lab, in a group of three
  2. Teaching the prerequisite concept of equality alongside equation solving
  3. Concrete–representational–abstract sequence instead of worked examples
  4. The speech-language pathologist delivering syntax instruction inside the class text
  5. A cognitive strategy taught to mastery as an object of instruction

Part C · Short response

13. A student with an IEP goal for generalizing a strategy scores 92% on familiar graph types and 30% on unfamiliar ones. His teacher proposes monitoring progress with a weekly quiz using the graphs from that week's instruction. Explain what is wrong with the measure and what should replace it.

14. A colleague says: “I run a really strong UDL classroom — multiple representations, choice in products, supports built into every task. My students with IEPs are covered.” Respond. Name what is right, what is wrong, and what you would ask to see.

15. Write an SDI plan summary for a student in your own course, in five sentences, covering: what is taught, how it is taught, who and when, the measure, and the fading criterion. It must be specific enough for a substitute teacher to implement.

Compare

Go back to the two pre-assessment items you flagged as least certain. Answer them again from memory before checking the key. The distance between the two answers is what this unit was for.

Section 15

Answer keys

Pre-assessment and post-assessment, with explanations. Module knowledge-check answers are inside each module.

Pre-assessment key

Pre 1Two purpose clauses — B
Answer · B

Both IDEA §300.39(b)(3) and 8 NYCRR §200.1(vv) state the two purposes conjunctively: address needs resulting from the disability, and ensure access to the general curriculum so the student can meet the standards that apply to all students. Option A describes remediation, which may be part of SDI but is not the definition. Option C confuses SDI with the documents that record it.

Pre 2Class-wide Cornell template — C
Answer · C

Provided identically to everyone, not determined by a documented need for this student, not taught to this student for this need, not measured. That is good universal practice. D is wrong for the opposite reason and is the more instructive error: a graphic organizer absolutely can be part of SDI, if it was built for a documented need, explicitly taught, monitored, and faded.

Pre 3Accommodations — B
Answer · B

Accommodations change access or response conditions; the standard and expectation are unchanged. A describes a modification. C is the core misconception this unit addresses. D is false — instructional accommodations operate daily, and testing accommodations are a subset with their own documentation requirements in New York.

Pre 4One-sided operations — C
Answer · C

The error pattern is conceptual, not computational, so the instruction has to target the concept. A reduces the volume of a misapplied procedure. B addresses neither the concept nor the procedure. D substitutes the placement for the instruction. Only C changes content and methodology in response to the specific error, and only C is likely to move the line.

Pre 52024 HLP domains — B
Answer · B

Collaboration; Data-Driven Planning; Instruction in Behavior and Academics; Intensify and Intervene as Needed. Option A is the 2017 edition — still widely circulating in district PD materials, which is why the item is here.

Pre 6Four points below the goal line — C
Answer · C

Diagnose before adapting. The item specifies fidelity has been confirmed, so the next step is finding out why rather than trying something else. A ignores a decision rule that has been met. B changes the measurement instead of the instruction and is the single most damaging response available. D adds an adult without changing the instruction.

Pre 7Novel graph types — C
Answer · C

The need is generalization, so the instruction must explicitly teach transfer using deliberately novel formats. A avoids the goal entirely and would produce excellent data on a skill he already has. B and D are accommodations aimed at time and attention, neither of which the data implicate.

Pre 8UDL and SDI — C
Answer · C

UDL is proactive design for predictable variability; SDI is individually determined and legally required for an eligible student. Strong UDL can reduce how much additional adaptation is needed, which is a genuine benefit, and it does not discharge the obligation. D is worth naming as wrong: both should be operating in the same room at the same time, and in an integrated co-teaching section they always are.

Pre 9Where SDI lives on a NY IEP — C
Answer · C

There is no SDI field on the New York IEP. It is assembled from present levels, annual goals, the recommended program and services, and program modifications / supplementary aids. This is why handing a content teacher only the accommodations page hands them the page with the least instructional information in the document.

Pre 10Sort
Answer

(a) Not SDI — a class-wide support; not individually determined. (b) Accommodation — conditions change, instruction does not. (c) SDI — need-driven, explicitly taught, novel-format practice, measured against the goal. (d) Insufficient information — you need to know whether the team authorized a reduced expectation (modification), whether it is a temporary acquisition-phase decision inside designed instruction (possible SDI), or whether it just happened (neither). (e) Modification — the expectation changes; note that it is only legitimate as a documented team decision. (f) Not SDI as described — same task, closer adult; no designed instruction. Compare WVDE FAQ 15.

Full credit on (d) requires marking it insufficient. Participants who confidently sort every item are demonstrating the exact habit the unit is trying to interrupt.

Pre 11HLP matching
Answer

HLP 6 → Is it working?  |  HLP 11 → What should the student learn?  |  HLP 16 → What will I do differently?  |  HLP 20 → What happens if the student still does not respond?  |  HLP 4 → What is interfering with access?

Pre 12Sequence
Answer

1 Grade-level standard → 2 Learning target → 3 Disability-related barrier → 4 IEP need → 5 Specially designed instruction → 6 Student response and data → 7 Instructional adjustment.

Pre 13Emailed notes — short response
Look for

The accommodation substitutes for the skill rather than building it, and the usage log shows it is not even substituting. Nothing about the instruction changed, so the disability-related need is untouched. SDI here would be explicitly teaching a note-taking routine matched to what his capture data show he loses — argument structure, not detail — with modeling, cued pauses to raise opportunities to respond, immediate comparison against a criterion set, self-monitoring, and weekly capture probes charted against the goal. Strong answers also note that the usage log is itself data worth bringing to the team.

Pre 14“Everyone benefits, so I'm already doing SDI”
Look for

Right: explicit vocabulary instruction is well supported by evidence, benefits nearly all students, and the fact that other students benefit from a practice never disqualifies it as SDI — WVDE says this directly. Wrong: a practice delivered identically to everyone is not individually determined, is not selected against a documented need, and is not monitored against a goal. The strongest answers add the constructive move: the same practice becomes SDI when the terms are selected for a specific student's barrier, taught with the components that student needs, and probed.

Pre 15“She's in a special class”
Look for

Placement is a delivery decision under §200.6 about the intensity of service and the least restrictive environment; it does not determine which standards apply. The student is enrolled in Regents Living Environment, so the Living Environment standards apply, and substituting a simplified summary fails the second purpose clause of the SDI definition and is inconsistent with Endrew F. What should happen instead: identify the specific barrier to the assigned text — syntax, vocabulary, background knowledge, or decoding are different problems with different instruction — and teach against it while keeping the grade-level text. Strong answers note that a substitution made on a Tuesday without team authorization is an undocumented modification.

Post-assessment key

Post 1Two conditions — B
Answer · B

The two-condition test. A confuses documentation with definition — being on the IEP does not make something SDI, or every accommodation would be. C confuses who delivers with what it is.

Post 2The built organizer — C
Answer · C

All four markers are present: built from analysis of this student's work, explicitly taught, measured against the IEP goal, and with a written fading criterion. Same artifact as pre-assessment item 2, opposite verdict, which is the point of the pairing.

Post 3Modifications — B
Answer · B

A modification changes what is expected, and it is a team decision rather than a Tuesday improvisation. D is a common and consequential error — a modification may be delivered in a general education setting where the IEP calls for it, and assuming otherwise drives unnecessary segregation.

Post 4Blank packet — C
Answer · C

The data locate the failure at initiation, and the routine must therefore terminate in an action he can start within seconds. A supplies the decomposition rather than teaching it. B gives more time to a student who does not have a time problem. D produces prompt dependence and is directly counter to an independence goal.

Post 5Pillar practices — A
Answer · A

HLP 1 and 3 (Collaboration), HLP 6 (Data-Driven Planning), HLP 7 and 16 (Instruction in Behavior and Academics), HLP 20 (Intensify and Intervene as Needed).

Post 66 of 21 occasions — C
Answer · C

This is a fidelity failure, not a non-response. The student has not had the opportunity to respond to the plan, so nothing about the plan has been tested. A, B, and D all treat the flat line as information about the student, and B in particular would move a student to a more restrictive setting on the basis of an adult scheduling problem.

Post 7Diagram vs. text — C
Answer · C

A supplies the representation and fails at the Regents, where nobody supplies it. B addresses decoding, which the data rule out. D addresses time, which the data also rule out. Only C teaches the student to build the representation herself, which is the documented need.

Post 8Scaffolding — B
Answer · B

HLP 15 describes scaffolds as calibrated, flexible, evaluated, and gradually removed once no longer needed. D contradicts the practice as defined, and describes the most common way a scaffold turns into a permanent dependency.

Post 9Who delivers SDI — B
Answer · B

General educators, related service providers, and trained paraprofessionals under supervision all implement SDI. Under §200.6 integrated co-teaching is explicitly a setting where specially designed instruction is delivered alongside academic instruction. The special educator's distinctive role is designing, coordinating, and monitoring it.

Post 10Sort
Answer

(a) SDI — taught, self-generated, scheduled, and being faded to independence. (b) Accommodation. (c) Not SDI — a poster; class-wide, untaught. (d) Modification — the expectation changes by team decision. (e) Insufficient information — you need to know whether it was built for a documented need, whether the student was taught to complete it, whether it is measured, and whether it is being faded. (f) SDI — grouped by error pattern, designed sequence, probed against goals.

Post 11Taxonomy matching
Answer

Attention to transfer → performs in resource room, not chemistry. Alignment → targets fluency, barrier is conceptual. Behavioral support → disengages when frustrated. Dosage → five opportunities in 30 minutes. Comprehensiveness → named but never modeled, no non-examples.

Post 12Content, methodology, or delivery
Answer

(a) Delivery. (b) Content. (c) Methodology. (d) Delivery. (e) Content and methodology — the strategy is a thing learned and teaching it explicitly is how instruction is delivered. Item (e) is the one to discuss; participants who insist on one label are missing why strategy instruction is so useful at secondary level.

Post 13Wrong measure — short response
Look for

The proposed measure samples familiar formats, where he is already at 92%, so it cannot detect change in either direction — a measure that cannot go down cannot go up. The goal specifies novel formats, so the probe must use graph types he has not been taught. Replacement: four interpretation items weekly on a graph type he has never seen, scored against the 80% criterion, with the trend judged over points rather than by the latest score.

Post 14“Strong UDL, so my students with IEPs are covered”
Look for

Right: proactive design genuinely reduces barriers, likely reduces how much individual adaptation is needed, and is exactly what should be happening first. Wrong: nothing in it is individually determined, tied to a documented need, or monitored against an IEP goal, and the obligation is individual. The diagnostic question worth asking is would this lesson be identical if those students were not enrolled? If yes, no SDI has occurred. What to ask to see: for one student, the barrier statement, what is being explicitly taught, the probe, and the fading criterion. Strong answers keep the tone collegial — the colleague is doing something good and is one step short, not wrong.

Post 15Five-sentence plan — scoring guide
Score 1 point each; 5 total
  1. What is taught — a named routine, concept, or skill, specific enough to reproduce. Not “strategies.”
  2. How — includes modeling and practice with feedback, and ideally the correction language for the predictable error.
  3. Who and when — person, setting, frequency, duration, group size.
  4. The measure — a probe that samples the goal behavior, with a schedule.
  5. The fading criterion — a performance level and what comes off first.

The substitute-teacher test applies: if a colleague would have to ask a question before starting, the corresponding point is not earned.

Section 16

Job aids and templates

Ten tools built to be printed and used at a planning table. Each one fits on a page.

These are the part of the unit that outlives it. Print them, put them in the folder you actually open, and use them during planning rather than after it. Use your browser's print function; the page is styled so each aid prints cleanly.

1 · What counts as SDI?

The one-page test

Job aid 1
The definition8 NYCRR §200.1(vv)
Adapting, as appropriate to the needs of an eligible student, the content, methodology, or delivery of instruction to address the unique needs that result from the student's disability, and to ensure access to the general curriculum so the student can meet the standards that apply to all students.
Condition 1Must be yes
Would I be doing this for this student if the documented disability-related need were not there?
Condition 2Must be yes
Does this move the student toward the grade-level standard rather than away from it?

The four markers

If any one is missing, it is a support, not SDI.

  • Caused — chosen because of a specific documented need, not because it is generally good
  • Taught — somebody explicitly teaches the student to do it: modeled, practiced, corrected
  • Measured — a probe samples the goal behavior on a schedule, charted against a goal line
  • Fading — a written criterion says what comes off, when

Three questions when you are stuck

  1. Provided or taught? If the student was handed it and nobody taught them to use it, it is not yet SDI.
  2. Would this lesson be identical if this student weren't enrolled? If yes, nothing has been specially designed.
  3. What comes off first, and when? If you cannot answer, the support has no exit and will still be there in June.

2 · SDI vs. accommodations vs. modifications vs. differentiation vs. UDL

 PurposeWho receives itWhat changesWho decidesLitmus question
SDIAddress a disability-related need and secure grade-level accessOne student, individually determinedContent, methodology, or delivery of instructionIEP team designs; many implementWas this caused by a documented need, taught, measured, and is it being faded?
AccommodationEqualize access or response conditionsOne studentConditions. Standard and instruction unchanged.IEP teamDoes the expectation stay exactly the same?
ModificationChange what is expected when the standard itself is not attainable as writtenOne studentThe expectation — standard or criterionIEP team, documentedDid a team decide this, and is it written down?
DifferentiationRespond to the range of readiness, interest, and profile in the roomGroups, fluidlyContent, process, or product across a classClassroom teacherWould I do this for a group regardless of any IEP?
UDLReduce predictable barriers by design, in advanceEveryoneThe design of the lesson before students arriveWhoever designs the lessonWas this decided before I knew who was enrolled?
InterventionRemediate a targeted skill deficitAnyone meeting criteria, IEP or notSupplemental instruction on a specific skillSchool team, by data rulesDid the student get this by meeting a threshold rather than by an IEP decision?
ScaffoldBridge the gap between the task and current performanceAnyone, situationallyTemporary support inside a taskWhoever is teachingIs it calibrated, evaluated, and coming off?
Say this out loud in meetings

These are not alternatives. A good lesson has all of them running at once. The error is counting one of them as SDI and concluding the obligation has been met.

3 · SDI planning template

Ten questions, ten decisions

Job aid 3 · one student, one unit
What should the student learn?Grade-level standard / learning target
Write the standard, then the target so two observers would agree whether it was met.
What can the student currently do?Baseline / performance data
A number, and the conditions it was collected under.
What is interfering with access or progress?Disability-related barrier
A mechanism, not a label. What does it predict? What would rule it out?
What does the IEP identify?Need / goal
Goal, program and service, program modification, accommodations.
What needs to be intentionally changed?Content, methodology, and/or delivery
Name the element(s). Circle: content  ·  methodology  ·  delivery
What will the teacher do differently?Specially designed instruction
What is modeled, what is guided, what is independent, and the correction language for the predictable error.
What will the student do?Observable learning behavior
The thing you would see them doing. Not “engage with.”
How will progress be measured?Data / evidence
Probe, schedule, goal line, where the data live.
Is it working?Student response
Decision rule, agreed in advance. Fidelity check before any conclusion.
What happens next?Maintain, fade, modify, or intensify
If intensifying, name the dimension. If fading, name what comes off first and at what criterion.

4 · IEP-to-instruction planning tool

Use this at the start of a marking period, one page per student. It converts four IEP sections into four instructional decisions.

Four passes through the IEP

Job aid 4
Pass 1 — Present levelsRead for the barrier and the baseline
What performance is described, under what conditions?
What is the number I will measure against?
If it says nothing usable, what will I collect myself this week?
Pass 2 — Annual goalsRead for the aim and the criterion
What is the condition clause? What routine or strategy is named?
What counts as success, and by when?
How does this goal connect to my course's standards?
Pass 3 — Program and servicesRead for delivery
Program type, frequency, duration, location, group size.
Related services, including indirect time — what is that time for?
Who else is teaching this student, and when could we co-plan?
Pass 4 — Modifications, aids, accommodations, supports for personnelRead for permission and obligation
What content changes are authorized?
What must I provide, every time, without being asked?
What training or consultation is written in for adults — and am I getting it?
The outputOne sentence
“Because ______ [barrier], I will teach ______ [what] by ______ [how], ______ [when/where/who], and measure it with ______ [probe], fading at ______ [criterion].”

5 · SDI classroom observation checklist

For coaches and administrators. Built for a coaching conversation, not an evaluation rating. Observe one student, not the room.

What you should be able to see

  • The target is stated and is observable
  • Explicit modeling occurs, with the thinking made audible
  • Non-examples are used deliberately, not incidentally
  • Guided practice occurs with feedback at the point of error
  • The observed student produces frequent observable responses — count them
  • Feedback is specific, task- or process-level, and tied to a named component
  • The student uses a taught routine, with or without a cue
  • Adult prompting is at a defined level rather than continuous
  • Grade-level demand is intact for the observed student
  • Some form of data is collected during or at the end of the lesson

What you should be able to ask afterward

  • What barrier is this instruction aimed at, and what data support that?
  • Which element are you adapting — content, methodology, or delivery?
  • What is the current support level, and what comes off next?
  • Where is the progress-monitoring chart?
  • Could your co-teacher run this tomorrow without asking you anything?
  • What would make you change the plan?

Tally box for opportunities to respond, observed student, 20 minutes: ______

A caution for observers

Most of what makes an adaptation SDI is invisible in a 20-minute observation. The lesson looks the same whether or not there is a barrier statement, a probe, and a fading criterion behind it. The observation tells you about delivery; the conversation afterward tells you about design. Do not score the first without the second.

6 · SDI lesson-planning checklist

Before the lesson

  • Grade-level target written and observable
  • Task demands listed — including the ones I usually assume
  • For each identified student: which demand is the broken one?
  • Element named: content / methodology / delivery
  • Think-aloud script written for the step the student skips
  • Two non-examples chosen and the misconception each prevents
  • Target OTR count for the identified student, and how I will get it
  • Correction language drafted for the most likely error
  • Current support level noted; fading criterion known
  • Probe identified — ideally the exit ticket
  • Co-teacher / TA role specified, not assumed

After the lesson

  • Did the designed instruction actually happen? On how many of the planned occasions this week?
  • Did I model, or did I hand it over?
  • Was feedback immediate, or on returned work?
  • Did the student produce the observable behavior?
  • Data recorded where the team can see it
  • Has the fading criterion been met without my noticing?
  • What is the one thing I would change tomorrow?

7 · Data-based individualization decision tree

  • Start: you have four or more progress-monitoring data points.

    Look at the trend, not the last point.

  • Q1. Is the trend meeting or exceeding the goal line?
    Yes → Maintain. Then check: has the fading criterion already been met? If performance is at ceiling across several points, fade a layer and raise the demand. A ceiling is a signal, not a success.
    No → Go to Q2. Do not add supports yet.
  • Q2. Is the data pattern highly variable with no discernible trend?
    Yes → Fix the conditions before judging the student. Check probe equivalence, who delivered it, on which days, and whether the period was interrupted. Re-collect four points.
    No, it is consistently flat or below → Go to Q3.
  • Q3. Was the instruction delivered as designed?

    Count occasions. Check explicitness, feedback timing, and consistency across adults.

    No → Stop. This is a fidelity finding, not a student finding. Deliver it as designed for four weeks, then re-evaluate. Do not intensify. Do not refer.
    Yes → Go to Q4.
  • Q4. Do you know why it isn't working?
    No → Assess diagnostically. Error analysis, brief skills inventory, FBA for behavior. Do not skip this step — skipping it is what produces years of trying things.
    Yes → Go to Q5.
  • Q5. Which single dimension does the diagnostic point at?

    Alignment — instruction targets the wrong skill · Comprehensiveness — explicit instruction components missing · Dosage — too few opportunities to respond and receive feedback · Attention to transfer — works in one setting only · Strength — the program itself lacks evidence · Behavioral / academic support — the other domain is interfering · Individualization — a protocol is being run where a tailored plan is needed

    Change one. Document what you changed and when. Continue monitoring for at least four points before judging.
  • Q6. Still flat after two well-implemented, well-targeted adaptations?
    → Convene the team. Bring the graph, the fidelity record, the diagnostic findings, and the adaptations tried. The question for the team is whether the goal, the service, or the program needs to change — and that is a team decision, not a Tuesday one.

8 · High-leverage practices quick reference

2024 second edition. Pillar practices in bold.

HLPPracticeDomainWhere it shows up in SDI planning
1Collaborate with professionals to increase student successCollaborationCo-planning with the content teacher, SLP, and TA
2Organize and facilitate effective meetings with professionals and familiesCollaborationCSE meetings; the co-planning conversation
3Collaborate with families to support student learning and secure needed servicesCollaborationWhat the family knows about how the student learns and what motivates them
4Use multiple sources of information to develop a comprehensive understanding of a student's strengths and needsData-Driven PlanningStep 3 — barrier analysis
5Interpret and communicate assessment information to collaboratively design and implement educational programsData-Driven PlanningStep 4 — making data usable by the team
6Use student assessment data, analyze instructional practices, and make necessary adjustments that improve student outcomesData-Driven PlanningStep 6 — including the fidelity check on your own teaching
7Establish a consistent, organized, and responsive learning environmentInstructionThe precondition for everything else
8Provide positive and constructive feedback to guide students' behaviorInstructionStep 5 — feedback on the routine, not the person
9Teach social behaviorsInstructionWhat to teach, when the barrier is social or behavioral
10Conduct functional behavioral assessments to develop individual student behavior support plansIntensify & InterveneStep 3 — barrier analysis for interfering behavior
11Identify and prioritize long- and short-term learning goalsData-Driven PlanningSteps 1–2 — standard to target
12Systematically design instruction toward a specific learning goalData-Driven PlanningSequencing across lessons
13Adapt curriculum tasks and materials for specific learning goalsInstructionStep 5 — the content element
14Teach cognitive and metacognitive strategies to support learning and independenceInstructionStep 5 — content and methodology; the workhorse of secondary SDI
15Provide scaffolded supportsInstructionStep 5 — calibrated, evaluated, removed
16Use explicit instructionInstructionStep 5 — the core methodology of SDI
17Use flexible groupingInstructionStep 5 — the delivery element; group by error pattern, not schedule
18Use strategies to promote active student engagementInstructionStep 5 — opportunities to respond, counted
19Use assistive and instructional technologiesInstructionStep 5 — delivery, and only when the tool matches the barrier
20Provide intensive intervention for academics and behaviorIntensify & InterveneStep 7 — when response is inadequate
21Teach students to maintain and generalize new learning across time and settingsInstructionStep 5 — the most skipped practice in the framework
22Provide positive and constructive feedback to guide students' learningInstructionStep 5 — immediate, specific, goal-directed

9 · SDI examples for secondary classrooms

Each row is a barrier, not a disability category. Find the barrier, then read across. These are starting points for design, not a menu to select from — the design still has to be built for the student in front of you.

BarrierElementSDIEvidence baseMeasure
Cannot organize a multi-part written argument; content available orallyMethodologySelf-regulated strategy development for written argument: memorize the routine, model with a repair, collaborative write, independent; organizer faded in three layersStrategy instruction for writing; HLP 14, 16Weekly paragraph on a novel prompt, department rubric
Procedures applied without the underlying conceptContent + methodologyConcrete → representational → abstract sequence with a spoken justification at each transformation; deliberate non-examples analyzedCRA in mathematics; HLP 13, 16Weekly probe scored twice: accuracy and valid justification
Loses argument structure across extended oral inputMethodology + deliveryTwo-column routine (argument / detail) taught explicitly; two cued pauses; comparison to a criterion set; self-graphingNote-taking strategy instruction; HLP 16, 18, 22Percent of criterion content captured, weekly
Strategy does not transfer to novel formatsMethodologyDeliberate practice across unfamiliar formats with explicit “same routine” statements; practice in a second courseGeneralization instruction; HLP 21Weekly items on a format never taught
Cannot initiate or decompose multi-day workContent + methodologyDecomposition routine taught to independence, terminating in a first physical action; systematic prompt fading on a written scheduleSelf-regulation and self-management; HLP 14Latency to initiation; percent of dated steps on time
Complex syntax not parsed in reading or discussionMethodology + deliverySentence deconstruction taught on the class text; discussion moves taught to mastery and faded; SLP indirect time used to co-planSyntactic awareness instruction; HLP 15, 1610 sentences weekly from the actual reading
Escape from evaluative demand at task onsetMethodology + deliverySelf-regulation routine taught outside the moment of demand; graduated ladder of evaluative pressure; check-ins on a thinning fixed scheduleSelf-regulation; FBA-informed; HLP 10, 14Percent of evaluative tasks begun within 2 minutes
Acquires normally, cannot retrieve after a delayContent + methodologySelf-generated retrieval questions taught explicitly; spaced retrieval at day 1/3/7/21; the rereading illusion demonstrated deliberately; reference notebook built as a cue indexSpaced retrieval practice; HLP 14, 21Free-recall probe on content taught 7 days earlier
Cannot build a representation from relation-heavy technical textMethodologyText-to-diagram routine taught explicitly: identify objects, mark relationships, label the unknown, verify against the sentenceSchema-based instruction; HLP 14, 16Weekly probe scoring the diagram, not the answer

10 · Is this actually SDI? — decision tree

  • Start with the practice you are about to use, or already use.
  • Q1. Does it change the content, methodology, or delivery of instruction?
    No — it changes conditions of access or response (time, seat, format, tool) → This is an accommodation. Keep it if it is on the IEP or needed. Do not count it as SDI. Now ask: what instruction addresses the underlying need?
    Yes → Go to Q2.
  • Q2. Does it change what the student is expected to learn or demonstrate?
    Yes → This is a modification. Legitimate only as a documented IEP team decision. If a team did not decide it, stop and bring it to the team — an undocumented modification is the most common way rigor quietly disappears.
    No, the expectation is unchanged → Go to Q3.
  • Q3. Was it chosen because of a specific documented disability-related need for this student?
    No — everyone gets it, or it is good general practice → This is UDL, differentiation, or good teaching. Valuable. Not SDI. It may become SDI if you now design its use around a documented need.
    Yes → Go to Q4.
  • Q4. Is the student being taught to use it — modeled, practiced, corrected — or just given it?
    Given → Not yet SDI. This is the most common gap in the whole framework. Add explicit instruction and you have crossed the line.
    Taught → Go to Q5.
  • Q5. Is it measured against the IEP goal on a schedule?
    No → Not yet SDI. Without data you cannot tell whether it is working, and you have no basis for the next decision. Add a probe.
    Yes → Go to Q6.
  • Q6. Is there a written criterion for fading it?
    No → SDI, at risk. It meets the definition now and will become a permanent prosthetic by default. Write the criterion today.
    Yes → This is SDI. Caused, taught, measured, and built to be removed.
Section 17

Facilitator implementation guide

Pacing options, session plans, the conversations that will go wrong, and how to handle them.

Three ways to run this

FormatStructureHow it works
Full-day institute × 2Two 6-hour days, ideally 4–6 weeks apartDay 1: pre-assessment, Modules 1–4. Between days: participants run the Module 3 reflection with a real student and bring data. Day 2: Modules 5–8, performance task started in session, submitted within two weeks.
Eight sessions75–110 min each, across a semesterOne module per session in a PLC or department meeting. Strongest option, because the between-session application is where the learning actually consolidates. Assign each module's reflection as the between-session task and open the next session with it.
AsynchronousSelf-paced, 12–16 hoursParticipants work through the page independently. Knowledge checks are self-scoring. Two things must stay live: the Module 7 substitute test needs a real partner, and the performance task needs a human scorer. Pair participants at the start and require one 30-minute call for each.
If you only have three hours

Run Module 1 (30 min, compressed), the “Is This Actually SDI?” section of Module 2 with the flip-the-card activity (60 min), Module 3's competing-hypotheses activity (45 min), and Module 7's substitute test (30 min). Send the rest as follow-up. Those four pieces carry most of the shift in practice. What you lose is Module 6, which is the part people most need and are least likely to complete alone — schedule it separately rather than dropping it.

Session plans for the eight-session model

SessionOpening (10 min)Core (45–70 min)Close (15 min)
1Pre-assessment, silently, no discussionModule 1 content; element sort activity in pairsCollect the two flagged items from each participant. Do not review answers.
2Share one accommodation from your caseload with no instruction behind itThe six neighbours; the sixteen pairs read in table groups; flip the cardChart the four markers. Assign: audit one student's accommodation page.
3Report from the accommodation auditBarrier analysis; competing hypotheses activityAssign: collect 15 errors from one student before session 4.
4Sort your 15 errors into factual / procedural / conceptualHLP framework; HLP audit of a real lessonEach participant names one pillar to strengthen and one IRIS module to complete.
5Watch one CEEDAR HLP 16 video togetherExplicit instruction sequence; lesson redesign with peer critiqueEveryone reads their think-aloud script aloud. This is the moment the module lands.
6Bring one progress-monitoring graph, real or inventedDBI cycle; the four-round decision simulationEveryone writes a decision rule for one student and puts four dates in a calendar.
7Task demand analysis of your hardest taskSeven students, seven Tuesdays; the substitute testRewrite one plan against the five-item documentation list.
8Post-assessmentRafael Ortiz design challenge with peer critique protocolLaunch the performance task. Due in two weeks; scored on the Section 13 rubric.

Grouping

  • Mix roles deliberately. A general educator, a special educator, and a TA at the same table produces better analysis than three of any one. The TA usually knows things about the student the other two do not.
  • Content-alike for Module 5 only. The lesson redesign needs subject partners.
  • Cross-content for Module 7's substitute test. The whole point is that the reader cannot fill gaps from subject knowledge.
  • Administrators participate, they do not observe. A principal at a table doing the sort learns more than a principal at the back of the room, and it changes what teachers are willing to say.

Six conversations that will happen, and what to do

“This is just good teaching. Why does it need a special name?”

Do not dismiss this. It is a serious objection and it is half right — the practices really are good teaching, and the 2024 HLPs say explicitly that they work for all students. Concede that fully, then move to the part that is different: SDI is an individually determined, legally required response to a documented need, monitored against a goal, and provided whether or not the class needs it. The diagnostic question that usually ends the argument productively: would this lesson be identical if that student were not enrolled? If yes, it is good teaching, and the student is still owed something more.

“I don't have time for this.”

Also true, and also not an argument against designing well. Two responses. First, most of what this unit describes occupies five to twelve minutes inside an existing lesson — go to the “forty-two minutes” section of Module 7 and work an actual schedule with them. Second, be honest that undesigned instruction is not free: the time currently going into responsive help, reteaching, and re-explaining is real time, and it produces less. Do not promise the workload drops. Promise it moves from reaction into design.

“Our IEPs don't say any of this.”

Very common, and the participants are usually right. Hold two things at once: teach the student now against your best barrier hypothesis, and bring the data to the team so next year's present levels are usable. The Module 3 reflection is built for this. If a whole department reports the same problem, that is a system finding and it belongs with the special education administrator, not with the teachers in the room.

“So I'm doing it wrong?”

Watch for this in Module 2, particularly among veteran teachers who recognize their own practice in the “not SDI” column. Say clearly and early that every practice in that column is defensible and most are helpful; what the module is separating is helpful from specially designed. Nobody in the room invented the confusion — it is built into how the field talks. The frame that works is you are one design decision away, not you have been doing it wrong.

“But he really can't do grade-level work.”

Slow down here. Distinguish three claims that get collapsed: cannot do it today unaided (true of most students, and the reason instruction exists), cannot do it with designed instruction (an empirical claim requiring evidence that designed instruction was tried), and should not be expected to (a team decision requiring documentation). Ask which one is being made. Most of the time the honest answer is the first, and the conversation becomes productive immediately.

“What about students with significant cognitive disabilities?”

A fair question, and worth answering directly rather than deferring. Everything in this unit applies; what changes is which standards apply. Students assessed against alternate achievement standards still have grade-level content, still deserve explicit instruction, still need generalization planned, and still need data. Two cautions: this unit's examples are deliberately drawn from students working at or near grade-level standards, because that is the population most often under-served by low expectations; and the decision to work from alternate standards is a team decision with specific criteria, never a shortcut reached because a task looked hard.

Materials to have ready

  • Printed copies of job aids 1, 2, 3, 7, and 10 for every participant. Do not project them — they are meant to be written on.
  • The HLP pillars chart, printed, one per person.
  • Participants bring: one real lesson plan (Modules 4, 5), one real IEP or de-identified excerpt (Module 3), and their current SDI documentation for one student (Module 7).
  • Chart paper for the Module 2 flip-the-card debrief.
  • For the Module 6 simulation, reveal rounds one at a time — printed strips or a slide per round.

Scoring the performance task

  • Two scorers where possible, and reconcile differences on criteria 2 and 4 specifically. Those two carry the most disagreement and the most learning.
  • Score criterion 2 first. If the barrier statement is a label, the rest of the plan usually cannot be strong regardless of how polished it is, and telling the participant this first makes the rest of the feedback land.
  • Revise and resubmit is the default for anything below proficient. This is professional learning, not certification, and a second attempt after specific feedback is where most of the growth occurs.
  • Return the three traps from Section 13 with every score. Participants who chose one of them need to see why, and participants who avoided them should know they did something hard.

Following through

A unit like this changes practice only if something happens after it. Three low-cost structures, in order of impact:

  1. One SDI plan reviewed per department meeting. Fifteen minutes, using the peer critique protocol from Module 8. Rotate. This is the single highest-yield follow-through.
  2. Coaching cycles using the Section 16 observation checklist, with the explicit rule that the post-conversation carries as much weight as the observation. Coaches should be trained on why an SDI observation cannot be scored on what is visible alone.
  3. A shared progress-monitoring folder per grade team. Not for compliance — so that the four-consecutive-points rule triggers a conversation rather than sitting in one teacher's notebook.
What success looks like six months out

Not that everyone can recite §200.1(vv). That teachers open a lesson plan next to an IEP and ask, without being prompted, what is the barrier and what am I teaching this student to do about it? — and that when the answer is “nothing yet,” they notice.

Section 18

References

Everything cited, with links and dates. Legal requirements are marked; everything else is research or professional guidance.

Legal requirements

  • Individuals with Disabilities Education Act regulations. 34 CFR §300.39, Special education. law.cornell.edu/cfr/text/34/300.39. Federal requirement. The source of the SDI definition at (b)(3).
  • New York State Regulations of the Commissioner of Education, Part 200. §200.1 Definitions — subdivision (vv) specially designed instruction, (ww) special education, (bbb) supplementary aids and services. nysed.gov/special-education/section-2001-definitions. New York requirement.
  • 8 NYCRR §200.4 Procedures for referral, evaluation, IEP development, placement and review — IEP content at (d)(2), transition planning at (d)(2)(ix), beginning with the first IEP in effect at age 15. nysed.gov. New York requirement; more stringent than the federal age-16 floor.
  • 8 NYCRR §200.6 Continuum of services — consultant teacher, resource room, integrated co-teaching, special class, and the associated group size and frequency requirements. nysed.gov/special-education/section-2006-continuum-services. New York requirement.
  • NYSED, Continuum of Special Education Services for School-Age Students with Disabilities. nysed.gov. State guidance interpreting §200.6.
  • Endrew F. v. Douglas County School District RE-1, 580 U.S. 386 (2017). Holding that an IEP must be reasonably calculated to enable a child to make progress appropriate in light of the child's circumstances. Binding federal case law.

New York State guidance

  • NYSED. Blueprint for Improved Results for Students with Disabilities. nysed.gov/special-education/blueprint-improved-results-students-disabilities. Seven principles; Principle 3 addresses SDI directly. State guidance, not regulation.
  • NYSED. Turnkey Guidance for Developing a Standards-Based IEP. PDF. The seven-step standards-based IEP process, gap analysis, and “point of access.” State guidance.
  • Narcisse, A., & Leone, V. Understanding Specially Designed Instruction. NYS Regional Special Education Technical Assistance and Support Center, presented at the Next Generation Learning Standards Conference, November 2017. PDF. Source of the content / methodology / delivery breakdown in New York terminology. Note the 2017 date; the regulatory definitions it cites are unchanged.

High-leverage practices

  • McLeskey, J., et al. High-Leverage Practices for Students with Disabilities (2nd ed.). Council for Exceptional Children & CEEDAR Center, March 2024. Free PDF, 221 pp. All HLP descriptions quoted in this unit are from this edition. Replaces the 2017 first edition.
  • HLP Pillars chart. CEEDAR Center, 2024. PDF. The four domains, six pillars, and sixteen embedded practices on one page.
  • CEEDAR Center. High-Leverage Practices. ceedar.education.ufl.edu/high-leverage-practices-2. Video hub, faculty guides, leadership guides.
  • High-Leverage Practices. highleveragepractices.org. The companion site maintained by CEC and CEEDAR.
  • IRIS Center. High-Leverage Practices. iris.peabody.vanderbilt.edu/resources/high-leverage-practices. Interactive alignment tool mapping IRIS resources to each HLP.
  • Clausen, A., Reyes, E. N., Wakeman, S., & Collins, B. High-Leverage Practices (TIES Center TIPS Series, Tip #9). University of Minnesota, TIES Center, 2020. Summarizes the HLPs most critical to inclusive practice and SDI.

Specially designed instruction — other state guidance

Used in this unit for the clarity of its framing. None of it is New York requirement.

  • West Virginia Department of Education. Specially Designed Instruction Guidance Document. August 2025. PDF, 25 pp. The elements of SDI (p. 2), myth-busting (p. 3), what SDI is and is not (p. 4), roles and co-teaching (pp. 8–10), twelve evidence-based strategies with research citations (pp. 11–18), and the FAQ (pp. 20–21).
  • WVDE. Additional SDI Resources. wvde.us. EBP Spotlights on explicit instruction, scaffolding, and corrective feedback; the SDI planning template; the SDI vs. DI vs. UDL comparison.
  • WVDE. Specially Designed Instruction Planning Template. PDF. Collaborative instructional design, collaborative delivery, and an SDI design checklist.
  • Kentucky Department of Education. Specially Designed Instruction Resources. education.ky.gov.
  • Maryland State Department of Education. Success for All Students in the General Education Classroom: A Guide for Inclusive Practices. 2024. Source of the guiding questions adapted in WVDE's guidance and in Module 2 of this unit.
  • Texas Education Agency & ESC Region 20. Specially Designed Instruction: A Resource for Teachers. 2022.
  • PROGRESS Center at the American Institutes for Research. IEP Tip Sheet: What Is Special Education? A Focus on Specially Designed Instruction. 2024. promotingprogress.org. Source of the content / methodology / delivery table structure used in WVDE's guidance.

Instructional practice and intensification

  • National Center on Intensive Intervention. Data-Based Individualization. intensiveintervention.org/data-based-individualization. The five-step DBI process.
  • NCII. Taxonomy of Intervention Intensity. intensiveintervention.org. The seven dimensions: strength, dosage, alignment, attention to transfer, comprehensiveness, behavioral or academic support, and individualization.
  • Fuchs, L. S., Fuchs, D., & Malone, A. S. The Taxonomy of Intervention Intensity. TEACHING Exceptional Children, 50(1), 2017. The peer-reviewed source for the taxonomy.
  • Riccomini, P. J., Morano, S., & Hughes, C. A. Big Ideas in Special Education: Specially Designed Instruction, High-Leverage Practices, Explicit Instruction, and Intensive Instruction. TEACHING Exceptional Children, 50(1), 20–27, 2017. doi.org/10.1177/0040059917724412. The argument that these four are one connected system.
  • IRIS Center. Intensive Intervention (Part 1): Using Data-Based Individualization to Intensify Instruction. iris.peabody.vanderbilt.edu/module/dbi1. Page 7, “Modify Delivery of Instruction,” is directly applicable to Module 6.
  • IRIS Center. Explicit, Systematic Instruction. Module page. Video-illustrated model of the sequence used in Module 5.
  • IRIS Center. Mathematics: Identifying and Addressing Student Errors. Case study unit. Assigned reading for Module 3.
  • IRIS Center. Universal Design for Learning: Creating a Learning Environment that Challenges and Engages All Students. Module.
  • Foxworth, L. L., Hashey, A. I., Dexter, C., Rasnitsyn, S., & Beck, R. Approaching Explicit Instruction Within a Universal Design for Learning Framework. TEACHING Exceptional Children, 54(4), 268–275, 2021. doi.org/10.1177/00400599211010190.
  • Hart, E. J., Doyle, L., Cantero, C., & Garrington, F. O. Practice Guide: Teaching Self-Regulation Skills to Students with Disabilities (K–12). U.S. Department of Education, Office of Special Education Programs, 2022.
  • Welch, M. The PLEASE Strategy: A Metacognitive Learning Strategy for Improving the Paragraph Writing of Students with Mild Learning Disabilities. Learning Disability Quarterly, 15, 119–127, 1992. One of the strategy-instruction studies underlying the written-expression SDI in this unit.
  • Bone, E., Bouck, E., & Witmer, S. Evidence-Based Systematic Review of Literature on Algebra Instruction and Interventions for Students with Learning Disabilities. Learning Disabilities: A Contemporary Journal, 19(1), 1–22, 2021. Evidence base for the concrete–representational–abstract sequence.
  • Kang, S. H. K. Spaced Repetition Promotes Efficient and Effective Learning. Policy Insights from the Behavioral and Brain Sciences, 3(1), 12–19, 2016. doi.org/10.1177/2372732215624708. Evidence base for the retrieval schedule in Rafael's plan.
  • TIES Center. The 5-15-45 Tool: Linking Available Time with Meaningful Collaboration. University of Minnesota, 2020. publications.ici.umn.edu. Co-planning structures for teams with no common planning period.
  • Texas Education Agency. Guidelines for Co-Teaching in Texas. 2021. Source of the co-teaching role comparison adapted in WVDE's guidance.
On citation integrity

Every quotation in this unit is taken from the source named beside it. Where a claim is professional judgment rather than a sourced finding — the four markers, the barrier-statement criteria, the sixteen paired examples, the seven fictional students — the unit says so or the framing makes it evident. Where the evidence is contested, Section 2 says which way and why. If you find a citation here that you cannot verify at the link given, treat that as a defect and correct it before using the material with colleagues.


Regulatory citations verified against NYSED and Cornell LII, August 2026. High-leverage practice wording quoted from the second edition (March 2024). West Virginia guidance current as of the August 2025 edition.