Skip to content

Least Restrictive Environment

Principle · All principles
Evidence · 25 claims (11 for, 12 mixed, 2 against) · 16 studies (7 associational, 3 causal, 3 quant-synthesis, 2 qualitative, 1 theoretical), q1–q4 · 2 of 16 report an effect size · 23 claims rest on one study

Conditional relationship

This page owns the placement decision for a student with a disability: in which mix of settings (the general class, shared school activities, a resource room or intervention group, a separate class, a separate school) the student is taught, and how that mix is decided and revised. The U.S. IDEA requirement is that the student be educated with non-disabled peers in regular classes and school programs as much as the specific student and setting allow. The design relationship behind it, as this page states it: for a student with a documented disability, bringing supports and specially designed instruction to the student inside the general class and school life, and moving the student to a separate setting only for a specific, documented need that the supported general class cannot meet, is expected to give learning (general-achievement) and adjustment (motivation-affect) no worse than a separate placement, more access to grade-level content, and fuller participation in the class and school as a member rather than a visitor. The mix is set case by case by a team (classroom teacher, special-education teacher, parents or guardians, and the student) and reviewed against data, not fixed by disability category or by grade level.

What the wiki's evidence tests, and what it does not. One systematic review (Dalgaard et al., 2022) compares inclusive with segregated placement for children with special educational needs: across 15 non-randomised studies (all but one at serious risk of bias), the pooled differences were negligible for math and for language and literacy and small and non-significant for psychosocial adjustment, with large effects in both directions in single studies, and moderator analyses (age, type and severity of need, part- or full-time inclusion, co-teaching) inconclusive. The authors conclude that inclusion in general is very unlikely to increase or decrease learning or adjustment. A non-significant difference does not establish that the two placements are equivalent for any one student, but it does mean the placement label alone does not predict outcomes: what a setting provides (the instruction, the supports, the time on grade-level content) has to carry the decision. One randomised trial of a reading intervention for struggling secondary readers (not identified as students with disabilities) found that students pulled out of other classes to receive it were particularly negatively affected, possibly from missing grade-level content. No claim in the wiki tests the case-by-case decision procedure itself, the IEP team process, or any rule for when to move a student to a more or less restrictive setting. The default design below is therefore the earlier page's guidance made concrete, each step labelled with its evidence status.

The relationship is conditional on four things: what the student needs that the general class cannot yet provide (a student whose disability is only physical may need access, not different curriculum; a student reading far below grade may need intensive instruction somewhere); what the separate setting actually delivers (specially designed, data-responsive instruction, or only a quieter room with the same worksheets); what the student loses by leaving (grade-level content, peers, membership); and who decides, and how often the decision is revisited. A perception that curriculum adaptation is harder at higher grades pushes older students toward more restrictive placements; the earlier page treats that as an assumption to question, and no claim here settles it.

How the converted siblings sit around this: Supporting Students with Intellectual Disabilities holds how ordinary class instruction is planned for a student with an intellectual disability once they are in the class; this page sets where that instruction happens. Universal Design for Learning holds planning the general class so that fewer demands need after-the-fact accommodation, which is what makes the general class workable for more students. Functional Behavior Assessment holds how a recurring behavior is understood and answered, and the due-process rule that a behavior plan in an IEP is changed through the team. Flexible Grouping holds temporary small groups inside a class, which is often how specially designed instruction can be delivered without removing a student. This page does not repeat their models.

Default design, while the relationship is untested

A default a team can act on and revise. Each step is a design proposal; its evidence status is stated beside it. Quantities are untested proposals unless a claim is cited for them.

  1. Start from the general class and write down the reason for every minute away from it. For each block of separate time, the plan names the need, the instruction the separate setting will deliver, and why that instruction cannot be delivered in the class with support. A block whose only reason is the disability category or the student's grade is not a reason. Basis: the earlier page; placement alone did not predict learning or adjustment in the one review here (Inclusion in general is very unlikely to increase or decrease learning and psychosocial adjustment of children with special needs [~M]), so the case for a setting has to come from what it provides.
  2. Bring support to the student before moving the student to support. In order: adjust the class design for everyone (options for how content is presented and how learning is shown; see Universal Design for Learning); add structured peer support (peer tutoring pairs, assigned roles in group work); add adult support in the class (co-teaching, an aide, a specialist working with a small group in the room); only then consider time out of the class. Evidence: school peer tutoring raises achievement for tutees and tutors, and in one meta-analysis of mathematics tutoring the tutee's disabled or at-risk status was not a significant moderator (Peer Tutoring Improves Achievement [+M], carried from general school populations to a placement decision). The order is an untested proposal.
  3. If separate instruction is needed, schedule it so it does not replace grade-level content in a core subject. Use an intervention period, an elective slot, or time before or after school, not the student's own English or mathematics class. Default dose (untested proposal): no more than 30 to 45 minutes a day of separate instruction for a student whose needs are mild, in a fixed block. Evidence: Students pulled out of other classes to receive LLI were particularly negatively affected [+M] (struggling secondary readers in one randomised trial, a subgroup comparison; the authors' explanation, missed grade-level content, is a hypothesis; carried to students with disabilities, whom it did not single out).
  4. Make the separate time specially designed and responsive to assessment data, with an entry and an exit criterion. State the skill targeted, the measure used to check it (a short curriculum-based measure is a common choice), and the score or pattern that ends the block. Evidence: Outcomes of students with disabilities improve when they receive specially designed instruction responsive to assessment data [+W] (a policy brief's statement citing four studies whose designs it does not describe). The entry and exit criteria are untested proposals.
  5. Build the plan as a team, in writing, with the student in it. The IEP team includes at least the classroom teacher, the special-education or resource teacher, and parents or guardians, and the student wherever possible. Before the meeting, teach the student to present their strengths, goals and needed supports and to lead part of the meeting. Evidence for the student's part: Self Directed IEP Increases Student Participation [+S] (secondary students in one randomised trial and four high-school students with moderate intellectual disability in a single-case study; the outcome is participation in the meeting, not placement or learning). The written, team-built plan is from the earlier page; untested here.
  6. Decide case by case, not by category or grade. Ask the same question for a kindergartner and an adolescent with similar needs: what does this student need that the supported general class cannot give? Do not move an older student to separate classes only because curriculum modification looks harder at higher grades. Untested; from the earlier page. Placement rates vary with how a district is organised, not only with student need (Single-district SELPAs had meaningfully lower inclusion rates, with a higher proportion of students with disabilities in general education less than 40 percent of the school day [~M], descriptive, California), which is a reason to ask for the case-specific reason.
  7. Review the mix on a short cycle, and move toward the general class when the data allow. Untested proposal, as a default arc across a school year:
  8. Weeks 1 to 6 (a trial period after any placement change): check every two weeks against the IEP goal: the curriculum-based measure, the share of class work completed, participation in group work, and any behavior incidents; also check that each support was delivered as written (an unreceived support is not a failed one).
  9. Moving toward less restriction: when the student meets the goal criterion at two consecutive checks, reduce the separate time by one block (for example one 30-minute block a day), or reduce in-class adult support (an aide sitting beside the student becomes an aide who checks in, then peers and a self-check).
  10. When progress stalls: if the criterion is missed at two consecutive checks with supports delivered as written, the team meets to add or change support in the class first; added separate time is the next step, with its reason written down (step 1).
  11. Mid-year and later: fade adult prompting toward peer support and the student's own self-monitoring; the annual IEP review is the legal minimum, not the review cycle.
  12. Transitions (preschool to kindergarten, elementary to middle school): start the new stage in the less restrictive setting unless the data say otherwise. Students who attend inclusive preschool and kindergarten are more likely to be in inclusive settings later in elementary school (A state funding report's research review, citing earlier studies, states that inclusive preschool special education benefits students with and without disabilities and that effective early education reduces later special education identification [+M], a report's research review, associational).
  13. Plan membership in the whole school life, not only academic minutes. Give the student a real role in assemblies, clubs, lunch, field trips and performances, and plan the supports for each in advance; treat the student as a member from the first day rather than a guest who earns access by showing readiness. Basis: the earlier page; Inclusion of students with disabilities often falls short because efforts start from a place of exclusion, making students guests who must earn access [~W] (an argument, not an empirical test).
  14. Change the plan only through the team. The IEP is a legal agreement: a teacher under classroom pressure documents the problem and asks for a team review rather than changing the placement or the behavior plan alone. From the earlier page; untested as a design rule; see Functional Behavior Assessment.
  15. Keep out-of-district and residential placements for last. Before placing a student outside the district, the team lists the in-district alternatives tried and the reason each did not meet the need. Evidence: one district that closed its separate school and rebuilt programs in regular schools reported two out-of-district placements in twenty years (Tacoma's progressive inclusion program pervaded district operations and limited out-of-district placements to two over twenty years [+M], a single district case description; it reports placements, not student outcomes).

When a condition is not met: if the general class has no adult support and none can be added, use structured peer support and class-wide design options (step 2) and concentrate specialist time on co-planning and the two-weekly data check; if the separate setting cannot say what it will teach and how it will check progress, it is not yet a reason to move the student.

Fitting the design to a situation

The three facts that most change the decision:

  • What this student needs that the supported general class cannot yet give. Access only (a wheelchair-accessible room, captioning, a reader), adapted curriculum on the same unit, or intensive instruction on a skill far below grade lead to different mixes. If unstated, ask: on this term's class tasks, with which supports, what can the student do now, and what can they not?
  • What the separate setting would deliver, and what it would displace. If unstated, ask: what instruction happens in the separate time, how is progress checked, and which class does the student miss to get it?
  • Who decides, and how soon the decision is reviewed. If unstated, ask: who is on the team, does the student take part, and when is the next data check?
If the brief says… Then change… Basis
The disability is physical or sensory only and the curriculum needs no change Full general-class placement; plan physical access, seating, assistive technology and materials in accessible formats before the first day; no separate academic time untested; from the earlier page
Learners are child, preschool or kindergarten age Default to an inclusive preschool or kindergarten class with an aide or specialist in the room; plan the move to first grade as a move to the general class A state funding report's research review, citing earlier studies, states that inclusive preschool special education benefits students with and without disabilities and that effective early education reduces later special education identification [+M] (a report's review of cited studies, associational)
Learners are adolescent in secondary school and teachers say curriculum modification is too hard at this level Keep the student in the subject class; the subject teacher and special-education teacher co-plan one modified version of the same unit (same topic, adjusted reading load and product); put any intervention block outside core classes untested for the co-planning; scheduling from Students pulled out of other classes to receive LLI were particularly negatively affected [+M] (struggling secondary readers, not specifically students with disabilities)
The student needs intensive skill instruction (for example, reads several grades below level) Add one fixed block of specially designed instruction (30 to 45 minutes a day, proposal) with a measure checked every two weeks and a written exit criterion; do not add more minutes by default when progress is slow, change the instruction first Secondary students who received more LLI or were pulled out of other classes were particularly negatively affected, possibly from missing grade-level content [~M] (one secondary reading trial; more dose was not better there); Outcomes of students with disabilities improve when they receive specially designed instruction responsive to assessment data [+W]
The main need is recurring behavior Assess what triggers and keeps up the behavior before changing placement; write a behavior plan into the IEP and change it only through the team; a move to a separate setting is a team decision with a dated review untested here; see Functional Behavior Assessment
classroom with one teacher, many learners, no aide Set up structured peer tutoring pairs and assigned group roles for the class; build presentation and response options into materials for everyone; the specialist co-plans for 15 minutes a week (proposal) instead of sitting in Peer Tutoring Improves Achievement [+M] (general school populations; carried to inclusion)
Scarce specialist or facilitator time (one special-education teacher across many classes, volunteers, a part-time aide) Spend specialist minutes on co-planning and the two-weekly goal check (about 10 minutes per student, proposal), not on supervising seatwork; give volunteers one scripted support (a reading partner routine) rather than open-ended help untested proposal
Mixed modes: the student learns from home or hospital, the class is partly online, or the family has only a phone Keep the student in the live class (joins the same lessons by video or audio, a named classmate as partner, captions or a transcript, work in the same format as peers); a separate packet or separate online course counts as time away from the class and needs its reason written down untested proposal
Single session only (one event, field trip, one-off workshop) Plan the student's role and supports for that event in advance so they take part in the same activity; offer an alternative activity only if the IEP names one untested proposal
The student or family has low confidence or low literacy Prepare the student to present part of the IEP meeting (a one-page strengths-and-goals sheet, rehearsed once); give the family a plain-language summary and an interpreter, and ask them first what they want for the student student part: Self Directed IEP Increases Student Participation [+S] (secondary students; meeting participation only); family part untested
Adults in work settings (transition to work, workplace or vocational training) Train the person on the real job alongside co-workers, with a job coach whose presence fades on the same two-check rule as step 7; a separate training room is time away with a stated reason untested; the earlier page's transition-planning source has no claim page in the wiki
A product or performance for a real audience (a school play, exhibition, concert) Give the student a real role in the shared production with the supports it needs (a script with cues, a partner, an adapted part); no separate showing untested proposal
The brief names its own routine (the district's IEP form, placement checklist, progress-monitoring tool or meeting agenda) Keep it and run it as given; change only what the evidence here bears on: a written reason for each block of separate time, scheduling that does not take a core class, and a dated review untested; scheduling from Students pulled out of other classes to receive LLI were particularly negatively affected [+M]
The brief uses its own categories or criteria (the district's continuum of placements, percentage-of-day bands, its own goal rubric) Use them; record where the student sits on that continuum now and which move is proposed, with the data that would justify the next move untested proposal
Private or sensitive information (diagnosis, IEP contents, family circumstances) The student and family decide what classmates are told; deliver supports that do not single the student out (offer the same options to everyone where possible); share the IEP only with the team untested proposal
A fixed short time budget (a 30-minute IEP meeting) 5 minutes: the student presents strengths and goals; 10: progress data against last goals; 10: supports and the placement mix, each separate block with its reason; 5: next review date and who checks what untested proposal
A whole-task performance the brief wants kept whole (a lab, a project, a team game) Keep the student in the whole task with an adapted role or tools, not a separate worksheet on part of it untested; see Universal Design for Learning
High stakes or legal dispute (due process, a contested placement) Document each support offered and delivered, the data at each check, and the team's reasons; separate the question "what does the evidence on placement show" (no average difference in the one review) from "what does this student need" The evidence base on inclusion effects is methodologically weak: no randomised studies and nearly all synthesised studies at serious risk of bias [~M]
The district's structure or habit, not the student, seems to drive placement Compare the student's proposed placement with what comparable districts or schools do for similar needs, and ask for the case-specific reason Single-district SELPAs had meaningfully lower inclusion rates, with a higher proportion of students with disabilities in general education less than 40 percent of the school day [~M] (descriptive, California SELPAs)
An out-of-district or residential placement is proposed List the in-district alternatives tried and why each fell short; ask whether a program could be built in a regular school Tacoma's progressive inclusion program pervaded district operations and limited out-of-district placements to two over twenty years [+M] (one district's case; placements, not outcomes)

Nine of the twenty rows cite a claim. Only the age, low-confidence and stakes rows rest on evidence about students with disabilities (and only the age and stakes rows on placement itself); the scheduling rows carry a trial of struggling secondary readers, the peer-support row carries general peer-tutoring syntheses, and the district rows rest on descriptive reports. The rest are untested proposals.

Observation, state and explanation

What can be observed: the minutes the student spends in each setting, the supports planned and the supports actually delivered, work completed and the score on the goal measure in each setting, participation (turns in group work, contributions, who the student works and eats with), behavior incidents with what came before them, and what the student and family say they want. "Is ready for the general class", "needs a separate setting" and "is not coping" are inferred states, and a placement label is not an observation of the student.

Response Plausible explanations Observation that could change the interpretation
Completes little class work in the general class The task is far from the student's current level; supports were planned but not delivered; an access barrier (cannot see, hear or read the material); a trigger in the setting Compare the support log with the plan for two weeks; give the same task one-to-one with the planned support; give it with the access barrier removed.
Progresses in the separate setting, not in the class The skill was learned in one format only; the class pace or materials differ; supports present there are absent in class Give a class-format item in the separate setting, and the separate setting's format in class; check which supports travel with the student.
Is socially isolated in the class Peers do not know how to work with the student; seating and grouping keep the student apart; a communication barrier; the student's own preference Observe a structured group task with assigned roles; ask the student, privately, who they would like to work with.
Behavior incidents rise after a move to the general class A specific trigger (noise, a demand, a transition); escape from work that is too hard; a short adjustment period Record what comes before and after each incident for two weeks before deciding anything; compare incidents in tasks matched and unmatched to the student's level.
The family asks for a separate setting An earlier experience of exclusion or bullying; safety; a specific service they believe only that setting offers; what they value for the child Ask what outcome they want and what worries them; describe what the separate time would deliver and what the student would miss.

These rows are proposals, untested with learners as diagnostics. A move between settings is itself an intervention that changes what is observed, so record each change and its date, and keep the explanations that remain open rather than choosing the one whose name fits.

Evidence and qualifications

The one direct comparison of placements in the wiki comes from a single review; the claims below describe it and its limits, then the neighbouring evidence the default design uses.

Populations, outcomes and designs differ across these claims (children with special educational needs in non-randomised comparisons, struggling secondary readers in one trial, secondary students in IEP meetings, district reports), so do not rank them by effect label. None measures what happens to an individual student moved between settings.

The evidence by question

Further evidence, not yet read against this model

Every claim the earlier page cited is used in the model above. These are further claims the wiki holds on placement, not yet read against the model:

Objective and learner-valued goal

The designer's or team's objective is usually progress on the IEP goals and grade-level content, with as much participation in the general class and school life as the student's supports allow, checked at a stated date. Elicit separately what the student wants: to be with friends or siblings, not to be singled out, to have a quiet place when the class is overwhelming, to get a particular job or go to a particular school. Elicit what the family wants too; it can differ from both. Agreement and divergence are observations to use. A student who values not being seen as different may refuse an in-class aide that works well for learning, and may accept the same support from a peer partner or a tool everyone has; a student who values a quiet place may do better with a scheduled break space than with a separate class.

For example, an adolescent with a learning disability may want to stay in the same English class as their friends, while the team's objective is reading fluency, which the resource-room block serves. Scheduling the fluency block in an elective slot serves both; scheduling it in place of English serves the team's objective at the cost of the student's, and of grade-level content. Record both, so that fluency gains are not read as the participation the student came for.

What would revise this model?

The expectation should change if well-designed comparisons (randomised or credibly quasi-experimental, with comparable students at baseline) show that for an identifiable group of students a separate setting with a stated kind of instruction produces better learning or adjustment than the supported general class; the model would then name that group and that instruction as a condition. If pull-out harm does not replicate for students with disabilities, or disappears when separate instruction is scheduled outside core classes, step 3 becomes a scheduling preference rather than a rule. If moderators (severity, age, part- or full-time inclusion, co-teaching) come to explain the heterogeneity the review found, the situation table should take its rows from them. If teams using the two-check review cycle move students toward restriction as often as away from it, the cycle is not doing what it claims. Do not protect the model by calling every poor result "supports not delivered" after the fact; record delivery as it happens.

Placement, the instruction delivered in each setting, achievement, adjustment and participation are separate claims. The present evidence in the wiki establishes no average learning advantage for either placement, no rule for when a student should move, and no dose of separate instruction.

Claims

  • Flexible Grouping — temporary small groups inside a class are often how specially designed instruction can be delivered without removing a student from it
  • Functional Behavior Assessment — IEP due-process requirements directly constrain how discipline plans can be changed for a student under a behavior-related IEP
  • Supporting Students with Intellectual Disabilities — inclusion in social and academic activities is one of that principle's three core teaching strategies, and is itself an LRE-style judgment
  • Universal Design for Learning — a more flexible-by-default core curriculum reduces how much a student's inclusion in the regular classroom depends on after-the-fact accommodation

Examples

Key Sources

  • Arduini-Van Hoose, N. (2020). Teaching students with disabilities. In Educational psychology. Retrieved from https://edpsych.pressbooks.sunycreate.cloud. CC BY-NC-SA 4.0.
  • West, J., et al. (1999). Transition planning under IDEA: Implications for practice. In Special education law and practice.
  • Special Education Facilities for Emotionally Disturbed Children. (1968). Ontario Department of Education. https://eric.ed.gov/?id=ED085907
  • Dowdy, E., & Kamphaus, R. W. (2007). A Comparison of Classification Methods for Use in Predicting School-Based Outcomes. The California School Psychologist, 12, 121-132. https://eric.ed.gov/?id=EJ896649. doi:10.1007/bf03340936
Page metadata
Field Value
type principle
id least-restrictive-environment
aliases special-classrooms-within-regular-schools-for-integration, gather-comprehensive-information-not-solely-categorical-eligibility
title Least Restrictive Environment
description For a student with a disability, bringing supports into the general class and school life and moving the student to a separate setting only for a documented need the supported class cannot meet, decided case by case by a team and reviewed against data, is expected to give learning no worse than separate placement and fuller participation; one review of non-randomised studies finds no average difference between inclusive and segregated placement, and no claim tests the case-by-case decision itself.
status review
canonical True
generated by claude/unspecified on 2026-10-15
trust unverified — no one has confirmed this page's content
sources 2 — special-education-facilities-for-emotionally-disturbed-children-1968, dowdy-2007

Frontmatter is the page's machine-readable record — see the Schema & Guide.