What the data shows
What The Research Says
This synthesis draws on 20 core sources: 10 peer-reviewed sources and 10 legal authorities. I read them asking one question: when does a student need specially designed instruction, and when can general education meet the need with the right support?
The one question behind the reading
Three threads run through all 20 core sources. Access to general education matters, and it is measurable. The line between need and benefit is a legal line, and data has to draw it. And none of it works without real collaboration between the adults. The three buckets below sort every source, study, and legal authority alike into the thread it speaks to, with each one cited where it lands. After the buckets comes the data, then the conclusion.
20
Core sources
10
Peer-reviewed sources
10
Legal authorities
Access & Outcomes
- Students with disabilities who spend more time in general education tend to show stronger achievement in reading and math. Cosier et al., 2013
- Time alone is not an automatic fix. The research is mixed on how much achievement moves without strong, well matched support in place. Fuchs et al., 2023
- Teams often drift toward a more restrictive setting without first documenting the supports already tried in general education. (Kurth et al., 2019) The law requires that every placement in general education begin with support and asks the team to justify moving a student out of it, not the reverse. Sacramento City USD v. Rachel H., 1994
- Placement runs along a continuum from general education outward, and a team moves only as far as a documented need requires. 34 C.F.R. section 300.115; Cal. Ed. Code section 56361
- When a disability is suspected, the school has a duty to look. Access begins with identifying the students who may need it. Timothy O. v. Paso Robles USD, 2016
Telling Need from Benefit
- Specially designed instruction follows the student, not the building. It is not dictated by a school's instructional model, curriculum, or staffing. Sayeski et al., 2023
- FAPE and LRE work as a pair, not a trade-off: a program reasonably calculated to provide appropriate progress, in the setting closest to general education that can meet the need. Yell & Bradley, 2025
- Response to Intervention is a general education framework first. A slow response should inform an eligibility decision; never decide it alone. Fletcher & Vaughn, 2009
- RTI implementation varies widely across classrooms, so a slow response can reflect a support gap as much as a student's true need. Alahmari, 2019
- Eligibility has two parts: a qualifying disability, and, because of it, a need for specially designed instruction. Qualifying is not the same as needing services everywhere. IDEA, 20 U.S.C. section 1400 et seq.
- An appropriate education is a floor of opportunity, not the best program available. Appropriate is the legal standard, not optimal. Board of Education v. Rowley, 1982
- That floor still has to be ambitious: a program reasonably calculated for progress appropriate to the child's circumstances, not merely more than nothing. Endrew F. v. Douglas County, 2017
Challenges & Collaboration
- Special educators nationally describe unclear roles and heavy workload when a shared framework is missing. Werts et al., 2014
- Monitoring intervention with fidelity depends on structured collaboration between general and special education, not assumed goodwill. Gates et al., 2013
- Tiered systems like MTSS only work when the adults running them have real training and real time, which is not automatic. Brown-Chidsey, 2024
- Arriving at a meeting already set on one outcome shuts out the family and is itself a procedural violation. Deal v. Hamilton County, 2004
- A team has to genuinely consider the full range of supports in general education, with the family, before choosing a more restrictive setting. Oberti v. Board of Education, 1993
- Biased assessment can place Black students into segregated classes at discriminatory rates. Identification itself can tip the scale the wrong way, so check the data, not the assumption. Larry P. v. Riles, 1984
What we can measure
The data
- Time in general education, measured: students with disabilities who spend more of the school day in general education show higher reading and math achievement. Cosier et al., 2013
- The honest counterweight: the achievement effect of more time is mixed when the support is weak, so time is necessary but not sufficient on its own. Fuchs et al., 2023
- A documentation gap shows up in the record: analyses of LRE statements find placements that move toward restriction without first documenting the general education supports already tried. Kurth et al., 2019
- Intervention response is not a clean signal: RTI is delivered unevenly across classrooms, so a slow response can reflect the instruction as much as the student. Fletcher & Vaughn, 2009; Alahmari, 2019
What it adds up to
The conclusion
Two ideas hold everything together. Specially designed instruction follows the student, not the building. And the right amount of support is a finding, drawn from data, not a feeling.
Access comes first. Document need with real data. Support is matched to what each student actually requires.