3.6 Disproportionality and Identification Issues in Special Education

What Disproportionality Is and How It Is Measured

Disproportionality is the over- or under-representation of a racial, ethnic, or linguistic group relative to its share of the overall student population, in three specific decisions: identification (including identification in particular disability categories), placement in more restrictive settings, and discipline (suspension and expulsion). IDEA requires each state to examine districts annually.

Risk and Risk Ratio

The standard metric is the risk ratio:

• Risk for a group = (number of students in that group identified) divided by (total enrollment of that group).
• Risk ratio = the risk for the target group divided by the risk for all other students combined.

A risk ratio of 1.0 means proportionate representation. A ratio of 2.5 means students in that group are identified at two and a half times the rate of their peers. When a comparison group is too small to be stable, states use an alternate risk ratio against the statewide figure. States set a risk ratio threshold, require the pattern to persist for a defined number of years, and apply a minimum cell size before making a determination — safeguards that prevent a single unusual year in a small district from triggering findings.

The Consequence of a Finding

A district identified with significant disproportionality must:

1. Reserve 15 percent of its IDEA Part B funds for comprehensive coordinated early intervening services (CEIS), targeted at — but not limited to — the group identified as disproportionate, and usable for students not yet identified as needing special education.
2. Review and revise policies, procedures, and practices related to identification, placement, or discipline to determine whether they comply with IDEA.
3. Publicly report the revisions.

The 15 percent reservation is the single most-tested consequence. Note the logic: the remedy is invested in general education prevention, not in more special education.

Where Disproportionality Concentrates

Over-representation is concentrated in the judgmental (high-incidence) categories — specific learning disability, intellectual disability, emotional disturbance, and other health impairment — rather than in the low-incidence categories established by medical or sensory criteria such as deaf-blindness, visual impairment, or orthopedic impairment. That distribution is itself the evidence that the problem is decision-making rather than biology: eligibility in the judgmental categories depends on professional interpretation of terms such as "marked degree," and interpretation is where bias operates.

Disciplinary disproportionality follows the same pattern and compounds it, since removals reduce instruction, which lowers achievement, which increases referral.

Identification Issues and the Practices That Reduce Error

Definitional Ambiguity Is the Root Problem

Objective 0005 opens with "issues in definition and identification procedures." Several IDEA criteria are irreducibly judgmental:

• "Adversely affects educational performance" is not defined by a cut score. Illinois broadens it to include social relationships and emotional development, so a student with passing grades may still meet it.
• "To a marked degree" and "over a long period of time" in the Emotional Disturbance definition have no regulatory numbers; the widely used six-month convention is practice, not law.
• The social maladjustment exclusion requires a distinction between conduct chosen for gain or peer status and conduct driven by an affective disorder — a distinction that is culturally loaded and frequently applied inconsistently.
• SLD identification method varies: Illinois permits an RtI/MTSS response-based approach, a pattern of strengths and weaknesses, and, historically, severe discrepancy. Different methods identify different students from the same population.
• Developmental Delay is an optional, age-bounded category (ages 3 through 9 in Illinois) that postpones categorical labeling; students identified under it must be reevaluated and given a new eligibility determination by the end of the school year in which they turn nine.

Under-Identification Is Also Disproportionality

The framework asks about disproportionality "in the identification of students with disabilities," which runs in both directions:

• English learners are sometimes under-identified because staff attribute every difficulty to language acquisition and wait, and sometimes over-identified because language difference is read as disorder. Both errors come from the same failure to compare the student to true peers with similar language exposure and schooling history.
• Girls with autism and ADHD are identified later and less often because masking and compliance suppress referral.
• Students with internalizing profiles are missed because they do not disrupt instruction.
• Students in under-resourced schools may go unidentified because no one has ruled out weak core instruction, so every struggling student looks the same.

Practices That Reduce Misidentification

1. Fix Tier 1 first. If a large share of a classroom or grade is below benchmark, the problem is core instruction, and referring individual students will not fix it. Teams should check the class-wide and school-wide picture before treating a student's performance as individual.
2. Document intervention fidelity. Weak progress in an intervention that was never delivered as designed is not evidence of a disability.
3. Compare to true peers. Judge a multilingual student against students with a similar first language, length of exposure, and schooling history rather than against monolingual norms.
4. Assess in the native language or primary mode of communication, using a qualified bilingual evaluator or trained interpreter, and use dynamic (test-teach-retest) assessment to gauge modifiability.
5. Review and record the exclusionary factors explicitly. IDEA prohibits identification when the determinant factor is lack of appropriate instruction in reading, lack of appropriate instruction in math, or limited English proficiency. The team must attach evidence — attendance, instructional history, language proficiency data — rather than simply checking a box.
6. Use culturally responsive pre-referral practice. Consider whether the classroom expectations, materials, and interaction norms match the student's cultural and linguistic experience before concluding the student is the source of the mismatch.
7. Watch the discipline data. Disproportionate removal predicts disproportionate referral, so behavior and identification data should be reviewed together, not in separate committees.

The Balance Special Educators Must Hold

Avoiding over-identification cannot become a reason to deny services to a student who genuinely needs them, and neither error is safer than the other. A student wrongly identified is placed under lowered expectations and a permanent label; a student wrongly denied loses years of intervention. The defensible position is the same in both directions: the decision rests on multiple sources of data, gathered in the student's language, compared to appropriate peers, with the exclusionary factors examined and documented.