From MTSS to referral: when intervention data says it is time

By the IEP Atlas editorial teamLast updated November 1, 2026

Short answer

Refer for a special education evaluation when a disability is suspected, not after a set number of tiers. In practice, that is when well-delivered intervention produces too little growth: the student stays far below peers and the slope is flat. Check core instruction, fidelity and attendance first, then refer promptly. RTI cannot be used to delay an evaluation.

The referral question is not "Has the student finished Tier 3?" It is "Do we suspect a disability, and do we have reason to think the student needs specially designed instruction?" Intervention data helps answer that question. It does not replace it.

This page is for principals, MTSS coordinators and school psychologists who run the meeting where that decision gets made. It gives a sequence of checks, a worked example and the mistakes that most often slow a referral down or speed it up for the wrong reasons.

What the law says about timing

Two rules frame every referral decision.

First, the school has a child find duty to identify students who may have a disability and need services.^[2] A parent can request an evaluation at any time.^[3]

Second, federal guidance says an RTI process cannot be used to delay or deny an evaluation of a student suspected of having a disability.^[1] For learning disability decisions, the regulations go further. If a student has not made adequate progress after an appropriate period of time, the school must promptly request parental consent to evaluate.^[5]

States set their own learning disability criteria. They must allow a process based on the student's response to research-based intervention, and they cannot require a severe discrepancy between ability and achievement.^[4] Your state's rules decide how intervention data is weighed in eligibility.

What you need before you start

Screening data for the student and the whole class or grade
Weekly progress monitoring for at least 6 to 8 weeks per intervention
The intervention plan: program, minutes per week, group size, start date
Fidelity and attendance logs
Vision and hearing screening results and attendance records
English proficiency data, if the student is an English learner
Parent input and any outside reports

Step-by-step: deciding whether to refer

Step 1: Rule out a core instruction problem

Look at the class before the student. If 60% of a second grade class is below the reading benchmark, the first fix is core instruction, not a referral for one child. If most classmates are on track and this student is not, the problem is more likely specific to the student. Either way, a student with a suspected disability can still be referred while the school fixes Tier 1.

Step 2: Check fidelity and dose

Confirm the intervention ran as planned. A plan of 30 minutes, 5 days a week that delivered 2 sessions a week because of testing and assemblies is not a fair test. Compare sessions planned with sessions delivered, and note absences.

Step 3: Look at level and slope together

Level is where the student is now compared with peers or a benchmark. Slope is how fast the student is growing. A student who is low but growing faster than peers is closing the gap. A student who is low with a flat slope is not. Most teams treat "low level plus slow growth" as the strongest signal. The rate of improvement calculation shows how to get a slope from weekly probes.

Step 4: Apply a decision rule

Use a rule written before the data came in. A common one: if four consecutive points fall below the goal line, change the intervention; if the slope stays below the goal after a change, consider referral.^[7] Writing the rule down keeps the team from arguing the data after the fact.

Step 5: Look at other explanations

Check attendance, vision, hearing, recent moves, trauma and language proficiency. These do not rule out a disability, and many students with disabilities also have one of these factors. They do change what the evaluation needs to include.

Step 6: Ask the disability question out loud

At the meeting, someone should ask: "Do we suspect a disability?" Record the answer and the reason. If the answer is yes, the next step is a referral and a request for consent. If it is no, write the next intervention change and the next review date.

When the school proposes to evaluate, it sends prior written notice and asks for consent.^[6] Give the parent the assessment plan with it. If the school decides not to evaluate after a parent's request, it must give written notice of that refusal and the reasons.

Worked example: one student, two intervention cycles

Darius is a fictional second grader. In the fall screening on September 8, he read 18 words correct per minute (WCPM). His school's fall benchmark for grade 2 was 50, and 21 of 24 classmates met it. Core instruction was working for most of the class.

Cycle 1, September 15 to November 7 (8 weeks). Small-group phonics, 30 minutes, 4 days a week, in a group of 5. He attended 30 of 32 sessions. Weekly probes: 18, 19, 19, 21, 20, 22, 21, 23. Growth was about 0.6 WCPM per week. His goal line called for 1.5. The last four points sat below it.

Change, November 10. The team moved him to a group of 2, added 15 minutes of daily decoding practice and kept the weekly probe.

Cycle 2, November 10 to December 19 (6 weeks). He attended 28 of 30 sessions. Probes: 23, 24, 23, 25, 24, 26. Growth was about 0.5 WCPM per week. His classmates in the benchmark group averaged about 1.2 per week over the same period.

Check	Darius	Signal
Class-wide performance	21 of 24 met benchmark	Core instruction is working
Fidelity and attendance	58 of 62 sessions	Intervention was delivered
Level	26 WCPM vs a winter benchmark of 72	Far below peers
Slope	0.5 to 0.6 WCPM per week vs peers at 1.2	Gap is widening
Other factors	Vision and hearing passed; 96% attendance; English is his first language	No other clear explanation

The team met on January 6 and agreed a disability was suspected. The school sent notice and a consent form that week, with an assessment plan covering reading, phonological processing, language and cognitive ability. The intervention continued during the evaluation. The numbers are illustrations; your benchmarks and decision rules come from your district.

Common mistakes

Weak: "He needs to complete Tier 3 before we can refer." Better: "We suspect a disability. We will refer now and keep Tier 3 going during the evaluation."

Weak: Referring after two weeks of data because the teacher is frustrated. Better: Collect 6 to 8 weekly points unless other information already points to a disability, then decide with the data.

Weak: Calling an intervention a failure when the student attended half the sessions. Better: Report sessions delivered and attended next to the growth rate, and fix delivery first.

Weak: Looking only at level. "She is still below benchmark, so refer." Better: Look at level and slope. A student below benchmark who is growing faster than peers may not need a referral yet.

Weak: Treating a parent's written request as an item for the next SST. Better: Respond to the request itself with consent forms or a written refusal and reasons.

Templates and tools

When consent arrives, enter the date the school received it in the IEP due date calculator, with your state's timeline. It returns the evaluation deadline so the team can schedule testing.
For how the clock runs, see the 60-day evaluation timeline.
The pre-referral intervention examples show how to document the intervention the referral will rely on.
The reading fluency data sheet and the oral reading fluency measurement page cover the weekly probe in the example.
Parents asking about timing can read can a school delay an evaluation for RTI.

The process after referral, from consent to eligibility, is in the IEP evaluation process.

Frequently asked questions

How many tiers must a student go through before a special education referral?

None are required by federal law. A school may refer at any point when a disability is suspected, and a parent may request an evaluation at any time. Intervention data is useful evidence, but finishing a tier is not a precondition.

How much progress monitoring data is enough for a referral?

Most teams want 6 to 8 weekly data points per intervention to judge a slope. Less can be enough when other information, such as a medical diagnosis or a parent's report of history, already points to a disability.

Should the intervention stop once the student is referred?

No. Keep it going during the evaluation. The extra weeks of data help the team, and the student keeps getting help while testing is done.

Use it in IEP Atlas

Sources

OSEP Memo 11-07: Response to Intervention (RTI) (January 21, 2011), U.S. Department of Education, Office of Special Education Programs
34 CFR §300.111: Child find, U.S. Department of Education (IDEA website)
34 CFR §300.301: Initial evaluations, Legal Information Institute, Cornell Law School
34 CFR §300.307: Specific learning disabilities, U.S. Department of Education (IDEA website)
34 CFR §300.309: Determining the existence of a specific learning disability, U.S. Department of Education (IDEA website)
34 CFR §300.503: Prior notice by the public agency; content of notice, U.S. Department of Education (IDEA website)
Decision Rules for Analyzing Academic Progress Monitoring Data, National Center on Intensive Intervention

Written and checked by the IEP Atlas editorial team under our editorial policy. This guide describes general practice and federal rules. It is not legal advice; your state may add requirements, and the IEP team makes the decisions for each student.