CaseloadEvaluateNotesLiteracyMySLPPricingSecurityFor districtsResources
Log inGet EASI at Northern Speech
White paper

Beyond the Cutoff Score: A Practical Guide for SLPs

Free SLP guide to speech language eligibility: cutoff score limits, what IDEA requires and how to add language sample and functional evidence.

Lois Jean Brady, M.A., SLP · November 4, 2026

An open blank notebook on a light wooden desk with a cup of tea, pastel sticky notes, a picture book, a toy car and a closed laptop.

Download the white paper (PDF)

Summary

No federal cutoff score decides speech language eligibility. IDEA bars any single measure from being the sole criterion and asks for functional, developmental and academic information, including parent input. This guide shows how to read a test manual, add a language sample and other functional evidence, and lay it all out for the team, which decides.

Key points

  • IDEA does not set a standard score cutoff for speech or language impairment. States and districts set criteria, and some leave it to local teams.
  • No single measure or assessment may be the only criterion for an eligibility decision, and evaluations must gather functional, developmental and academic information.
  • Research shows that many children with language impairment score above common cutoffs on some tests, and that few test manuals have historically reported diagnostic accuracy.
  • A language sample, parent and teacher input, classroom observation and dynamic assessment add the functional evidence a score cannot provide.
  • Educational performance is not limited to academic performance.
  • For preschoolers, the choice between speech or language impairment and developmental delay depends on whether and how your state has adopted developmental delay.
  • The group of qualified professionals and the parent decides eligibility. EASI never decides, diagnoses or qualifies a child.

The question behind this guide

In early November 2026, a school SLP asked colleagues online what standard score cutoff they use when the state does not set one. Replies described a wide range of practices across states and districts. Several clinicians made the same point in different words: the score is a starting place, and the extent of the impact on the child is what the team must weigh. A second discussion asked how to classify preschoolers whose scores sit near average while concerns remain.

These are good questions, and they deserve answers grounded in regulation and research. This guide pulls those sources together and offers a practical workflow you can use on your next evaluation.

Part 1: What federal regulations require

The definition names an effect, not a score

IDEA regulations define speech or language impairment as "a communication disorder, such as stuttering, impaired articulation, a language impairment, or a voice impairment, that adversely affects a child's educational performance" (34 CFR 300.8(c)(11)). A child is a child with a disability only after an evaluation that follows 34 CFR 300.304 through 300.311 and only if, because of that disability, the child needs special education and related services (34 CFR 300.8(a)(1)).

There are two parts to that test: a disability, and a need. ASHA frames the school team's work as three questions: Is there a disability? If so, is there an adverse effect on educational performance resulting from the disability? If so, are specially designed instruction and related services needed to help the student make progress in the general education curriculum? (ASHA, Eligibility for School Based Speech Language Services). A cutoff score speaks, at most, to part of the first question.

Evaluation procedures

The evaluation procedures in 34 CFR 300.304 are the backbone of a defensible eligibility process. The public agency must:

  • use a variety of assessment tools and strategies to gather relevant functional, developmental and academic information about the child, including information provided by the parent (300.304(b)(1))
  • not use any single measure or assessment as the sole criterion for determining whether a child is a child with a disability (300.304(b)(2))
  • use assessments in the form most likely to yield accurate information on what the child knows and can do academically, developmentally and functionally (300.304(c)(1)(ii))
  • use assessments for purposes for which they are valid and reliable (300.304(c)(1)(iii)), administered according to the producer's instructions (300.304(c)(1)(v))
  • assess the child in all areas related to the suspected disability, including communicative status (300.304(c)(4)), in a way that is sufficiently comprehensive to identify all of the child's special education and related service needs (300.304(c)(6))

Reviewing existing data

Before deciding what new testing is needed, the IEP team and other qualified professionals review existing evaluation data, including information from the parents, current classroom based assessments and observations, and observations by teachers and related services providers (34 CFR 300.305(a)). In practice, this is where functional evidence often starts.

Determining eligibility

The determination itself is made by "a group of qualified professionals and the parent of the child" (34 CFR 300.306(a)(1)). In interpreting evaluation data, the group must draw upon information from a variety of sources, including aptitude and achievement tests, parent input and teacher recommendations, and must ensure that information from all of these sources is documented and carefully considered (34 CFR 300.306(c)(1)).

Note that 34 CFR 300.307 through 300.311 contain additional procedures specific to specific learning disabilities. They are not the rules for speech or language impairment, though the general procedures above apply to every evaluation.

Developmental delay

States may use developmental delay for children ages 3 through 9, or any subset of that range, whose development is delayed in one or more areas, including communication development, as defined by the state (34 CFR 300.8(b)). A state that adopts the term must decide the age range, and a local agency may not use it if the state has not adopted it (34 CFR 300.111(b)). The regulations also state that nothing in IDEA requires that children be classified by their disability, so long as each eligible child is regarded as a child with a disability (34 CFR 300.111(d)).

Part 2: What a cutoff score can and cannot tell you

The arithmetic of cutoffs

Many standardized language tests report standard scores with a mean of 100 and a standard deviation of 15. In a normal distribution, about 16 percent of scores fall more than 1 standard deviation below the mean, about 7 percent fall more than 1.5 standard deviations below and about 2 percent fall more than 2 standard deviations below. These are properties of the distribution, not of any child. A cutoff sets how many typically developing children fall below the line, but it does not by itself tell you whether a particular child has a disorder.

Do children with language impairment score below common cutoffs?

Not reliably. Spaulding, Plante and Farinella (2006) reviewed 43 commercially available child language tests. On the majority, children with language impairment scored above 1.5 standard deviations below the mean on average, and on more than a quarter of the tests their mean score fell within 1 standard deviation of the mean. Only 9 of the 43 manuals reported sensitivity and specificity, and only 5 reported accuracy at 80 percent or better. The authors argued that a fixed low score cutoff is not always appropriate, and that clinicians should look for evidence of a test's diagnostic accuracy at a given cutoff.

The concern is not new. Plante and Vance (1994) evaluated 21 preschool language tests against psychometric criteria and found that 38 percent met half or more of them. When they gave four tests to 4 and 5 year olds with and without specific language impairment, only one test reached acceptable accuracy in identifying the children.

How clinicians choose tests

Betz, Eickhoff and Sullivan (2013) surveyed 364 SLPs about 55 standardized tests used to diagnose specific language impairment. Of the test characteristics they examined, publication year was the only one related to how often a test was used. Psychometric quality, including diagnostic accuracy, did not predict use. That finding is a useful reminder to read the manual before choosing a battery.

Reading the manual like a clinician scientist

Dollaghan (2004) laid out questions for judging diagnostic evidence, including whether the test was compared with a valid reference standard in a representative sample, and whether results are reported as sensitivity, specificity or likelihood ratios with confidence intervals. When you pick a test, look for:

Question Where to look in the manual
Who was in the normative sample, and does this child resemble them? Standardization or norms chapter
Was diagnostic accuracy studied, and with what reference standard? Validity chapter, clinical groups
What were sensitivity and specificity at the cut point your district uses? Validity chapter, classification tables
Was the test given exactly as standardized? Administration chapter, and your own notes

ASHA's guidance on late language emergence makes a related point for young children: SLPs must ensure that standardized measures show robust psychometric properties, and when a child is not represented in a test's normative group, the test cannot be used to determine the presence or absence of a communication disorder (ASHA, Late Language Emergence).

Research criteria are not eligibility criteria

Some readers will know the research system used in the large epidemiological study of kindergartners by Tomblin, Records and Zhang (1996), which classified children using scores 1.25 standard deviations or more below the mean on two or more of five language composites. That system was built for research classification in a specific study. It was not designed as a school eligibility rule, and it should not be borrowed as one.

A note on terminology

The CATALISE consortium recommended the term developmental language disorder and concluded that a mismatch between verbal and nonverbal ability is not required for the diagnosis (Bishop et al., 2017). In schools, ASHA does not support cognitive referencing, the practice of requiring language scores to fall below IQ scores before a student can receive services (ASHA, Eligibility for School Based Speech Language Services; ASHA, Spoken Language Disorders). Clinical diagnostic terms and IDEA eligibility categories are different systems. The team applies the IDEA category under state rules.

Part 3: Adding functional evidence

What counts as functional evidence?

ASHA's Practice Portal describes assessment of spoken language as using a variety of formal and informal tools, and lists sources that complement standardized assessment: language sampling, dynamic assessment, curriculum based assessment, interviews and observation (ASHA, Spoken Language Disorders; ASHA, Assessment Tools, Techniques, and Data Sources).

Evidence source What it adds IDEA hook
Language sample Everyday language in play, conversation or narrative, with measures such as MLU and number of different words 300.304(b)(1), (c)(1)(ii)
Parent input How the child communicates at home and in the community 300.304(b)(1), 300.305(a)(1)(i)
Teacher input and classroom observation Participation, following directions, peer interaction 300.305(a)(1)(ii), (iii)
Dynamic assessment How the child responds to teaching 300.304(b)(1)
Curriculum based assessment The language demands of the child's own classroom 300.304(b)(1)

Why a language sample?

A language sample shows how a child actually uses language in a natural context. ASHA describes it as a way to derive measures such as mean length of utterance and other grammatical and lexical measures "to complement data obtained from standardized language assessments" (ASHA, Assessment Tools, Techniques, and Data Sources).

Research supports pairing the two. Ebert and Scott (2014) compared narrative language samples with norm referenced test scores in 73 school age children. The two sources identified overlapping but different groups of children, and agreement depended on the cutpoint used. The authors concluded that language samples can complement norm referenced tests.

For younger children, Eisenberg and Guo (2013) found that the percentage of grammatical utterances in picture description samples differentiated 3 year olds with and without language impairment in their sample of 34 children. Rice and colleagues (2010) published MLU reference values in six month intervals for children with and without language impairment, which can help you interpret MLU when your sample follows comparable procedures.

How much talk do you need?

Heilmann, Nockerts and Miller (2010) compared 1, 3 and 7 minute cuts of conversation and narrative samples and found measures quite consistent, with productivity, lexical diversity and utterance length most reliable in short samples. Guo and Eisenberg (2015) studied parent elicited conversational samples from 3 year olds and concluded that 7 to 10 minutes of talk is desirable, warning that very short samples may lead to misdiagnosis. When the stakes are high, as with eligibility, more talk is safer. For details, see How Long Should a Language Sample Be? at https://easi-as.com/resources/how-long-should-a-language-sample-be-what-research-says. For MLU counting rules, see https://easi-as.com/resources/mlu-analysis-software-how-to-calculate-mlu.

Context and norms

If you compare a sample measure to reference values, use the same elicitation context and transcription rules those values used. Play, conversation and story tasks draw out different language (Southwood & Russell, 2004). Report the context, the length and the number of utterances analyzed along with each measure.

Dynamic assessment

Dynamic assessment uses a test, teach, retest approach. In a study of culturally and linguistically diverse preschool children, Peña, Iglesias and Lidz (2001) found that measures of how children responded to mediated teaching of word learning distinguished children with typical and low language ability better than static pretest scores. ASHA notes that dynamic assessment can be used in conjunction with standardized assessments and language sampling (ASHA, Assessment Tools, Techniques, and Data Sources).

Adverse effect is broader than grades

ASHA's chronology of U.S. Department of Education letters, including a 2007 letter and a 2023 response from the Office of Special Education Programs, emphasizes that educational performance is not limited to academic performance, and that the adverse effect of a speech or language disorder must be considered case by case (ASHA, Eligibility for School Based Speech Language Services). Functional evidence is how you show what that effect looks like for this child: joining play, answering questions, retelling events, asking for help, or being understood by peers.

Part 4: A practical workflow

  1. Check your state and district rules. Use ASHA's interactive map of state speech and language eligibility guidelines, then confirm against your state education agency and district procedures.
  2. Review existing data. Gather parent input, teacher observations and classroom work before choosing new tests (300.305(a)).
  3. Choose tests on evidence. Read each manual for the normative sample and diagnostic accuracy at the relevant cut point.
  4. Plan the functional evidence. Decide which language sample context fits the child, who will be the partner, and how much talk you need.
  5. Collect and review the sample. Record it so you can check your transcript. Review each line before any measure is calculated.
  6. Look for convergence. Lay the sources side by side. Do they point the same way? Where they differ, explain why.
  7. Write the evidence clearly. Report scores with their limits, sample measures with context and length, parent and teacher input, and the functional impact at school.
  8. Bring it to the team. The group of qualified professionals and the parent decides.

Convergence worksheet

Source Finding Points toward concern? Notes on limits
Standardized test 1 Normative sample, accuracy at cut point
Standardized test 2
Language sample Context, length, utterances analyzed, reference values used
Parent input
Teacher input or observation
Dynamic assessment
Impact at school, academic and beyond

Report language you can adapt

These examples describe evidence. They do not state an eligibility conclusion, which belongs to the team.

  • "Standard scores on [test] fell in the [range] range. The manual reports diagnostic accuracy at [cut point]; results are interpreted together with the functional evidence below."
  • "A [length] minute play based language sample with a parent yielded [number] complete and intelligible utterances. Transcript measures, calculated after clinician review of each line, are reported below with the context used."
  • "Classroom observation and teacher report describe [specific functional observations]."

ASHA's guidance on documentation in schools notes that the specific definition of eligibility varies by state and district and is decided by the IEP team, and that recommendations for eligibility are a team decision typically not included in the evaluation report (ASHA, Documentation in Schools).

Part 5: Where EASI fits

EASI is language sample analysis software made by iTherapy, LLC. You can record a sample live using the built in pictures, upload audio or bring in an existing transcript. EASI analyzes English language samples. The workflow is Draft, Review, Attest, Score: EASI drafts the transcript, you review the words and speaker labels and attest each line, and only then, when the sample supports it, does EASI calculate measures such as MLU, number of different words, vocabulary diversity and the Index of Productive Syntax for you to review. You decide whether the sample is adequate, which norms apply and what the results mean. Learn more about EASI language sample analysis software at https://easi-as.com/language-sample-analysis.

MySLP, the secure AI assistant built into EASI, answers SLPs' clinical, scientific and eligibility questions (https://easi-as.com/myslp). For your state's rules, use ASHA's state map and your state education agency's current guidance. Report drafting (in beta) can bring your reviewed findings together into a draft that you edit and sign (https://easi-as.com/solutions/slp-report-writing).

EASI is HIPAA and FERPA compliant, with a BAA and insurance. EASI never decides eligibility, diagnoses or qualifies a child. The clinician and the team stay in charge.

Get EASI, with MySLP included, at NorthernSpeech.com/EASI: https://www.northernspeech.com/EASI. Northern Speech Services is the exclusive distributor.

References

Download the white paper (PDF)

Your caseload shouldn't take over your life.

EASI connects evaluations, reports, progress notes, planning and scheduling, with the SLP in control.

Get EASI at Northern Speech More resources