How to measure stuttering: percent syllables stuttered for IEP present levels

Percent syllables stuttered is the standard way to count how often a student stutters. It is useful, and it is also easy to misuse. On an IEP, %SS belongs in present levels as a description of the speech, next to measures of how the student feels about talking and how much they take part. It works poorly as the goal itself, for reasons covered below. This page shows how to take a clean %SS from start to finish, with one fictional student and real numbers.

For goals that target strategy use, participation and attitudes, see fluency goals examples. For the other speech areas in this cluster, start with the articulation goals examples pillar.

Why %SS stays in present levels

Stuttering changes from hour to hour, by listener, by topic and by how much is riding on the moment. A student can stutter on 3% of syllables with the SLP and 12% when called on in class. ASHA's fluency page describes stuttering as affecting attitudes, emotions and participation, not only the speech itself, and lists all of these as parts of assessment.^[1]

A goal that only says "fewer than 2% syllables stuttered" has three problems. It measures the easiest setting. It can reward avoidance, because a student who switches words or stays quiet will lower the count. And it asks the student to control something that varies for reasons outside their control. Use %SS to describe the speech; write goals about what the student does.

What you need before you start

• A recorder. A phone or tablet is fine. You will count from the recording, not live.
• Two or three speaking situations. Conversation with the SLP, a class discussion or small group, and one more that matters to the student, such as reading aloud or a phone call.
• A tally sheet. A grid where each box is one syllable, or a simple two-column count. The frequency data sheet works if you add a total column.
• Clear rules for what counts. Decide before you listen, and use the same rules every time.

What counts as a stuttered syllable

Teams differ on single-syllable whole-word repetitions without tension. Some count them, some do not. Either is defensible; what matters is writing the rule down and keeping it all year. Count each stuttered syllable once, however many repetitions it had: "b-b-b-ball" is one stuttered syllable.

Step-by-step: taking %SS

Step 1: Record in more than one setting

Record at least two settings on different days. One should be a school situation the student finds hard. A therapy-room-only number describes the student's best case.

Step 2: Collect at least 300 syllables per sample

Shorter samples give jumpy numbers: in a 100-syllable sample, each stuttered syllable moves %SS by a full point. Three to five minutes of conversation usually yields 300 to 500 syllables. If the student talks little, extend the sample rather than accept a short one.

Step 3: Count the total syllables

Play the recording and make one tally mark per syllable. Or transcribe it and count syllables in the text. Count syllables, not words: "elephant" is 3 syllables. Leave out interjections and revisions from the total if you are leaving them out of the stuttered count.

Step 4: Mark each stuttered syllable

Play it again and mark each syllable that had a stuttering-like disfluency. Note the longest blocks in seconds and any physical signs, such as eye closing, head movement or facial tension.

Step 5: Calculate

Stuttered syllables divided by total syllables, times 100. Calculate each sample separately. Do not pool settings into one number; the difference between settings is the useful part.

Step 6: Check your counting

Have a second SLP count one sample a year, or recount one of your own a week later. If the two counts differ by more than about 1 percentage point, tighten the rules. The interobserver agreement guide shows the math.

Step 7: Add the student's view

Pair the numbers with a measure of the student's experience: a self-rating of how hard speaking was, a published communication attitude scale if your district uses one, or a short interview. Ask what situations they avoid. This is what goals will target.

Worked example: one student from sample to present levels

Sam is in sixth grade. The SLP takes three samples in September.

In the science group, the SLP also notices Sam saying "I don't know" twice when called on, then giving the right answer to a peer afterward. On a 1-to-5 self-rating ("How hard was talking today?"), Sam rates conversation 2, reading 2 and the science group 5. He says he avoids raising his hand and never orders at restaurants.

Present levels statement: "In September, Sam stuttered on 5.1% of syllables in conversation with the SLP (412 syllables), 11.0% during a science small group (318 syllables) and 3.0% reading aloud (300 syllables). Moments include blocks up to 4 seconds with eye closing. Sam rates small-group talk as very hard (5 of 5) and reports avoiding raising his hand and ordering food. In class, he said 'I don't know' to 2 of 3 questions he could answer."

Goals built from it (not from %SS):

• When called on and he knows the answer, Sam will give it, stuttered or not, in 4 of 5 opportunities across 3 observed lessons. Baseline: 1 of 3. Template.
• During conversation, when Sam stutters, he will use a pull-out to ease out of the moment in 4 of 5 moments across 3 sessions. Baseline: 0 of 5. Template.
• After each weekly speaking challenge, Sam will rate the experience and explain the rating, with his average rising from 4.5 to 3 on the 5-point "how hard" scale by spring. Template.

At the annual review, the SLP repeats the same three samples. %SS goes in the report as description, and the goals are judged on their own measures.

Common mistakes

• Weak: Goal: "Sam will speak with fewer than 2% syllables stuttered." Better: Report %SS in present levels and write goals for strategy use, answering in class and attitudes.
• Weak: One sample in the speech room. Better: Two or three settings, including one the student finds hard.
• Weak: Counting words instead of syllables. Better: Count syllables; a student who uses longer words is not penalized.
• Weak: Counting "um" and revisions as stutters. Better: Write down which disfluency types count and keep the rule all year.
• Weak: Averaging all settings into one number. Better: Report each setting; the gap between them is the finding.
• Weak: A 90-syllable sample. Better: 300 or more syllables per sample.
• Weak: %SS with no context. Better: Add block length, physical signs and the student's own rating.

Templates and tools

All fluency templates, including participation, strategy and attitude goals, are on the speech fluency goals page. Syllable counting rules are in syllable types for anyone new to it. For classroom supports that go with these goals, see accommodations for students who stutter, and for choosing a therapy approach, see fluency shaping vs stuttering modification. The goal checker will flag a fluency goal with no measure.

Frequently asked questions

Use it in IEP Atlas

• Speech fluency goals (speech language)
• Stuttering in the IEP glossary
• Syllable types in the IEP glossary
• Frequency data sheet
• IEP goal checker

Related guides

• Articulation goals examples: sound by sound, with data methods
• Accommodations for students who stutter: 24 examples written for the IEP
• Stuttering therapy techniques: fluency shaping vs stuttering modification
• Fluency goals examples for students who stutter

Sources

1. Stuttering, Cluttering, and Fluency, ASHA

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.