Articulation data collection: percent correct vs opportunities
By the IEP Atlas editorial teamLast updated
Short answer
Percent correct scores a fixed set of trials you elicit, such as 20 picture words. Opportunities scores the chances that come up on their own, such as every /r/ word in a conversation. The deciding test: did you control how many chances the student got? If yes, report percent correct of trials. If not, report correct target words out of the opportunities that occurred.
Most articulation data comes in one of two forms, and the IEP needs to say which. Percent correct is the familiar one: you present 20 words, the student gets 16 right, the score is 80%. Opportunities data counts how often the student uses the sound correctly when a chance comes up on its own, such as every /r/ word in a 3-minute conversation, where the number of chances changes every time. Mixing the two in one goal is the most common reason articulation progress reports are hard to read.
For the same question across all IEP goals, the percent accuracy vs percent of opportunities guide covers the general case. This page is the speech-room version, with articulation levels, group sessions and conversation samples.
The deciding question
Ask: did I control how many chances the student got?
- Yes. You chose 20 words, 10 sentences or one passage with 10 marked words. Report percent correct of a set number of trials: "16 of 20 words."
- No. The chances came up in talk, play or class, and their number changed. Report correct uses out of the opportunities that occurred: "31 of 40 /r/ words in a 3-minute sample (78%)." Always keep the count next to the percent.
Both report a percent. The difference is whether the denominator is fixed in advance or found in the sample.
Side-by-side comparison
| Percent correct of trials | Opportunities | |
|---|---|---|
| Definition | Correct productions out of a set number of elicited items | Correct uses out of the chances that came up naturally |
| Denominator | Fixed before the probe (10, 20) | Varies with each sample |
| Articulation levels | Isolation, syllables, words, phrases, sentences, marked reading | Conversation, play, class talk, home routines |
| Basis in law or practice | The IEP must say how progress is measured and reported;[1] the SLP picks the method | Same rule; the method must match the goal's setting |
| Who decides | The SLP, when writing the goal | The SLP, with the teacher if class data is used |
| Example goal | Given 20 untrained words with initial /s/, 18 of 20 correct | In a 3-minute conversation, /s/ correct in 80% of /s/ words |
| Data sheet | Opportunity trial data sheet with 10 or 20 boxes | Tally of correct and total target words, or the same sheet with a count column |
| Where documented | Session notes, probe log, progress report | Sample record (date, setting, length, listener), progress report |
| Main risk | A memorized item set inflates the score | Too few chances make the percent jumpy |
ASHA's guidance on documentation in schools expects data that ties directly to the IEP goal and its measure.[2]
Three scenarios
Scenario 1: a word-level /s/ goal in a group of three
The SLP sees three students for 30 minutes. Each has a word-level goal. She gives each student 10 untrained probe words at the start, then practices.
Answer: percent correct of trials. The SLP chose the items and the number. With three students, 10 trials each is realistic; 20 would eat the session. Report "8 of 10" and keep the denominator the same on every probe. If the goal says 20, run the probe every other session with 20 words rather than mixing 10 and 20.
Scenario 2: a conversation goal for vocalic /r/
A fifth grader's goal is vocalic /r/ in conversation. The SLP records 3 minutes about a class topic each month. One sample has 38 /r/ words; the next has 51.
Answer: opportunities. The student, not the SLP, decided how many /r/ words came up. Report "29 of 38 (76%)" and "42 of 51 (82%)." If a sample has fewer than about 20 target words, extend it; one error in 10 words moves the percent by 10 points.
Scenario 3: reading aloud in class
A third grader's carryover goal is /l/ during class reading. The SLP pushes in once a week.
Answer: it depends on how you set it up. If the SLP marks 10 /l/ words in the passage beforehand, that is percent correct of a fixed set: "7 of 10 marked words." If she scores every /l/ word the student happens to read, it is opportunities. Pick one and write the goal to match. Marked words are easier to score during a busy push-in and give a steady denominator.
Mistakes that blur articulation data
- No denominator. "80% accuracy" on a goal with no item count. Write "16 of 20 words" or "80% of target words in a 3-minute sample."
- Practicing the probe items. Probe with untrained words at the start of the session. The cold probe guide explains why.
- Counting all words instead of target words. In a conversation sample, the denominator is the target words, not every word the student said.
- Changing the trial count. 10 trials one week and 25 the next makes the graph hard to read. Keep it fixed.
- Scoring a near miss as correct. A distorted /r/ is incorrect. Code the type of error so you can show progress inside the incorrect column.
When you need both
Most articulation plans use both, at different levels. A student working on /r/ might have percent correct for sentences (20 untrained sentences every 2 weeks) and opportunities for conversation (one 3-minute sample a month). That is fine as long as each goal names its own measure. The progress report then shows the structured score near the top and the conversation score climbing toward it, which is the picture of carryover. Articulation carryover strategies covers that stage.
For intelligibility, where the measure is words a listener understands out of all words attempted, see intelligibility IEP goals. For setting the first number, see articulation baseline. The opportunities measurement guide and the percent accuracy guide cover scoring rules, and the goal checker flags a goal with no denominator.
Frequently asked questions
How many trials should I take for an articulation probe?
Twenty for an individual session and 10 per student in a group of three or more are common. What matters most is keeping the number the same on every probe for that goal, so the scores can be compared.
Is a conversation sample percent correct or opportunities?
Opportunities. The student decides how many target words come up, so the denominator changes each time. Report the count with the percent, such as 31 of 40 target words, 78%.
Should articulation data be scored during practice or only on probes?
Use probes for the goal score. Practice data with cues and feedback runs higher than what the student does alone. Many SLPs log practice for planning and graph only the untrained probes.
Use it in IEP Atlas
- Opportunities measurement guide
- Free IEP data sheets
- Opportunities data in the IEP glossary
- Data sheet in the IEP glossary
- Articulation goals (speech language)
Related guides
- Articulation goals examples: sound by sound, with data methods
- Voice therapy data collection: how to measure voice goals
- Articulation baseline: 24 baseline statements with the goals they support
- Articulation carryover strategies: moving a sound into class and home
- Articulation goals for 3 year olds: 20 play-based examples
Sources
- 34 CFR §300.320: Definition of individualized education program, Legal Information Institute, Cornell Law School
- Documentation in Schools, 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.