How to choose a data collection method for an IEP goal
By the IEP Atlas editorial teamLast updated
Short answer
Choose the method from the goal's verb and the question you need answered. If the student answers items, use percent accuracy or trials. If the behavior is counted, use frequency. If it lasts, use duration. If the issue is starting, use latency. Use interval recording for long periods, a rubric for products and a task analysis for routines.
The data collection method is set by what the goal asks the student to do. A goal about how often a student greets peers needs a count. A goal about how long the student stays engaged needs a stopwatch. Pick the wrong method and you get clean-looking numbers that say nothing about the skill.
This guide gives a decision tree from the goal verb to the method, a trade-off table for the seven main methods, and the measurement page and data sheet for each. For the full process of running data across a caseload, start with the IEP data collection guide.
Start with the question the goal asks
The IEP has to say how progress on each goal will be measured,[1] so the method is part of the goal, not a detail for later. Progress monitoring works when the same brief measure is repeated over time,[2] which means the method you pick now is the one you will run every week until the annual review.
Every method answers one question. Before you look at data sheets, write down the question the goal asks.
| The goal asks | Method |
|---|---|
| How many items does the student get right? | Percent accuracy |
| Does the student use the skill when the chance comes? | Trials or opportunities |
| How often does the behavior happen? | Frequency or rate |
| How long does it last? | Duration |
| How quickly does the student start? | Latency |
| Is the behavior present across a long period? | Interval recording |
| How good is the product? | Rubric scoring |
| How much of the routine does the student do alone? | Task analysis |
If two questions fit, the goal is probably trying to do two jobs. Split it or pick the one that matters most for this student.
The decision tree: from goal verb to method
Work down the list. Stop at the first "yes."
- Does the student respond to a set of items you present? (solves, spells, reads words, answers, identifies) Use percent accuracy with the accuracy data sheet. Keep the item count the same each probe.
- Does the student need to use the skill when a natural chance comes up? (requests, greets, follows a direction, asks for help) Use opportunities with the opportunity data sheet. Each chance is one trial, marked independent, prompted or missed.
- Is it reading aloud for speed and accuracy? Use oral reading fluency with the reading fluency data sheet.
- Is it a behavior with a clear start and end that you want to count? (calls out, raises hand, initiates play, hits) Use frequency recording. If session lengths vary, report rate per minute. Both go on the frequency data sheet.
- Does the goal care how long the behavior lasts? (stays engaged, remains seated, tolerates, cries) Use duration recording with the duration data sheet.
- Is the problem getting started? (begins, responds within, transitions within) Use latency recording with the latency data sheet.
- Is the behavior too frequent or too continuous to count, over a long period? (on task, engaged, in area) Use interval recording with the interval recording data sheet.
- Does the student produce something judged on several features? (writes a paragraph, gives a presentation, draws a diagram) Use rubric scoring with the rubric scoring data sheet.
- Is it a routine made of steps? (washes hands, packs up, makes a snack, checks out a book) Use a task analysis with the task analysis data sheet.
- Is it about finishing assigned work? (completes, turns in) Use work completion, scored as tasks done out of tasks assigned on the accuracy sheet.
When you run the same probe for a group, the group data sheet keeps every student on one page.
Trade-offs: what each method is good at and what it misses
No method sees everything. The table shows the cost of each one in time and what it leaves out.
| Method | Time to collect | What it misses | Best goal types |
|---|---|---|---|
| Percent accuracy | 3 to 10 minutes per probe | Speed, and whether the skill is used outside the probe | Math facts, spelling, comprehension questions |
| Trials or opportunities | Seconds per trial, spread over the day | Quality of the response beyond yes or no | Requesting, following directions, social initiations |
| Frequency or rate | Seconds per tally, whole session watched | How long each episode lasted | Call-outs, hand raising, aggression with clear episodes |
| Duration | A stopwatch for each episode | How often episodes start | Engagement, tantrums, time on a task |
| Latency | One timing per cue | What happens after the student starts | Task initiation, transitions, response to directions |
| Interval recording | Needs a timer; partial or whole needs a dedicated observer | Exact counts and exact durations | On task, engagement, stereotypy across long periods |
| Rubric scoring | 2 to 5 minutes per product | Process, unless the rubric includes it | Paragraph writing, presentations, projects |
| Task analysis | 1 to 2 minutes per routine | Fluency and speed of the routine | Daily living, arrival and dismissal, vocational tasks |
Two of these pairs are confused often enough to have their own guides: frequency vs duration recording and percent accuracy vs percent of opportunities.
Step-by-step: choosing for one goal
Step 1: Underline the verb and the criterion
Take the goal as written. Underline the action and the number. "Given a direction to begin independent work, Leo will start the task within 60 seconds in 4 of 5 work periods." The verb is "start." The criterion is a time. That points to latency.
Step 2: Run the decision tree
Go down the list and stop at the first match. Leo's goal fails the first five questions and matches question 6. Latency it is. If the goal had said "will stay on task," it would match question 5 or 7 instead.
Step 3: Check it against the baseline
The baseline must use the same method. If present levels say Leo is "off task 60% of the time," that is interval data, not latency. Either take a new latency baseline or rewrite the goal to match the data you have. For how to take one, see how to collect baseline data.
Step 4: Test it for a week
Collect three sessions with the chosen method before you commit. Ask two questions. Can the person collecting do it while teaching? Does the number move when the behavior changes? If either answer is no, switch methods now, not in March.
Worked example: one student, three goals
A fictional 5th grader, Mia, has three draft goals. Here is how each method was picked.
| Goal | Verb | Tree stop | Method | Data sheet |
|---|---|---|---|---|
| Will solve two-step word problems with 80% accuracy | Solves items | 1 | Percent accuracy, 10 problems per probe | Accuracy |
| Will ask for help when stuck in 4 of 5 opportunities | Asks when a chance comes | 2 | Opportunities | Opportunity |
| Will stay engaged in independent writing for 15 minutes | Stays (lasts) | 5 | Duration | Duration |
Baselines from three probes each: 4 of 10 word problems, 1 of 5 help requests, and a median of 5 minutes engaged. Each baseline uses the goal's own unit, so the first progress report can compare like with like.
The writing goal first looked like an interval goal. The team chose duration because Mia either writes steadily or stops for long stretches, and the length of the writing streak was the real concern.
Common mistakes
- Weak: Percent accuracy for "will remain on task." There are no items to be accurate on. Better: Duration, or interval recording for long periods.
- Weak: Frequency for tantrums that last 2 to 25 minutes. Three short tantrums and one long one look the same as four. Better: Duration per episode, plus a count.
- Weak: "In 4 of 5 trials" for a skill you can only observe once a week. Five trials take five weeks. Better: Plan trials you can present in one session, or count opportunities over several days.
- Weak: A rubric with "good, fair, poor." Two adults will score it differently. Better: A rubric where each point has a visible description.
- Weak: One method for every goal on the caseload because one sheet is easier. Better: Two or three sheet types that match the goals.
Tools for the next step
Once the method is chosen, set up the matching sheet from the free data sheets and write the operational definition at the top. Check the goal wording in the goal checker so its criterion names the same unit you will record.
Frequently asked questions
What is the most common IEP data collection method?
Percent accuracy and trial or opportunity data are the most widely used, because many academic and communication goals are scored item by item. They are not right for every goal. Behavior that lasts or needs to start faster calls for duration or latency instead.
Can one goal use two data collection methods?
It can, but it usually means the goal is doing two jobs. If you need both a count and a length for the same behavior, record both, and say in the goal which one the criterion uses. Otherwise split it into two goals.
What if the method in the IEP does not fit the goal?
Raise it with the IEP team. Changing how a goal is measured is a change to the IEP, so it needs the team's agreement, often through an amendment. Bring a week of data showing why the current method does not capture the skill.
Use it in IEP Atlas
- Measurement methods
- Free IEP data sheets
- Data sheet in the IEP glossary
- Latency recording guide
- Check a goal you wrote
Related guides
- IEP data collection: methods, sheets and schedules that hold up
- Frequency vs duration recording: which one fits the behavior
- Percent accuracy vs percent of opportunities: what the denominator means
- Partial interval, whole interval and momentary time sampling compared
- Digital vs paper IEP data collection
- Generalization data across settings
- Cold probe data vs trial-by-trial data
Sources
- 34 CFR §300.320: Definition of individualized education program, Legal Information Institute, Cornell Law School
- Progress Monitoring: Data & Goal Setting, 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.