IEP baseline data: how to collect it, how much you need, and what it should look like
Quick answer: a defensible baseline is three to five data points, collected with the same measure the goal will be scored with, summarized as the median, and written into the PLAAFP in the goal's own units. That standard comes from the IRIS Center's progress-monitoring brief (Vanderbilt, developed under a U.S. Department of Education grant), and it exists because everything else in the goal depends on this number: the annual target is set from it, the goal line is drawn from it, and every progress decision is a comparison against it. Federal law never uses the word "baseline" — but it requires a statement of present levels of academic achievement and functional performance (34 CFR §300.320(a)(1)) and measurable annual goals (§300.320(a)(2)), and a goal can't be measurable relative to a starting point nobody measured.
The five-step method
- Define the skill in observable terms. "Reading fluency" isn't observable; "words read correctly per minute on an unfamiliar grade-level passage" is. If two adults could score the same session and get different numbers, the definition isn't tight enough yet.
- Pick the measure the goal will live on. The IRIS brief's test is worth memorizing as an example/non-example pair: for a 115-wcpm reading goal, an oral reading fluency probe measures the goal; an end-of-chapter comprehension test does not. For a peer-interaction goal, an observation frequency count measures it; the teacher's general impression does not.
- Collect 3–5 data points. Enough for a stable picture, per IRIS. Spread them across different days — three probes in one sitting measure one bad or good day three times.
- Take the median, not the mean. Scores of 5, 7, and 8 give a baseline of 7. The median resists the outlier day; the mean doesn't. This median is the point the goal line starts from.
- Write it into the PLAAFP in the goal's units. Baseline 41 wcpm → goal in wcpm. Baseline 40% of intervals on task → goal in % of intervals. If the units change between baseline and goal, progress can't be computed — this is one of the most common findings when IEPs are audited.
Weak vs. strong baseline statements
All students below are invented composites — sample statements, not real records. The pattern to copy: probe type, dates or window, the raw scores, the median, and a comparison point that shows the size of the gap.
| Area | Weak (unusable) baseline | Strong baseline (sample) |
|---|---|---|
| Reading fluency | Marcus struggles with reading fluency and reads below grade level. | Across three 1-minute oral reading fluency probes on unfamiliar 3rd-grade passages (9/8, 9/10, 9/12), Marcus read 41, 38, and 44 words correct per minute — median 41 wcpm (3rd-grade fall benchmark: 83). |
| Behavior (on-task) | Dani is frequently off task during independent work. | During five 15-minute independent-work observations using 1-minute momentary time sampling, Dani was on task in 40%, 33%, 47%, 40%, and 53% of intervals — median 40% (classwide comparison peer: 87%). |
| Written expression | Awa has difficulty writing complete sentences. | On three curriculum-based writing probes (10/1–10/15), Awa wrote 3, 5, and 4 complete sentences in 15 minutes, with correct capitalization and end punctuation in 2 of 12 total sentences — median 4 sentences, 17% mechanics accuracy. |
| Math computation | Leo is behind in math facts. | On three 2-minute mixed addition/subtraction probes (fall week 3), Leo scored 8, 11, and 9 digits correct — median 9 digits correct per 2 minutes (2nd-grade fall norm: 20). |
From baseline to goal line
The baseline median is the left end of the goal line — the straight line drawn from the baseline score to the end-of-year target on the student's progress graph. Once instruction starts, each new data point lands above or below that line, and the IRIS decision rule takes over: with six or more data points collected, look at the four most recent. Mostly on or above the line — stay the course. Mostly well above — the goal may be under-ambitious; reconvene and raise it. Mostly below — change the instruction, not the data collection. The full cadence-by-goal-type system, decision rules included, is in IEP progress monitoring.
Baseline also disciplines the target itself. A 3rd grader at a 41-wcpm median with a 83-wcpm benchmark needs to close 42 wcpm in a year — about 1.2 wcpm per instructional week, an ambitious but real growth rate. A goal of "115 wcpm" for the same student isn't rigor, it's fiction, and the mid-year data will say so. Set the target by adding a defensible growth rate to the measured baseline — not by copying the grade-level benchmark into the goal.
Where baseline lives in the paperwork
The baseline numbers go in the PLAAFP — they are the "academic achievement and functional performance" the section is named for — alongside the impact statement and parent input the section also requires. The weak-vs-strong rewrite of a full present-levels section is in how to write a PLAAFP statement, and the goal built on top of that baseline follows the four-part formula in how to write measurable IEP goals. The chain is baseline → PLAAFP → goal → progress monitoring, and it's only as strong as the first link.
FAQ
How many data points do you need for an IEP baseline?
Three to five, per the IRIS Center's progress-monitoring guidance — enough to establish a stable, clear picture of current performance. Use the median (middle score), not the average, so one outlier day doesn't distort the starting point. A single data point is not a baseline; it's an anecdote.
What is baseline data in an IEP?
The measured level of a student's academic or behavioral performance before the IEP's supports are implemented — the starting point the annual goal is calculated from and the number the first progress report is compared against. It belongs in the PLAAFP in the same units as the goal.
Can you write an IEP goal without baseline data?
You can, but you shouldn't — and it's the fastest way to write an indefensible goal. Without a measured starting point there is no way to tell whether the annual target is ambitious, trivial, or impossible, and no way to compute whether the student is on track. If a goal is challenged, the first question is 'what was the baseline?'
Is baseline data the same as the PLAAFP?
No — the baseline is the number; the PLAAFP is the paragraph. The PLAAFP (present levels statement) contains the baseline data plus the context: how the disability affects involvement and progress in the general curriculum, strengths, parent input. Baseline data with no narrative is an incomplete PLAAFP; a narrative with no numbers is an unmeasurable one.
What should you use to collect baseline data?
The exact measure the goal will be scored with. If the goal is words correct per minute, baseline is an oral reading fluency probe — not an end-of-chapter comprehension test. If the goal is frequency of a behavior, baseline is an observation count — not a teacher's impression. Baseline and progress monitoring must be the same yardstick or the goal line means nothing.