DRAGEN Academy · Free lesson · BCBAs, teachers, parents reading an IEP · 8 minutes
How to write a goal a stranger could measure.
What it is. A measurable goal has four parts: the condition (when, where, with what), the behavior (what the person will do, in words you can see), the criterion (how well, how often, over how long, with whom), and the date. Plus the baseline: what they do now.
Why it matters. “Will improve social skills” cannot be met or missed. Reviewers send it back, and worse, nobody can tell whether the child is better. A goal a stranger can measure is a goal the team can be honest about.
The method
How to write one yourself: eight steps, then the stranger test.
- Start from the problem in plain words. “He does not ask for help; he tears the paper.”
- Name the behavior you will see instead, as a “do.” “Asks for help by saying ‘help please’ or handing over the help card.”
- Add the condition. “When given an independent task he cannot finish, in the classroom…”
- Set the criterion — accuracy (80% of opportunities), consistency (across three consecutive weeks, or five sessions), and spread (two settings, two adults). The spread is what makes a skill real instead of a trick for one room.
- Add the date. Six months out, or the end of the IEP year. A date already past is a goal already failed.
- Write the baseline directly under it: what they do now, with the number and the dates it was measured. No baseline, no progress — there is nothing to compare to.
- For funded treatment, add the “so that.” Why this skill matters for this person: “…in order to replace paper-tearing, which currently ends four of five work periods.” Reviewers ask for it every cycle.
- The stranger test. Hand the goal to someone who has never met the child. Can they say yes or no, from data alone, whether it was met? If they have to ask you anything, rewrite.
Worked example
Noah, before and after.
| Before | After |
|---|---|
| “Noah will improve communication.” | “By March 2027, when given an independent task he cannot complete in the classroom, Noah will request help by saying ‘help please’ or handing the help card in 80% of opportunities across three consecutive weeks, with two different adults, in order to replace paper-tearing. Baseline: 0 of 12 opportunities, observed 9/8–9/12/2026.” |
Mistakes to avoid
The goals that come back from review.
- “Will decrease behaviors.” Which behaviors? From what, to what?
- A percentage with no number of opportunities. 80% of two chances is one good afternoon.
- “With 80% accuracy” on something that is yes-or-no once a day. Use days out of days.
- A criterion nobody is collecting data for. If it will not be measured, it is a wish.
- Last year’s goal with the date moved. The stranger test catches this: the baseline will be a year old.
- Upgrading the words when the data did not move. “Emerging” does not become “acquired” because a report is due.
Try it
One question, then the answer.
“Mateo will improve his listening skills with 80% accuracy by June.” What fails the stranger test?
Show the answer, and why
Almost everything. “Listening skills” is not something you can see happen; there is no condition, no number of opportunities, no consistency or spread, and no baseline. A stranger could not say yes or no. One rewrite: “By June 2027, when given a one-step instruction in the classroom, Mateo will begin the instruction within ten seconds in 80% of ten opportunities across five consecutive days, with two different adults. Baseline: 2 of 10 opportunities, measured 9/14–9/18/2026 (made up).”
Something to keep
Goal-writing template.
Fill the blanks, read it aloud, then hand it to someone who has never met the learner.
DRAGEN Academy · Goal-writing template · dragen.cloud
| By [date] | when [condition: where, with what, after what] | [name] will [observable behavior, a “do”] |
|---|---|---|
| Criterion: how accurate | How consistent (sessions / weeks in a row) | Spread (settings, people) | …in order to [why it matters] |
|---|---|---|---|
| Baseline (number, dates measured) | Data collected by whom, how, how often | Stranger test passed — who read it, date |
|---|---|---|
Print shows only the sheet, in black on white. Nothing you write on it is sent anywhere; the page has no forms.
Where DRAGEN helps
Taught to teams, with practice.
Academy workshops teach goal-writing the way this page does — write one, have it read by a stranger, fix it, again. DRAGENgrade, in development, applies the same criterion thinking to classroom rubrics.
Source
Where this comes from.
Mager, Preparing Instructional Objectives (3rd ed., 1997), for condition–behavior–criterion; Cooper, Heron & Heward, Applied Behavior Analysis (3rd ed., 2020), on operational definitions and selecting target behaviors. Funders publish their own report requirements; where yours asks for more, it wins.
Examples on this page are made up. They contain no client, student or family information. This lesson teaches a method; it does not replace your child’s care team, your BCBA, or your payer’s written rules.