Writing intervention goals in observable, measurable terms

A goal is a promise to the client and family: this is what will change, and this is how everyone will know. If the goal is vague, nobody can keep that promise or show it was kept.

Why it matters

Every later decision in a treatment plan rests on the goal. Your measurement system, your mastery decisions, your progress reports, and your conversations with caregivers and funders all depend on a goal that two people would read the same way. "Ellie will improve her social skills" gives an RBT nothing to teach and gives you nothing to graph. "During recess, Ellie will start a play exchange with a peer at least 3 times per 20-minute recess across 5 consecutive school days" tells everyone what to look for and when to celebrate.

In practice items on this task, you spot the missing piece of an objective, pick the best rewrite of a vague goal, or tell a condition from a criterion.

Key ideas

• Name the behavior, not a label or an inner state. Use action verbs that someone can see or hear: say, write, point to, hand, sit, complete. Avoid understand, know, appreciate, be aware of, or labels like aggressive or respectful.
• A complete objective has four parts.
  1. The learner (who).
  2. The target behavior, stated in observable terms (what).
  3. The conditions: the setting, materials, instructions, or level of help under which the behavior should occur (when and where).
  4. The criterion: the level of performance that shows the goal is met, such as accuracy, rate, duration, or latency, often across several sessions so you know it is stable (how well).
• Many teams add a target date and plans for generalization and maintenance, such as performance with a new person or in a new place, or at a follow-up check.
• Pass the dead man's test. If a dead man could "do" it, it is not behavior. "Will not hit peers" or "will stay quiet" describe the absence of behavior. Pair any reduction goal with what the learner will do instead.
• Set criteria that matter in daily life. A criterion should match what the learner's environment needs. Looking at how same-age peers typically perform, or at what a job or classroom requires, helps you choose a level that is socially valid rather than arbitrary.
• Measurable goals build on operational definitions. The goal says what change you want; the operational definition spells out exactly what counts as an instance of the behavior.

Definitions

• Behavioral objective: a written statement of the learner, the observable target behavior, the conditions, and the criterion for success.
• Conditions: the antecedent circumstances under which the behavior is expected, such as setting, materials, instructions, or prompt level.
• Criterion: the standard of performance, stated as a measurable level, that shows the objective has been met.
• Dead man's test: a quick check that a target names something a person actively does; anything a dead man could do is not behavior.

Examples from a session

Notice that each revision answers who, what, when and where, and how well. Read your goal aloud and ask whether a new RBT, a caregiver, and a funder would all agree on whether today's session met it.

Common mistakes / exam traps

• Writing a goal for staff instead of the learner. "The RBT will prompt Leah to wash her hands" describes what staff do, not what Leah will do.
• Goals that only describe stopping. "Will not elope" fails the dead man's test on its own. Add the replacement behavior.
• Arbitrary criteria. A goal of "80% accuracy" for crossing the street safely is far too low. Pick the level the real situation demands.

Quick check

Related topics

• Writing operational definitions (C.1)
• Choosing a measure that fits the critical dimension (C.9)
• Prioritizing socially significant, client-informed, culturally responsive goals (F.8)
• Choosing socially valid replacement behavior (H.3)
• Data-based decisions about effectiveness and when to modify (H.7)