cumulative

C.1 Writing operational definitions of behavior

What the task asks: Write a definition of a target behavior precise enough that different observers would count the same instances, and choose between defining it by its function or by its form.

Measurement, Data Display, and Interpretation5 min read4 practice questionsOfficial outline (opens in a new tab)

The big idea

An operational definition describes a target behavior in terms that can be observed and measured, so that anyone who reads it would count the same instances. It turns a label like “aggression,” “off task” or “meltdown” into a description of what the person actually does.

Every other step of measurement depends on it. If two observers picture different behaviors, their data will disagree, the graph will mislead, and nobody can tell whether the intervention worked. A precise definition also lets another practitioner repeat the work, which is part of what makes an intervention technological (A.5).

Breaking it down

Objective, clear and complete

A widely used standard (Hawkins and Dobes, 1977) holds that a good definition is:

  • Objective: it refers only to what can be seen or heard, never to feelings, intentions or traits. “Throws objects” is objective. “Is angry” is not.
  • Clear: it is easy to read and leaves little room for interpretation, so a new observer could use it without asking questions.
  • Complete: it marks the boundaries, saying what counts, what does not, and how to handle borderline cases.

What a working definition contains

  1. A short, descriptive label.
  2. A description of the response in observable terms.
  3. Examples: instances that count, especially ones near the boundary.
  4. Non-examples: responses that look similar but do not count.
  5. When needed, onset and offset rules that say when an episode starts and ends, for example “a crying episode ends after 10 seconds without crying.” These matter for duration (C.4) and for deciding whether two bursts are one episode or two.

Function-based vs. topography-based

A function-based definition counts any response that has the same effect on the environment, whatever it looks like. “Moving more than 3 meters from the group without permission” covers walking, running and crawling, because each produces the same outcome. Function-based definitions tend to be simpler and to capture the whole response class (B.1), so they are usually the first choice when the effect can be observed.

A topography-based definition describes the form of the response: its shape, movement or sound. “Striking another person with an open or closed hand” is topography-based. It is the better choice when the form itself is the goal (letter formation, pencil grip, the steps of hand washing) or when the effect of the behavior cannot be observed.

Function-based A response counts by its effect Walks off Runs off Out a door Over 3 m from the group without permission All three count Topography-based A response counts by its form Open hand Closed fist Kick Counts Counts Not named
Ms. Estrada's elopement definition counts walking off, running off and slipping out a door alike, because each has the same effect. A topography-based definition of hitting counts only the forms it names, so a kick is left out even when it has the same effect.

“Function” here means the effect the response has, not the person’s reason. “Leaves the area” is function-based. “Leaves the area to avoid work” adds a guess that observers cannot see and that belongs in assessment (F.5, F.6).

The dead man test

Ogden Lindsley’s quick check: if a dead man could do it, it is not behavior. “Sits quietly,” “does not hit” and “stays calm” all fail, because a dead man could do each of them. Define what the person does instead, such as “keeps hands on the desk and writes,” or measure the problem behavior directly.

Seeing it in practice

LabelWeak versionOperational definition
Aggression (Tomás, 7, clinic)“Gets aggressive when upset”Hitting, kicking or pushing another person with enough force to make a sound or move their body. Not included: high-fives, tapping a shoulder to get attention.
On task (Aisha, grade 5)“Pays attention in class”Eyes on the teacher or the assigned material while writing, reading or raising a hand as the lesson requires. Not included: drawing on the worksheet, looking out the window.
Elopement (Ms. Estrada, 34, day program)“Runs away”Moving more than 3 meters from the group, or through any exit, without staff permission, at any speed.
Hand biting (Lin, 12, home)“Self-injures when frustrated”Closing teeth on any part of the hand or wrist. Not included: sucking a finger with no tooth contact.

Each better version names actions, sets a boundary with non-examples, and drops guesses about feelings. The elopement definition is function-based. The others are topography-based, because the form of contact is what observers must judge.

Before collecting baseline, try the definition out: have two people observe the same session and compare their records (C.8). Disagreements point to the part of the definition that needs another example or non-example.

Common mix-ups

  • A label is not a definition. “Noncompliance” names a category. The definition says what to count.
  • Inferred states slip in. Words like frustrated, anxious, wants attention or on purpose cannot be observed. Describe the action and leave function to assessment.
  • Function-based does not mean “includes the reason.” It groups responses by their effect on the environment.
  • Too narrow or too broad. “Hitting with a closed fist” misses slaps. “Touching others” sweeps in hugs. Test the definition against real examples and non-examples.
  • Dead man behaviors. Goals stated as the absence of behavior fail the test. Restate them as something the person does.

Exam tip: When a question asks what is wrong with a definition, look for a word that names a feeling, intention or trait. When it asks what type a definition is, ask whether a response counts because of what it looks like (topography) or because of what it does (function).

Quick recap

  • An operational definition describes behavior in observable, measurable terms so observers count the same instances.
  • Good definitions are objective, clear and complete, with examples, non-examples and onset and offset rules when needed.
  • Function-based definitions group responses by their effect; topography-based definitions group them by their form.
  • If a dead man could do it, it is not behavior.
  • Try the definition with two observers before relying on the data.

Further reading

  • Cooper, J. O., Heron, T. E., & Heward, W. L. (2020). Applied behavior analysis (3rd ed.), chapter 3.

Key terms

Operational definition
A description of a target behavior in observable, measurable terms, written so that different observers would agree on which instances count. A good definition is objective, clear and complete, and often includes examples, non-examples and rules for when an episode starts and ends.
Function-based definition
An operational definition that counts a response as an instance of the target behavior because of the effect it has on the environment, whatever form the response takes. It describes the effect, not the person's reason for behaving.
Topography-based definition
An operational definition that identifies instances of the target behavior by their form: the shape, movement or sound of the response. It fits goals where the form itself matters or where the effect of the behavior cannot be observed.
Response class
A group of responses treated as one unit because they share something. Members of a functional response class produce the same effect on the environment, however different they look; members of a topographical response class share the same form. Used without a qualifier, the term usually means a functional class.

Check yourself

4 questions on C.1. Choose an answer to see whether you're right and why.

  1. Question 1A teacher drafts this definition: 'Disruption: Ravi acts out and gets frustrated when the work is too hard.' What is the main problem with it?
    Show the answer

    Answer: D. 'Acts out' and 'gets frustrated' are a label and a guess about feelings, so two observers could count very different things. A usable definition names what Ravi does, such as tearing paper or leaving his seat. Length and the measurement procedure are not the issue, and a definition never needs to name a reinforcer.

  2. Question 2A school team defines elopement as 'moving more than 3 meters from the assigned area, or through any exit, without adult permission, whether walking, running or crawling.' What kind of definition is this?
    Show the answer

    Answer: A. The definition counts a response because of its effect (the student ends up away from the area), and walking, running and crawling all qualify. That makes it function-based. A precise distance does not make it topography-based, and a function-based definition describes the effect on the environment, not the person's reason, which observers cannot see.

  3. Question 3Two observers keep disagreeing about whether Ana tapping a peer's shoulder to get attention counts as 'hitting.' What change to the definition would help most?
    Show the answer

    Answer: C. The disagreement is about a boundary, so the fix is to make the definition complete: state which kinds of contact do not count and how to score unclear cases. Averaging the counts hides the disagreement instead of resolving it, and longer sessions or a guess about function do not tell observers where the boundary is.

  4. Question 4When is a topography-based definition the better choice?
    Show the answer

    Answer: D. Topography-based definitions describe the shape of the response, so they fit goals about form, such as letter formation, pencil grip or the steps of hand washing. Including every form with the same effect is the strength of a function-based definition. Permanent products and high rates are measurement questions, not reasons to define by form.

Adds a step to your study record. Saved in this browser only.

Type to search all 104 guides and the glossary.