Continuous measurement in ABA refers to the tracking of every instance of a target behavior during a specified time frame. This process can help therapists determine proper treatment.

Applied behavior analysis (ABA) is a type of therapy based on the way a person learns from the consequences of their own behavior.

ABA is well known as a therapy for autism. Allistic people can also receive this treatment, for a variety of issues including substance use disorders and brain injuries. ABA is also used in areas like education, health and fitness, and gerontology.

A key component of ABA is identifying behaviors to change and tracking those behaviors. The target behaviors are ones that interfere with a person’s well-being, like combative social interactions or self-harm.

For an ABA therapist to design a successful intervention for a person, they need a systematic and structured behavior-tracking method.

A registered behavior technician (RBT) uses continuous measurement in the field of ABA. Continuous measurement is the tracking of every instance of a person’s target behavior, for a specified duration.

Timeframe examples include a school day, a single class, or a therapy session.

Collecting continuous measurement data requires more effort than discontinuous measurements, which are samples taken from small time intervals. Continuous measurement can be challenging in busy environments, and it requires the RBT to be there for the entire timeframe.

But continuous measurement is more accurate than discontinuous methods, which can miss behavior instances and underestimate their impact.

Continuous vs. discontinuous measurement

Continuous measurement is when an RBT records every instance of a target behavior that occurs within a specified time frame, like an entire class.

Discontinuous measurement is a behavior sample recorded during a short time span, divided into smaller, equal intervals.

There are three types of discontinuous measurement samples:

  1. Partial interval: A response is logged when the target behavior occurs at any time during the interval.
  2. Whole interval: A response is logged when the target behavior occurs during the whole interval.
  3. Momentary time sampling: A response is logged when the target behavior occurs at the end of the interval.
Was this helpful?

There are several characteristics of a target behavior that an RBT measures and logs.


Frequency is how often something occurs. To obtain the frequency of a behavior, the RBT simply counts the number of times the target behavior occurs.

The RBT might use a tool like tally marks, a phone counter app, or a clicker counter. Or, they might keep track of the count in their head.

For frequency to be an appropriate measure, the behavior should have a clear start and finish.


Rate is frequency divided by time.

For example, if a student runs out of the classroom six times in 2 hours, the rate is three times per hour.

The unit of time doesn’t have to be per hour. It can be per minute, or whatever unit of time works with the measurement being done.


Duration is how long a behavior lasts. To determine the duration, an RBT starts a timer when the behavior begins and stops it when the behavior ends.

The total duration is the sum of individual duration measurements. For example, if a child has a 5-minute tantrum and a 7-minute tantrum, the total tantrum duration is 12 minutes.


Latency is the time between the discriminative stimulus (SD) and the start of the person’s response.

An SD is the prompt that starts a behavior, like an instruction given from an adult to a child. The time between that instruction and the child’s response is the latency.

Interresponse time (IRT)

IRT is the time between two responses to the same SD.

The responses should be the same type of behavior. For example, a child who yells then kicks is exhibiting two different types of responses, so IRT wouldn’t apply. But if the child kicked twice, the IRT is the time between the first kick and the second.

Trials to criterion

This is the number of times it takes a person to reach mastery of a skill. Trials mean attempts, and criterion is what the therapist has chosen for the person to master.

For example, the criterion might be for a student to raise their hand rather than shouting out an answer. If the teacher asks four questions to which the student shouts but then raises their hand to the fifth question, this represents five trials to criterion.

Percentage of occurrence

This is the number of responses divided by the number of opportunities.

For example, imagine an RBT is observing a student’s responses to a classmate walking past their chair. If the seated student grabs the arm of the passing student 6 out of 10 times, the percentage of occurrence is 60%.

Continuous measurement is a valuable part of ABA data collection because it allows the therapist to tell whether behavior is changing, and if planned interventions are having the intended effect.

A person’s treatment may continue as is if the data indicates that they’re making progress. If not, their therapist can modify their strategies.