Skip to Main Content

The Process of Positive Behavior Support (PBS)

Step 6: Monitoring Outcomes

previous page page 5 of 7next page

Direct Measurement

Direct measurement provides the most accurate representation of a child’s behavior or skill acquisition and is the preferable approach to use to measure change. However, the use of direct measurement by busy families and providers is not often possible. Indirect measurement may offer a more user-friendly approach.

Event Recording

The number of times a skill is observed. This is measured by counting the number of times a behavior occurs. It is important that a specific behavior has a clear beginning and ending point in order to ensure accurate measurement. An example of even recording might be recording the number of times a child leaves the table in a 20-minute snack time. This method is also referred to as frequency recording.


Another relatively easy way to collect data is to determine the percentage in which a behavior occurs. In order to do this, simply identify the number of times a behavior occurs, divide it by the total number of chances the child had to perform the behavior, and multiply by 100. For example, if a child is given 10 chances to stack three blocks and successfully stacks 3 blocks seven times, the percentage would be 70%.


Rate entails measuring the number of times a behavior occurs relative to a period of time. This is calculated by measuring the number of occurrences divided by a number of time units. For example, a teacher using rate measurement might calculate the number of times per minute a boy bites his hand. Although a useful tool, this can be difficult to track when measuring high frequency behaviors.

Interval Recording

Interval recording is similar to event recording and measuring rate but differs in that behavior is recorded within small blocks of time (e.g., 10-seconds) called intervals that together represent a longer period of time (e.g., 10-minutes). Observers use a data collection form that contains a series of intervals. Behaviors can be scored as occurrences if they occur at least once at any time during the interval (i.e., partial interval recording) or if they occur for the entire duration of an observation interval (i.e., whole interval recording). In general, partial interval recording is most often used to measure behaviors that are likely to be short in duration (e.g., hitting, biting), whereas whole interval recording is most often used to measure behaviors that are expected to occur continuously for a period of time (e.g., sitting in seat).

Unlike event recording, each discrete behavior being observed is measured on relative to distinct intervals of time. At the end of the interval period, the observer makes a check mark in the corresponding area to mark the occurrence of a behavior. The observation continues to the end of the next interval and another mark is made if the behavior recurs. Scores for interval recording are calculated as a percentage—the number of occurrences are divided by the number of opportunities, and then multiplied by 100.


The amount of time behavior occurs in an observation period (e.g., the amount of time the child sits at the table to eat). Measured by recording the amount of time a behavior occurs. Unlike the previous measures, duration is a measure that is useful for measuring how long a continuous-ongoing behavior lasts. Like event recording, duration is a direct measure of the actual behavior, not an estimate like interval recording.

The most precise way to observe and record duration of a behavior is to use a stopwatch or, if this is not possible, a watch with a second hand. As discussed previously, as with all behaviors, it is important that the behavior being measured with duration be accurately defined with a clear beginning and end so that the observer knows when the record should begin and when it should end.


The elapsed time between the instruction and the time the child initiates the behavior (e.g., how quickly the child initiates playing with a toy once it is presented). The way this is measured is by beginning timing once a cue is presented and continue timing until the child begins to correctly respond to the cue. A common example where latency is used is in school, when a teacher might be interested in learning how long it takes a student to begin work after she gives a direction to begin working.

Momentary Time Sampling

Similar to interval recording, momentary time sampling provides an estimate of behavior occurrence, but in this case, the behavior is only scored at the end of a interval of time. Using this procedure, observers simply check whether a behavior occurred precisely at the end of each interval (e.g., at the 10th second of a 10-second interval). Scores are calculated as percentages of the total number of observed intervals.

previous page page 5 of 7next page