Interval Recording and Time Sampling

Introduction to Discontinuous Measurement and Session Setup

In behavior analysis, continuous measurement captures every occurrence or the full duration of a behavior. However, continuous observation of every behavior instance is not always feasible during everyday service delivery. In these situations, supervisors often select discontinuous measurement. Discontinuous measurement records whether a behavior occurs during or at specific moments across divided periods of time. Rather than generating an exact count or duration, discontinuous measurement produces an estimate of behavior. These procedures are frequently used for behaviors that do not have discrete, easily counted start and end points.

Interval recording requires dividing an observation session into equal, predetermined blocks of time. The supervisor specifies the interval length suited to the behavior and setting. Common interval lengths range from brief intervals like five or ten seconds to several minutes. The three primary discontinuous measurement procedures used by Registered Behavior Technicians are partial-interval recording, whole-interval recording, and momentary time sampling. Using discontinuous measurement when an exact frequency is needed can lead to incorrect conclusions about client progress, so the technician must always follow the specific system chosen by the supervising BCBA or BCaBA.

Before collecting interval data, the RBT reviews the supervisor-provided operational definition of the target behavior and asks for clarification if needed. Clear, supervisor-provided operational definitions help RBTs score intervals consistently. During the session, the RBT uses a supervisor-approved timing tool, such as a vibrating interval timer or approved app, to track the specified intervals accurately. An RBT records a positive mark for an occurrence and a negative mark for a non-occurrence based on the clinic's data sheet format. To maintain high accuracy, an RBT must collect and record interval data immediately during the session rather than relying on memory later.