What is interval recording in ABA?
Interval recording is a family of time-sampling methods that divide a session into short intervals and score each one for whether the behavior occurred, producing an estimate of behavior when counting every instance isn't feasible.
Continuous measurement is the gold standard, but an observer running a classroom or a group cannot count every instance of everything. Interval methods trade some precision for feasibility: the session is chopped into brief, equal intervals, and each interval is scored yes or no.
The three variants differ in what earns a yes. Partial-interval scores the interval if the behavior occurred at any point in it - sensitive, but it tends to overestimate, since one second of behavior scores the same as thirty. Whole-interval scores only if the behavior lasted the entire interval, which tends to underestimate and suits behaviors you want more of, like on-task engagement. Momentary time sampling looks only at the instant each interval ends - remarkably practical, since the observer attends just at the beeps, at the cost of missing everything between them.
The choice is a bias decision made on purpose. Overestimating a dangerous behavior is a safe error, so partial-interval fits reduction targets; underestimating engagement pushes teams to keep teaching, so whole-interval fits acquisition of sustained states; momentary sampling fits busy environments and multiple learners at once.
Interval data is reported as the percentage of intervals scored, and it is an estimate, not a count - a fact worth remembering when comparing across methods or when a payor report needs precise language about what the number represents.
Interval sizes are kept short and consistent, and the method is documented with the data, because the same behavior sampled with different methods or interval lengths produces different numbers. Changing the measurement mid-program without noting it is one of the classic ways a graph lies.