How interval recording works
First, you divide the observation period into equal intervals, often 10 to 30 seconds long. Then, for each interval, you record whether the behavior met the criterion for that method. The result is usually reported as the percentage of intervals scored, which serves as an estimate of how much the behavior occurred.
Because it yields an estimate rather than an exact count, interval recording trades some precision for practicality. It is well suited to behaviors that occur too often or too continuously to count directly.
Whole-interval recording
In whole-interval recording, you score an interval only if the behavior occurred for the entire interval. This method tends to underestimate behavior, because any pause within the interval means it is not scored. Whole-interval recording is often used for behaviors you want to increase and maintain continuously, such as staying on-task.
Partial-interval recording
In partial-interval recording, you score an interval if the behavior occurred at any point during it, even briefly. This method tends to overestimate behavior, because a single brief instance counts the whole interval as occurring. Partial-interval recording is often used for behaviors you want to decrease, such as disruptions.
Momentary time sampling
In momentary time sampling, you score the interval based only on whether the behavior is occurring at the moment the interval ends. Because the observer only needs to glance up at set moments, this method is efficient and tends to neither systematically over- nor under-estimate as strongly as the other interval methods, though it can miss behavior between samples. It is covered in more detail in its own article.