SessionTap

Recording guides

Partial interval recording, whole interval recording and momentary time sampling

Interval methods split an observation into short, equal intervals (often 10 to 30 seconds) and record for each interval whether the behavior met a rule. The result is the percentage of intervals, an estimate of how much the behavior happened.

The three methods

Partial interval recording. Mark the interval + if the behavior happens at any point during it, even for a second. Once it has happened, you can stop watching for that interval. It suits behaviors that are frequent, brief or hard to count, such as hand flapping, vocal stereotypy and call-outs in a busy room. It tends to overestimate how much of the time the behavior is happening, because a one-second occurrence fills the whole interval. Because it overestimates, it is the usual choice for behaviors you want to reduce: if the percentage still falls, the improvement is unlikely to be an artefact of the method.

Whole interval recording. Mark + only if the behavior lasts the entire interval. It suits behaviors that should continue, such as on-task behavior, engagement and cooperative play. It tends to underestimate, because a brief break spoils the interval, so it is the usual choice for behaviors you want to increase, for the same reason in reverse.

Momentary time sampling (MTS). Look only at the moment each interval ends and mark + if the behavior is happening right then. It suits longer-lasting behaviors and observers who can't watch continuously, such as a teacher checking engagement every 2 minutes while teaching. It can miss brief behaviors entirely.

Partial interval Whole interval Momentary time sampling
Mark + when It happens at any point It lasts the whole interval It is happening at the interval's end
Error Overestimates Underestimates Either way; misses brief behaviors
Observer effort Continuous, but can stop once it happens Continuous A glance per interval

Scoring

percent of intervals = intervals marked + ÷ total intervals × 100

24 of 60 intervals marked + is 40%. Report it as "40% of intervals", never as "40% of the time" or a count: it is neither.

Choosing the interval length

Shorter intervals give a closer estimate but are harder to keep up with. Ten seconds is common for partial interval in a session; MTS in a classroom often uses 1 to 5 minutes. Whatever you pick, keep it the same from one observation to the next, or the percentages can't be compared.

On paper

The free interval recording data sheet has 60 intervals for each of two behaviors and a box to tick the method. Use a vibrating or beeping interval timer so your eyes stay on the person.

Is there a more direct measure?

Interval methods are estimates. When you can count every occurrence or time every episode, frequency or duration recording gives a direct number instead. That is what SessionTap records on iPhone: it does not do interval recording or momentary time sampling yet, but counting or timing on a phone is often no harder than marking intervals on paper, and the app keeps the timestamps.