Partial interval vs whole interval vs momentary time sampling

All three methods split an observation into equal intervals and score each one, instead of recording every occurrence. They differ only in what earns a score, and that is why they give different numbers for the same session.

MethodScore the interval whenTends toOften used for
Partial intervalThe behavior happens at any point in itOverestimateBehaviors to decrease
Whole intervalThe behavior lasts the entire intervalUnderestimateBehaviors to increase, such as on-task
Momentary time samplingThe behavior is happening at the moment the interval endsEither wayGroups, or when you can only glance up

One session, three answers

A 6-minute observation uses 1-minute intervals. A client is out of their seat from 0:40 to 2:10 and from 4:30 to 4:50. That is 110 seconds out of 360, or about 31% of the session.

IntervalPartialWholeMomentary (at the end)
0:00 to 1:00YesNoYes
1:00 to 2:00YesYesYes
2:00 to 3:00YesNoNo
3:00 to 4:00NoNoNo
4:00 to 5:00YesNoNo
5:00 to 6:00NoNoNo
Result4 of 6, 67%1 of 6, 17%2 of 6, 33%

Partial interval reads well above the real 31% and whole interval well below it. Momentary time sampling lands close here, but it missed the 20-second episode completely, so it can miss short behaviors.

How exam-style questions ask this

Try the free 20-question practice test, or read about DRA, DRI, DRO and DRL and the four functions of behavior.

Want three full-length practice exams? The Behavior Technician Exam Prep Pack costs $32 on Gumroad: a PDF study guide, three 85-question practice exams with answer keys and rationales, and 96 flashcards. It is our own paid product, so this link is a commercial link and not a neutral recommendation.

Report a problem — tell us what failed; we reply to your email when it is fixed.