Current measurement in applied behavior analysis.
Short-interval momentary time sampling is now evidence-based, so drop the old ban.
01Research in Context
What this study did
Springer et al. (1981) wrote a think piece. They said stop using momentary time sampling. They wanted every second of behavior recorded.
The paper is theoretical. No kids, no clinic data, just argument.
What they found
The authors claimed discontinuous methods lose real changes. They said only continuous data show true patterns.
How this fits with other research
LeBlanc et al. (2020) ran the numbers. Three-minute momentary time sampling matched continuous data. Their evidence flips the 1981 advice.
Grauerholz-Fisher et al. (2019) used 15-second momentary time sampling to track care quality in an ABA clinic. It worked. The clash is only skin-deep: the 1981 paper worried about basic behavior properties, the 2019 paper tracked staff and room hazards.
Shimp (2020) still loves continuous data but adds a twist. Plot both 10-second counts and session totals. This bridges the tiny moments with the big picture.
Why it matters
You can relax. Short interval momentary time sampling is fine for most clinic work. Keep continuous if you study fast behavior or publish basic charts. Try a 3-minute MTS next time you watch staff or self-help responses.
Want CEUs on This Topic?
The ABA Clubhouse has 60+ free CEUs — live every Wednesday. Ethics, supervision & clinical topics.
Join Free →Set a 3-minute momentary time sampling schedule for your next staff fidelity check
02At a glance
03Original abstract
The analysis of behavior began with a form of data, rate of responding, which allowed for efficient study and for the description of the basic principles of behavior. Especially important were the facts that rate of responding was a direct reflection of fundamental properties of behavior, and that rate of responding was measured continuously within an experimental session. As behavior analysts moved from purely experimental to applied settings, discontinuous, time-based methods of measurement evolved, which neither directly reflect fundamental properties of behavior nor continuously record behavior within an experimental session. This paper offers a critical discussion of current measurement practices, and discusses factors possibly related to the use of discontinuous, time-based observing/recording procedures. A theoretical basis for observing/recording procedures is presented which emphasizes continuous measurement of response dimensions directly related to fundamental properties of behavior.
The Behavior analyst, 1981 · doi:10.1007/BF03391849