Counting a behavior sounds like the simplest thing in the world. Tally each time it happens, add it up, done.
Then the session runs long one day and short the next, and the count stops telling you the truth.
That’s the moment frequency recording needs rate.
Frequency recording is the most common way ABA teams measure how often a behavior occurs, and for good reason. It’s fast, it’s intuitive, and it fits the discrete behaviors that fill a lot of programs. It also has a blind spot. A raw count says nothing about the time it took to collect. This guide covers what frequency recording measures, how it relates to count and rate, how to calculate rate, and when each is the right call.
Frequency recording in ABA measures how many times a behavior occurs during a set observation period. You tally each occurrence as it happens, then total them at the end. If a learner requests a break five times in a session, the frequency is five.
It works best for discrete behaviors, meaning behaviors with a clear beginning and end that are quick to count. Hand raises, prompted responses, aggressive acts, and vocal requests are all good candidates. Behaviors that last a long time or blur together are a poor fit, and the sections that follow cover better options for those.
Before anyone collects data, the behavior needs an operational definition precise enough that two people would count the same event the same way. A useful check is the new-staff test. If a new RBT reads the definition, would they know exactly what to tally? A definition like “interruptions or speaking out of turn, calling out answers, or talking over someone” passes. “Disruption” on its own does not.
Here is where terms get slippery. In everyday use, people say “frequency” to mean both a plain count and a count per unit of time. The distinction matters, and the field has weighed in. A 2018 review in the Journal of Applied Behavior Analysis found the predominant use of frequency is as a count, not a rate, and recommended keeping frequency as a synonym for count.
Using those definitions, frequency (or count) is the raw number of occurrences. Rate is that count divided by the time observed, expressed as occurrences per minute or per hour. Ten requests in a session is a count. Ten requests in 30 minutes is a rate.
|
Dimension |
Frequency (count) |
Rate |
|---|---|---|
|
What it measures |
How many times a behavior occurs |
How often a behavior occurs per unit of time |
|
Includes time? |
No, it’s a raw count |
Yes, the count divided by observation time |
|
Best when |
Observation periods stay a consistent length |
Session lengths vary, or you compare across sessions |
|
Example |
10 requests in a session |
10 requests in 30 minutes, or 0.33 per minute |
|
Watch out for |
Counts aren’t comparable if session length changes |
Keep time units consistent when you calculate |
Rate is simple math. Divide the number of occurrences by the length of the observation, and keep your time units consistent.
Ten hand raises in a 30-minute session is 10 divided by 30, or 0.33 per minute.
Ten hand raises in a 60-minute session is 10 divided by 60, or 0.17 per minute.
Same count, very different picture. That is why rate matters when your sessions are not a fixed length.
Reach for frequency when the behavior is discrete, brief, and easy to count, and your observation period is a consistent length. Reach for rate when session lengths vary, or when you want to compare how dense a behavior is across sessions of different durations. When the behavior lasts a long time or its length is the point, duration and latency recording is the better measure. When a behavior happens so fast or so continuously that counting every instance is not realistic, an interval recording and time sampling method gives a workable estimate.
Frequency, rate, duration, and latency are all [internal link: continuous vs discontinuous measurement] methods, which record every instance rather than sampling. Continuous recording is also the approach behavior-analytic research relies on most, used in 55% of reviewed JABA studies of free-operant behavior. For broader context, read our ABA data collection guide.
A count is only as good as the definition behind it. These practices keep frequency and rate data reliable across a team.
If you are the one tallying during a session, the mechanics matter as much as the definition. Keep the count somewhere you can update instantly, tally each occurrence the moment it happens, and do not try to reconstruct it later. The hands-on procedure for tallying discrete behaviors has its own guide.
The weak points in count data are consistency and time-tracking, and that is where the right tool helps. A one-tap counter in an ABA data collection app records each occurrence without breaking your attention, timestamps it, and pulls the session length automatically. This allows a count to convert to a rate with no math. Shared operational definitions keep every RBT counting the same event the same way, and built-in agreement checks make calibration routine instead of a special project. Office Puzzle brings frequency, rate, and the rest of your session data into one place, so the numbers flow straight into graphs and reports.
If a desk reference would help your team, grab our frequency and rate data sheet (downloadable).
Frequency is a count of how many times a behavior occurs. Rate is that count divided by the time observed, such as occurrences per minute. Frequency answers how many. Rate answers how many per unit of time, which lets you compare across sessions of different lengths.
In current ABA usage, yes. A 2018 Journal of Applied Behavior Analysis review found frequency is most often used to mean count, and recommended keeping it that way. Rate is the term for a count standardized by time.
Divide the number of occurrences by the length of the observation, keeping time units consistent. Eight occurrences in 40 minutes is 8 divided by 40, or 0.2 per minute. Multiply by 60 to express it per hour if that reads more clearly.
Use frequency when how often a behavior happens is the clinically meaningful part and the behavior is quick to count. Use duration when the length of each episode is what matters, like time spent in a tantrum or on task.