Duration and Latency Recording in ABA

Some behaviors matter because of how long they last. Others matter because of how long it takes to get started.

Two timing methods that answer two very different clinical questions. When you’re running a session, a raw count doesn’t always tell the real story. A tantrum that lasts 30 seconds and one that lasts 20 minutes can both show up as a single tally mark. Duration recording and latency recording capture what a simple count misses. They measure time. This guide explains what each method measures, when to reach for one over the other, and how to collect both cleanly while you’re with a client.

Key takeaways

• Duration recording measures how long a behavior lasts, from the moment it starts to the moment it stops. Use it when the length of a behavior matters more than how often it happens.
• Latency recording measures the time between an instruction or cue and the moment a behavior begins. Use it to track how quickly and independently a learner responds.
• Both are continuous measurement methods, so each needs a clear operational definition with a defined start and stop before you begin collecting data.

What is duration recording in ABA?

Duration recording in ABA measures how long a behavior lasts, from onset to offset. You start timing when the behavior begins, stop when it ends, and record the elapsed time.

It’s the right choice when the length of a behavior is what matters clinically. Think about goals like building independent work time, or reducing how long a meltdown lasts. A frequency count would miss the point, because it treats a brief episode and a long one the same way.

How duration recording works

You watch for the behavior, start a timer at onset, and stop it at offset. If the behavior happens again in the same session, you time that episode too. Each entry is one clean end-to-end measure.

Total duration versus duration per occurrence

Total duration is the combined time a behavior occurred across the whole observation period. Duration per occurrence is the length of each separate episode. Total duration gives you a session-level summary. Duration per occurrence shows patterns inside the session, like whether episodes are getting shorter even when they happen just as often.

What is latency recording in ABA?

Latency recording in ABA measures the time between a specific instruction, cue, or event and the moment the behavior begins. If you say “time to clean up” and the learner starts cleaning up eight seconds later, the latency is eight seconds.

Latency answers a different question than duration. It’s about the delay before a response, not the length of the behavior itself. That makes it useful for goals around task initiation, following instructions, and responding without heavy prompting.

How latency recording works

Define the cue and the response clearly first. Deliver the instruction, start your timer the instant you finish giving it, and stop the timer when the response begins. Record the elapsed time for that opportunity.

What latency tells you

A latency that shrinks over time can show a learner is responding more quickly and more independently. A long or growing latency can point to prompt dependency, low motivation, or a skill that isn’t fluent yet. That’s why latency is often paired with prompt-fading goals.

Duration versus latency at a glance

Both methods put time on the record, but they answer different clinical questions. The table that follows shows the difference.

How to collect duration and latency data

Clear definitions and consistent timing are what let two observers record the same behavior and get the same numbers, which is the foundation of accurate duration and latency data.

Follow these steps:

Duration

1. Write a clear operational definition, including exactly what counts as the start and the end of the behavior.
2. Start your timer the moment the behavior begins.
3. Stop it the moment the behavior ends. If it happens again, start a new entry.
4. Log the start, the stop, and the elapsed time for each episode.
5. Summarize as total duration or average duration per occurrence, based on the goal.

Latency

1. Define the cue and the response, so every observer starts and stops at the same point.
2. Deliver the instruction, then start the timer as you finish giving it.
3. Stop the timer when the response begins, not when it finishes.
4. Record the elapsed time for each opportunity.
5. Watch the trend across sessions to see whether response time is improving.

Common mistakes to avoid

• Fuzzy start and stop definitions, which leave two observers disagreeing about the same event.
• Starting the latency timer while you’re still giving the instruction instead of at the end of it.
• Using latency for a behavior that doesn’t follow a clear cue.
• Letting timing slip because you’re juggling a stopwatch and a paper data sheet mid-session.

When to use duration, latency, or another method

Reach for duration when the length of a behavior is the clinically meaningful part. Reach for latency when the delay before a response is the target. When what matters is how often a behavior happens, frequency and rate recording is the better fit. For very brief, high-rate behaviors, or behaviors without a clean start and end, an interval recording and time sampling method often gives a more workable estimate.

Duration and latency are both continuous vs discontinuous measurement methods, which capture every instance rather than sampling. Continuous recording is also the approach behavior-analytic research leans on most. Continuous recording was used in 55% of reviewed JABA studies of free-operant behavior. Whichever method you pick, the BACB Ethics Code expects a measurement system appropriate to the behavior that produces accurate and reliable data. For the full picture of how the methods fit together, read our ABA data collection guide.

How Office Puzzle makes duration and latency recording easier

Timing is where duration and latency data slip. A handheld stopwatch competes for the attention you owe the client. Built-in timers in an ABA data collection app let you capture onset and offset with a tap, calculate total and average duration for you, and keep latency records without a separate stopwatch. Collecting on a phone or tablet also means you record in the moment across settings, instead of reconstructing times from memory after the session.

Office Puzzle brings duration, latency, and the rest of your session data into one place, so the numbers you collect flow straight into graphs and notes.

If you want something to keep at your desk or on your clipboard, grab our duration and latency data sheet (downloadable).

Frequently asked questions

What is the difference between duration and latency recording?

Duration recording measures how long a behavior lasts, from start to finish. Latency recording measures how long it takes a behavior to begin after a cue or instruction. Duration is about the length of the behavior. Latency is about the delay before it starts.

Is latency recording continuous or discontinuous measurement?

Latency recording is a continuous measurement method, and so is duration recording. Both capture the actual timing of a behavior rather than sampling it at set intervals.

When should you use duration instead of frequency?

Use duration when the length of a behavior is more meaningful than how often it happens. A count treats a 30-second episode and a 20-minute episode the same. Duration shows the difference.

What tools do you need for latency and duration recording?

The essentials are a reliable timer and a clear operational definition. A stopwatch and a data sheet can work, though a digital data collection app with built-in timers reduces missed starts and stops and calculates totals for you.

Sources

1. Behavior Analyst Certification Board. (2020). Ethics code for behavior analysts. https://www.bacb.com/wp-content/uploads/2022/01/Ethics-Code-for-Behavior-Analysts-240830-a.pdf
2. Mudford, O. C., Martin, N. T., Hui, J. K. Y., & Taylor, S. A. (2009). Assessing observer accuracy in continuous recording of rate and duration: Three algorithms compared. Journal of Applied Behavior Analysis, 42(3), 527–539. https://pmc.ncbi.nlm.nih.gov/articles/PMC2741073
3. Mudford, O. C., Taylor, S. A., & Martin, N. T. (2009). Continuous recording and interobserver agreement algorithms reported in the Journal of Applied Behavior Analysis (1995–2005). Journal of Applied Behavior Analysis, 42(1), 165–169. https://pmc.ncbi.nlm.nih.gov/articles/PMC2649837/