Imagine a parent tells you: “The tantrums aren’t happening as often anymore, but when they do happen, they last forever.”
Your next question isn’t “How many tantrums occurred?” It’s actually “How long did they last?” That question is the foundation of duration recording, and turns “His tantrums seem shorter.” into data you can graph, analyze, and defend.
But knowing why duration matters is the easy part. The harder calls are which behaviors it actually fits, when frequency or latency would tell you more, and what corrupts things once your team collect duration data in session.
This duration data collection guide covers all three at a glance, then dives deep into duration recording documents and the operational decisions to make before the first data point.
Duration vs. Frequency vs. Latency
Three continuous measurement systems often get confused for one another, and picking the wrong one is a common reason data fails to answer the clinical question.
| Measurement | What it captures | The question it answers |
| Duration Data | Elapsed time from behavior onset to offset | How long did it last? |
| Frequency Data | The number of times specified behavior occurs | How often did it happen? |
| Latency Data | Elapsed time between an antecedent and behavior onset | How long until it started? |
If a Learner is given an instruction, screams two seconds later, and keeps screaming for four minutes, latency is two seconds and duration is four minutes.
Despite their connection, they’re still two separate clinical problems. A short latency points toward the antecedent and the demand. A long duration points toward regulation and recovery.
Total duration tells you how much of the session a behavior occupied. Duration per occurrence tells you whether individual episodes are getting shorter.
When Duration Recording Works Best
Duration recording is the right measurement system when the length of time a behavior occurs is more clinically meaningful than the number of occurrences, and when the behavior has a clear beginning and end that observers can identify consistently.
It suits behaviors that start and stop cleanly but run for varying lengths of time:
- Tantrums
- Crying
- On-task behavior
- Independent work
- Social engagement
- Elopement episodes
- Self-stimulatory behavior
- Sleeping
- Physical exercise
- Waiting appropriately
- Tolerating denied access
- Cooperative play
- Screen time
- Time seated during instruction
Duration is especially valuable when you expect behavior to get shorter or longer over the course of treatment.
Consider a Learner who still tantrums daily.
In week one, the average duration is 38 minutes. By week six, it’s only 7 minutes. The frequency suggests almost no change, but the duration shows remarkable improvement.
The same logic works in reverse for behaviors you want to increase.
Let’s say a Learner’s average independent work engagement moves from 4 minutes at baseline to 21 minutes during behavior intervention. The assignment count may not budge, but that jump in sustained engagement is real clinical progress.
Emotional regulation, persistence, attention, coping skills, and leisure participation are all better represented by time than by count.
An ABA Clinical Example
A seven-year-old Learner struggles to stay engaged during independent classroom work. Teachers report that they leave their seat after a few minutes and rarely finish assignments without adult support.
Rather than counting how many times the Learner becomes distracted, the BCBA measures the duration of sustained on-task behavior. Baseline shows an average of 4 minutes before disengaging.
After introducing visual schedules, scheduled movement breaks, and differential reinforcement for sustained engagement, you collect duration data across four weeks:
- Week 1: 8 minutes
- Week 2: 12 minutes
- Week 3: 17 minutes
- Week 4: 24 minutes
When graphed, the data collection shows a clear upward trend. The Learner still needs occasional prompts, but the increasing duration indicates improved attention, persistence, and independence, supporting continued intervention while gradually raising work expectations.
Why Duration Matters for ABA Clinics
Duration often reveals treatment effects that frequency cannot.
For example, a Learner who reduces their tantrums from 40 minutes to 5 still has one tantrum per day. Frequency data shows no change at all, while duration shows the time that behavior occupies, dropped by 87%.
This is a substantial reduction in its impact on that child’s day, and since duration measures time, it’s often the clearest indicator of quality of life, participation, and functional improvement.
When Duration Recording Is Not the Right Choice in ABA
Three situations call for a different measure.
Brief, Discrete Responses
Recording the duration of every greeting, request, or hand raise tells you almost nothing, because the behavior itself is so brief. Frequency represents these behaviors far better.
Unclear Onset & Offset
What ends a tantrum: when crying stops, when the Learner is quiet, when they resume instruction, or when they independently regulate? Without an operational definition to answer this, two observers will record very different durations for the same event.
Continuously Overlapping Behavior
When a Learner cries while engaging in aggression, screaming, dropping, and property destruction at once, it’s not clear whether you’re measuring crying, aggression, tantrums, or dysregulation. Settle those definitions before data collection starts, not after.
Common Data Collection Mistakes
Even when duration is the right choice, accuracy also depends on execution.
Here are the errors that show up most often:
- Timer discipline. Starting late or stopping early systematically underestimates duration.
- No operational definition of onset and offset. This is the largest source of disagreement between observers.
- Measuring total session duration instead of individual occurrences.
- Not deciding how to handle brief pauses. If a Learner stops crying for fifteen seconds and starts again, is that one occurrence or two?
- Comparing total duration across sessions of different lengths without accounting for percentage of session time.
- Using duration when frequency would answer the clinical question better.
By clarifying these decisions in writing before the first session, you can hold every clinician on the case to the same definitions. That’s also what lets your behavior analysts spend their time interpreting the data rather than questioning it.
💡 Clinician Tip: Before you commit to duration recording, ask yourself: If the behavior occurred the same number of times tomorrow but lasted twice as long, would I consider it changed? If the answer is yes, then duration is the more meaningful measurement system.
How We Collect Duration Data
Portia ABA Clinic software records duration with simple start-and-stop timers in session, capturing both total duration and duration per occurrence for every behavior without manual calculation.
A Learner’s total tantrum time can fall simply because fewer tantrums happened, which says nothing about whether any single tantrum became easier to regulate.
For example, consider a Learner’s tantrum data across six weeks:
| Week | Frequency | Total Duration | Duration Per Occurance |
| Baseline | 5 | 100 minutes | 20 minutes |
| Week 6 | 5 | 40 minutes | 8 minutes |
Frequency alone suggests no improvement at all, while duration per occurrence shows each tantrum is now less than half as long. This is a meaningful gain in emotional regulation that frequency recording would have hidden completely.
Portia graphs duration data automatically in reports, so you can compare intervention phases, track progress toward treatment goals, and pull objective duration metrics into supervision notes, progress reports, and funding documentation.
Start Your Free Trial Today
Get a better experience for your customers, and fewer headaches for your team. You’ll be set up in minutes.
Start Your Free Trial Today
Better experience for your customers, fewer headaches for your team. You'll be set up in minutes.