Written to BACB's RBT Test Content Outline (3rd ed.)
- Data Collection and Graphing 17.3%
- Behavior Assessment 10.7%
- Behavior Acquisition 25.3%
- Behavior Reduction 18.7%
- Documentation and Reporting 13.3%
- Ethics 14.7%
Free · 2026 · 17.3% of the RBT RBT Data Collection and Graphing practice questions
Data Collection and Graphing is 17.3% of the RBT. Answer 10 questions (6 on Data Collection and Graphing) and see your score by domain, with every answer explained. Free, no sign-up.
An RBT uses 5-minute momentary time sampling to record on-task behavior. During one interval, the client is off-task for about 4 minutes. When the timer signals the end of the interval, the client is writing on his worksheet. How should the RBT record this interval?
Tap an answer, or press A to D
Continuous and discontinuous measurement, permanent products, operational definitions, calculating and graphing data.
What the RBT exam tests in Data Collection and Graphing
- Use continuous measurement: frequency, duration, latency, interresponse time
- Use discontinuous measurement: partial-interval, whole-interval, momentary time sampling
- Use permanent-product recording
- Enter data and update graphs
- Describe behavior and environment in observable, measurable terms (operational definitions)
- Calculate and summarize data several ways: rate, mean duration, percentage
- Identify trends in graphed data
- Explain the risks of unreliable data collection and poor procedural (treatment) fidelity
High-yield data collection and graphing facts
- Continuous measurement records every instance (frequency, rate, duration, latency, IRT). Discontinuous measurement samples behavior and only estimates it (partial, whole, MTS).
- Latency = from the SD or cue to the START of the response. IRT = from the END of one response to the START of the next.
- Duration: record total time or time per episode. Mean duration = total time / number of episodes.
- Permanent product = measure the lasting result afterward (worksheet, assembled items). You did not see who produced it.
- Operational definition: observable and measurable, so two observers score it the same. Include what it looks like, examples and non-examples, and when an episode starts and stops.
- Dead-man test: if a motionless person could do it ('sits quietly', 'doesn't hit'), it is not a behavior definition.
| Measure | Record | Best for |
|---|---|---|
| Frequency/count | Every occurrence | Discrete behaviors with a clear start and end |
| Rate | Count / time | Comparing sessions of different lengths |
| Duration | How long each episode lasts | Tantrums, on-task, play |
| Latency | SD to start of response | Compliance, slow starters |
| IRT | End of one response to start of the next | Pacing (e.g., bites while eating) |
| Permanent product | Outcome measured afterward | Worksheets, tasks completed |
The full RBT High-Yield Cram Sheet comes with full prep.
Common data collection and graphing traps
- Latency starts at the instruction. IRT starts at the end of the previous response.
- Partial interval overestimates and whole interval underestimates. Candidates often swap them.
- MTS checks only the instant the interval ends, not the whole interval.
- When session lengths differ, compare rate, not raw count.
- Do not connect data points across a phase-change line.
- A decreasing trend is the goal for problem behavior, so 'decreasing' does not automatically mean 'worse'.
- An inference such as 'was anxious' is not an operational definition.
Free RBT data collection and graphing practice questions
Question 1
A client has about two tantrums per day. Some tantrums last 5 minutes and others last 40 minutes. The team wants to know whether the intervention is making the tantrums shorter. Which measurement should the RBT use?
- AInterresponse time recording
- BLatency recording
- CDuration recording
- DFrequency recording
Show answer and explanation
Answer: C. Duration recording captures how long each tantrum lasts, which directly answers whether the tantrums are getting shorter.
Why not A: IRT measures the time between tantrums, not the length of each tantrum.
Why not B: Latency measures how long it takes a response to start after a stimulus, not how long the response lasts.
Why not D: A count would show about two tantrums per day whether they lasted 5 minutes or 40, so it would miss the change in length.
Remember: Use duration recording when the question is how long a behavior lasts rather than how often it happens.
Question 2
A supervisor reviews four draft target definitions and asks the RBT to flag the one that fails the test "Could a motionless person do it?" Which definition should be flagged?
- ASays "help please" when a task is too difficult
- BWrites their name at the top of each worksheet
- CRemains quiet and still during circle time
- DRaises a hand and waits until called on before speaking
Show answer and explanation
Answer: C. A motionless person could "do" this, because it describes the absence of behavior rather than an action to teach and measure.
Why not A: A vocal request is an active, observable response that can be counted.
Why not B: Writing a name is an active behavior, and it even leaves a permanent product.
Why not D: Raising a hand is an observable action that a motionless person could not perform.
Remember: If a motionless person could meet a definition, it describes the absence of behavior and should be rewritten as an active response.
Question 3
During a group activity, an RBT's timer vibrates every 5 minutes. Each time it vibrates, the RBT looks at the client and records only whether the client is engaged with the materials at that instant. Which method is this?
- AWhole-interval recording
- BDuration recording
- CPartial-interval recording
- DMomentary time sampling
Show answer and explanation
Answer: D. The RBT records whether the behavior is occurring only at the moment each interval ends.
Why not A: Whole-interval recording requires watching the whole interval and scoring it only if the behavior lasts the entire time.
Why not B: Duration recording times how long the behavior lasts, which requires watching continuously rather than checking at moments.
Why not C: Partial-interval recording requires watching the whole interval and scoring it if the behavior happens at any point.
Remember: In momentary time sampling, the RBT records only whether the behavior is happening at the exact moment the interval ends.
Question 4
While updating a graph, an RBT notices that two days ago she entered 18 responses for a session, but the original paper data sheet shows 8. What should she do?
- ADelete that session's data point from the graph entirely
- BChange the value and throw away the original data sheet to avoid confusion
- CCorrect the entry following the agency's error-correction procedure and inform the supervisor
- DLeave the value as it is, since one data point will not change the overall trend
Show answer and explanation
Answer: C. The graph must match the source data, and documented, transparent corrections keep the record trustworthy for the supervisor's decisions.
Why not A: Removing a session with valid data creates a new inaccuracy. The correct value of 8 should be shown instead.
Why not B: Original records should be kept. Discarding source documents and making undocumented changes undermines data integrity.
Why not D: A single wrong point can distort the visual analysis and lead to poor clinical decisions. Known errors must be fixed.
Remember: When you find a data-entry error, correct it transparently according to policy and tell your supervisor, so the graph always matches the original data.
Question 5
A supervisor asks an RBT to use partial-interval recording to track a client's brief, frequent vocal outbursts, which the team hopes to decrease. Which statement about this method is accurate?
- AIt tends to overestimate the behavior, which is why it is often used for behaviors targeted for decrease.
- BIt records only whether an outburst is happening at the end of each interval.
- CIt tends to underestimate the behavior, so it is best for behaviors targeted for increase.
- DIt gives an exact count of outbursts, so it is equivalent to frequency recording.
Show answer and explanation
Answer: A. Because one brief instance scores the whole interval, the data inflate the behavior. Any progress shown is therefore conservative for a reduction goal.
Why not B: That describes momentary time sampling, not partial-interval recording.
Why not C: That describes whole-interval recording. Partial-interval recording scores an entire interval even for a brief occurrence.
Why not D: Several outbursts in one interval are scored only once, so partial-interval recording cannot give an exact count.
Remember: Partial-interval recording tends to overestimate behavior and is commonly used for behaviors targeted for reduction.
Full RBT prep
Practice every RBT domain
7-Day Cram Access
Lowest upfront$157 days
$2.14 a day
One-time payment. We will never charge you again.
- 300+ RBT practice questions, every answer choice explained
- Unlimited timed mock exams (85 questions)
- Interval & IRT Data Lab: all 15 rounds
- Full 8-page printable High-Yield Cram Sheet (PDF)
100% Pass Guarantee + 7-day money-back. How it works
RBT Exam Pass
Best value$39$29120 days
$0.24 a day
One-time payment. Nothing renews.
- 300+ RBT practice questions, every answer choice explained
- Unlimited timed mock exams (85 questions)
- Interval & IRT Data Lab: all 15 rounds
- Full 8-page printable High-Yield Cram Sheet (PDF)
100% Pass Guarantee + 7-day money-back. How it works