Topic Quiz — Data Collection
RBT Data Collection & Graphing Quiz
Content reviewed by Dr. Sarah Mitchell, BCBA-D
This quiz drills measurement systems, continuous and discontinuous data collection, interobserver agreement, and how to read behavior graphs — skills you use in every session.
A data path that shows points steadily decreasing session after session is best described as:
Why data collection is a daily skill
Applied behavior analysis is a data-driven science, and as an RBT you are the person who produces the data. Every clinical decision your supervising BCBA makes, whether to advance a target, change a prompt level, or adjust a behavior plan, rests on the numbers you record session after session. On the RBT Test Content Outline (3rd Edition), the measurement domain tests whether you can choose the right recording method for a given behavior, collect it accurately, and read the graph that results. This quiz drills all three.
The exam leans heavily on scenarios: it describes a behavior and a session, then asks which measurement fits, or gives you a small set of numbers and asks for a rate or an agreement score. Because these are the same judgments you make in real sessions, getting comfortable here pays off twice. Every question includes a plain-English explanation reviewed by a BCBA-D so the reasoning sticks.
Continuous versus discontinuous measurement
The first big distinction is whether you capture every instance of a behavior or only sample it. Continuous measurement records every occurrence, giving the most complete and accurate picture. Discontinuous measurement samples behavior within intervals, which is easier to run when you are teaching at the same time but sacrifices some accuracy. Knowing which category a method belongs to, and what it trades off, is a frequent exam target.
Continuous measures: frequency, rate, duration, latency, IRT
Continuous methods measure a dimension of the behavior itself. The ones you must know cold are:
- Frequency (count). A simple tally of how many times the behavior occurred. Frequency is meaningful only when the observation time is constant across sessions.
- Rate. Count divided by time, expressed as responses per minute or per hour. Rate lets you compare sessions of different lengths, which is why it is often preferred over raw count. If a learner hit 12 times in a 30-minute session, the rate is 0.4 per minute.
- Duration. How long a behavior lasts, from onset to offset. Best for behaviors defined by their length, such as tantrums or on-task engagement.
- Latency. The time from a cue or instruction to the start of the behavior. Useful when the concern is how long it takes a learner to begin responding, such as complying with a direction.
- Interresponse time (IRT). The time between the end of one response and the start of the next. Shorter IRTs mean a higher rate; it is often paired with DRL-style goals.
Discontinuous measures: interval recording and time sampling
Discontinuous methods divide the observation into intervals and record whether behavior happened within them. Three are commonly tested, and each biases the estimate in a predictable direction:
- Partial-interval recording. You mark an interval if the behavior occurred at any point during it. This tends to overestimate duration but works well for capturing whether low-rate or brief problem behavior happened.
- Whole-interval recording. You mark an interval only if the behavior occurred for its entire length. This tends to underestimate the behavior and fits skills you want sustained, like staying on task.
- Momentary time sampling. You record whether the behavior is occurring at the specific moment an interval ends. It lets you teach between observations, which makes it practical for busy sessions and group settings.
Interobserver agreement (IOA)
Data are only useful if they are believable, and interobserver agreement is how the team checks that. Two observers independently record the same behavior at the same time, and their records are compared to produce an agreement percentage. A common method is total-count IOA: divide the smaller count by the larger count and multiply by 100. If one observer counted 8 and the other counted 10, IOA is 8 divided by 10, or 80%. High agreement supports the accuracy of the data; consistently low agreement signals that the behavior definition may be unclear or that observers need retraining. As an RBT you often serve as the second observer during these checks.
Reading line graphs: level, trend, and variability
Behavior analysts display data on line graphs, most often equal-interval line graphs, so patterns are visible at a glance. The x-axis usually shows time or sessions and the y-axis shows the behavior's measure. Three features describe what a graph is telling you:
- Level is the overall value of the data on the y-axis, roughly how high or low the behavior is. A change in level between phases suggests the intervention shifted the behavior.
- Trend is the direction of the data path over time: increasing, decreasing, or flat. For a skill you are building, an upward trend is good; for a problem behavior, a downward trend is the goal.
- Variability is how much the data points bounce around. Low variability means stable, predictable responding; high variability makes it harder to judge whether an intervention is working and often prompts a closer look at conditions.
Vertical phase-change lines separate conditions (for example baseline from intervention) so you can compare level, trend, and variability across phases. As an RBT you read these graphs to understand where a program stands, but decisions about phase changes belong to your supervisor.
Where this fits in the full exam
Measurement is one of six domains, and it underpins every other one. Use the table below to see how it relates to the rest, then drill any weak area with a focused quiz.
| RBT Task List Domain | What it covers | Study |
|---|---|---|
| Data Collection and Graphing | Measurement systems, continuous and discontinuous data, interobserver agreement, and reading behavior graphs. | Guide |
| Behavior Assessment | Preference assessments, the RBT's role in functional assessment, and supporting the BCBA during assessment. | Guide |
| Behavior Acquisition | Discrete-trial teaching, naturalistic teaching, prompting and fading, shaping, chaining, and skill acquisition plans. | Guide |
| Behavior Reduction | Functions of behavior, differential reinforcement, extinction, antecedent strategies, and crisis procedures. | Guide |
| Documentation and Reporting | Objective session notes, incident reports, communication with the supervisor, and maintaining records. | Guide |
| Ethics and Professional Conduct | The RBT Ethics Code, professional boundaries, scope of practice, confidentiality, and supervision requirements. | Guide |
Key concepts to lock in before test day
- Continuous captures every instance; discontinuous samples. Know which bucket each method falls in.
- Rate beats raw count when session lengths differ, because it standardizes to responses per unit of time.
- Remember the interval biases. Partial-interval overestimates, whole-interval underestimates, momentary time sampling snapshots the moment.
- Total-count IOA is smaller divided by larger, times 100.
- Describe graphs with level, trend, and variability rather than vague impressions.
- Record, do not interpret. You collect accurate data; the BCBA makes phase-change and treatment decisions.
Where to go next
Measurement is the engine behind every other domain, so put it to work: see how data drives teaching in the behavior acquisition quiz and how it guides intervention in the behavior reduction quiz. Because clean data starts with good notes, pair it with the documentation and reporting quiz. Review core terms in our free RBT study guide, then test everything on a free sample RBT exam.