Study for the RBT credential by practicing decisions, not just reciting definitions. For every core concept, learn one definition, one decision rule for when it applies, and one scenario where the wrong choice changes the data. Close each study block by writing an objective session note and listing what you would report to your supervisor.
Choosing Between Continuous and Discontinuous Measurement
Continuous measurement records every instance of a behavior; discontinuous measurement estimates it from samples. Choose continuous methods when counting is feasible, and discontinuous methods when you must divide attention across learners, programs, or long sessions.
Continuous procedures include frequency or rate counts, duration recording, latency, and inter-response time. They capture everything, so they fit low-frequency behaviors or behaviors whose length matters, such as tantrum duration or time to begin a task. Discontinuous procedures sample instead: partial interval marks the behavior if it occurs at any point in an interval, whole interval requires it across the entire interval, and momentary time sampling records only whether it is happening at the instant each interval ends.
Consider a worked scenario. During a two-hour session you must run several teaching programs, so counting every instance of hand-flapping while delivering prompts is not feasible. A plausible mistake is to attempt a frequency count and quietly miss instances during prompts, producing data that looks like a real decline. The better decision is partial or momentary time sampling, selected in advance with your supervisor, plus a note stating the method and its known limits. Partial interval tends to overestimate occurrence, and momentary estimates depend on sampling density, so the label on the data sheet changes how the numbers should be read.
- Frequency/rate: discrete, countable responses; best when an observer can watch continuously
- Duration: how long each episode lasts; best when episode length drives decisions
- Latency: time from instruction to response onset; useful for compliance and initiation programs
- Partial interval: overestimates occurrence; practical when behavior is brief or sporadic
- Whole interval: underestimates occurrence; often used for continuous behaviors like on-task sitting
- Momentary time sampling: sparsest method; reasonable for long sessions with multiple demands
Preference Assessments and Function-Based Assessment Do Different Jobs
Preference assessments identify items or activities that may work as reinforcers. Function-based assessment examines why a behavior occurs. An RBT helps gather both kinds of data but does not interpret function or select interventions independently.
Common preference procedures include free operant observation, paired-choice (two-item) presentations, and multiple-stimulus formats with or without replacement. These produce a ranked hierarchy of potential reinforcers, which is a statement about what the learner selected under specific conditions, not a guarantee that an item will reinforce a particular behavior. A functional behavior assessment, by contrast, uses records review, interviews, descriptive ABC data, and sometimes experimental analysis to evaluate patterns in antecedents and consequences.
A frequent confusion shows up like this: an RBT runs a paired-choice assessment, sees a toy chosen on nearly every trial, and starts offering it after every correct response. When the target behavior does not improve, the temptation is to conclude the learner is unmotivated. The better decision is to treat the hierarchy as a hypothesis to verify with reinforcer probes or mand trials, and to keep collecting ABC data on the challenging behavior for the supervisor's analysis. Preference tells you what a learner gravitates toward; function tells you what a behavior accomplishes. Mixing them leads to reinforcers that fail and intervention plans built on a guess.
Prompting and Fading: Picking the Level and Planning the Transfer
Prompting hierarchies range from full physical assistance to independent responding. Least-to-most prompting minimizes intrusiveness; most-to-least and errorless approaches reduce early errors. The fading plan, not the starting level alone, prevents prompt dependence.
A typical least-to-most sequence moves from independent opportunity through verbal, gestural, model, partial physical, and full physical prompts. It suits learners with some existing skill who tolerate errors. Most-to-least starts with maximum assistance and thins it across sessions, which fits brand-new skills where repeated errors could build a chain of mistakes. Response prompts also sit alongside stimulus prompts such as color coding or positional cues, which are faded by removing or exaggerating the cue rather than by touching the learner less.
A scenario worth tracing: an RBT teaches handwashing using a verbal prompt at every step, and after several weeks the learner waits silently for each instruction. The plausible mistake is to keep prompting, because sessions look smooth and the learner rarely errs. The better decision is a transfer trial: deliver the prompt, then after a brief pause re-present the step and allow independent responding, reinforcing the unprompted response more strongly. If independence stalls, report the pattern to the supervisor so the plan can change. Prompt dependence is a data problem visible on the data sheet, not a flaw to hide with more assistance.
DRA, DRI, DRO, and DRL: One Goal, Four Different Procedures
These differential reinforcement procedures differ in what earns reinforcement: a specific alternative response (DRA), a response incompatible with the target behavior (DRI), the absence of the behavior during an interval (DRO), or a reduced rate (DRL).
The differences are easier to hold onto with a scenario. A learner calls out roughly every few minutes during group instruction, and the program goal is to have the learner request attention by raising a hand. A plausible mistake is implementing a DRO: reinforcing any interval without calling out. That can reduce calling out, but it does nothing to strengthen the replacement skill the plan targets, and the data sheet will show no progress on the actual goal. The better decision is a DRA: reinforce hand-raising with immediate attention and withhold reinforcement for call-outs, per the behavior plan.
Why the distinction matters operationally: each procedure dictates different data collection and different expectations. DRA data track both the target behavior and the alternative response, because an alternative that is never emitted cannot contact reinforcement. DRO data track interval-by-interval absence, so the timer and the definition of an interval matter. DRI requires you to know which appropriate response physically cannot occur at the same time as the target. DRL targets rate reduction for behaviors that are acceptable in smaller doses. Choosing the wrong label means collecting the wrong data, which means the supervisor's weekly review starts from a distorted picture.
| Procedure | What gets reinforced | Target behavior's fate | Data emphasis |
|---|---|---|---|
| DRA | A specific appropriate alternative response | Withholds reinforcement (typically with extinction) | Both the alternative and the target response |
| DRI | A response physically incompatible with the target | Withholds reinforcement (typically with extinction) | Frequency of the incompatible response |
| DRO | Any interval with zero occurrences of the target | Withholds reinforcement | Interval-by-interval absence |
| DRL | Responses occurring at or below a set rate | Reinforced at a reduced rate | Response rate against the criterion |
Session Notes and Data Reports a Supervisor Can Actually Use
Effective session notes are objective, measurable, and free of interpretation. They record what happened, which procedures ran, and any changes from the plan. Anything affecting safety or data validity is reported to the supervisor promptly, not saved for the next meeting.
Compare two sentences about the same session. Interpretive version: the learner had a rough day and was defiant. Objective version: the learner engaged in screaming at an average volume on six occasions during the ten-minute table-work block, completed four of nine presented targets, and required gestural prompts on five trials. The second version lets a supervisor compare today against the baseline, spot a trend, and adjust the plan. Objective notes avoid motive words, use the operational definitions from the plan, and quantify wherever the measurement system allows.
Escalation has its own logic. A new topography of challenging behavior, any injury, a sudden spike outside typical variability, refusal of a previously accepted reinforcer, or an illness in the home all change how the data should be read, so they belong in a timely report to the supervisor rather than only in the daily note. A quick self-check before submitting: could a stranger reconstruct the session from your note alone, and did you flag every deviation from the written plan? If either answer is no, revise before you file it.
Scope of the Technician Role: Implement, Record, Communicate
An RBT implements plans written by the supervising behavior analyst, collects data as designed, and communicates observations. Selecting targets, altering procedures, designing interventions, and interpreting function fall outside the technician role.
A boundary scenario: a parent says hand-raising is not working and asks you to switch to a token system next session. The plausible mistake is agreeing, because the request seems practical and you know token systems from training. The better decision is to thank the parent, share the observation, and explain that procedure changes go through the supervisor, who can assess the data and adjust the plan. Then report the parent's concern to the supervisor the same day. You are the observer closest to the data, which makes your communication valuable precisely because you do not act on it unilaterally.
Professional standards also cover boundaries and confidentiality. Avoid multiple relationships with client families, decline gifts or social arrangements that could compromise objectivity, and protect session information in conversations, on social media, and in shared spaces. The BACB publishes an ethics code specific to RBTs and requires ongoing supervision for certified technicians; treat those documents as part of the syllabus rather than paperwork to skim once. When you are uncertain whether a request is in scope, the default is to implement the written plan, document what occurred, and ask your supervisor.
A Preparation Sequence and a Readiness Rubric You Can Score
Build preparation in passes: definitions first, then decision rules, then scenarios, then note-writing under time pressure. Score yourself against observable milestones rather than a feeling of familiarity, and re-test weak decision points instead of rereading everything.
A realistic adaptable sequence: week one, define every measurement procedure and reinforcement procedure and write one decision rule for each, such as use whole interval when the behavior must persist across an interval. Week two, drill prompting hierarchies and preference assessment formats until you can name the order and the rationale without notes. Week three, work case scenarios: read a situation, state which procedure applies and why, then check against your definitions. Week four, practice objective note-writing and boundary responses, then cycle back to any decision rule you missed.
Now add a hands-on exercise. With permission, observe a safe, low-stakes behavior for twenty minutes, such as a household member's phone-checking. Take a duration record, then score the same period using partial interval and momentary time sampling. Expected observation: the partial-interval percentage exceeds the duration percentage, and the momentary estimate wobbles with sampling density. If your numbers match that pattern, you have internalized the methods rather than the labels. Self-check rubric, scored one to three on each line: I can define each procedure from memory; I can state a decision rule for each choice point; I can complete an objective note in five minutes; I can identify in-scope versus out-of-scope requests. These scores are learning milestones for you, not predictions of any exam outcome or passing standard.
References and further reading
Use these references to explore the concepts and check the latest information from the relevant organizations.