RBT (Registered Behavior Technician) — All Questions

Maya's program says to record how long it takes her to start brushing her teeth after the instruction 'Time to brush' is delivered. The RBT starts a stopwatch as the instruction ends and stops it the instant Maya picks up the toothbrush. Which measurement procedure is the RBT using?

• a.Latency recording, timing from the instruction to the start of the response✓
• b.Duration recording, timing from the start of the response to its end
• c.Interresponse time, timing from one response to the next response
• d.Frequency recording, counting each toothbrushing response in the session

(a) is correct: latency is the elapsed time between the onset of a stimulus — here the instruction — and the beginning of the response, which is exactly what the stopwatch captures. (b) is plausible because duration is also a timed measure, but duration would begin when Maya starts brushing and end when she stops; the RBT stops the watch at the moment brushing begins, so no duration is recorded. (c) is plausible because interresponse time is timed too, but IRT runs from the end of one response to the start of the next, and only one response is involved here. (d) is wrong because frequency produces a count, not a time value; it would tell the team how often Maya brushed, not how long she took to begin.

Devon's plan asks the RBT to record how much time passes between one hand-raise and the next during a 20-minute lesson. The RBT starts timing when Devon lowers his hand and stops when he raises it again, repeating this for every pair of raises. Which measure is being recorded?

• a.Latency, the time from the teacher's question to Devon's hand-raise
• b.Duration, the total time Devon's hand stays in the air each time
• c.Interresponse time, the time between the end of one raise and the next✓
• d.Rate, the number of hand-raises divided by the total minutes of the lesson

(c) is correct: interresponse time is the interval between the end of one response and the beginning of the next, which is precisely what the RBT times. (a) is plausible because latency is also measured in seconds, but latency is anchored to a stimulus such as the teacher's question, and no stimulus is being timed from here. (b) is plausible because duration is timed as well, but duration would run from the moment the hand goes up to the moment it comes down — the RBT is timing the gap between raises instead. (d) is wrong because rate is a count divided by time and would yield hand-raises per minute, not the spacing between individual responses.

Nora's crying episodes vary from a few seconds to nearly ten minutes, and the team wants to know whether the episodes are getting shorter over time. What should the RBT record?

• a.Frequency, by counting each crying episode that occurs in the session
• b.Duration, by timing each crying episode from its onset to its end✓
• c.Latency, by timing from the demand to the start of each crying episode
• d.Rate, by dividing the number of crying episodes by the session length

(b) is correct: the team's question is about how long episodes last, and duration recording times each episode from onset to offset, so a shrinking duration answers it directly. (a) is plausible because counting is the most common measure, but a count would stay the same whether an episode lasted five seconds or five minutes. (c) is plausible because latency is a timed measure, but it captures how quickly crying begins after a demand, not how long it continues. (d) is wrong for the same reason as (a): rate standardizes a count for session length but still says nothing about episode length.

An RBT counts Ellis's blurt-outs in a 30-minute session on Monday and in a 50-minute session on Tuesday. The BCBA wants the two days compared fairly on a single graph. Which summary should the RBT plot?

• a.Total count of blurt-outs, because the raw number is what was observed
• b.Duration of blurt-outs, because it controls for how long each one lasts
• c.Percentage of intervals with blurt-outs, because percentages compare well
• d.Rate of blurt-outs per minute, because it controls for session length✓

(d) is correct: when observation periods differ in length, a raw count is not comparable, and dividing the count by the minutes observed produces a rate that puts both days on the same footing. (a) is plausible because the counts are what the RBT actually collected, but a larger count in a longer session may reflect more time, not more behavior. (b) is plausible because duration also controls for something, but it measures how long each blurt-out lasts, which is not what was counted here. (c) is plausible because percentages are comparable across sessions, but interval data were never collected; the RBT cannot report a percentage of intervals from a simple tally.

Which description defines interresponse time?

• a.The time from a stimulus to the beginning of the first response
• b.The time from the beginning of a response to the end of that response
• c.The time from the end of one response to the beginning of the next one✓
• d.The number of responses divided by the time the behavior was observed

(c) is correct: interresponse time measures the gap between consecutive responses — from the end of one to the onset of the next — and is used when the team cares about how closely responses are spaced. (a) defines latency, which is anchored to a stimulus rather than to a preceding response. (b) defines duration, which measures the length of a single response rather than the space between two. (d) defines rate, which is a count standardized by observation time; it can be high while individual interresponse times still vary widely.

Kai claps once, briefly, several times an hour, and each clap has a clear beginning and end. Sessions are always exactly 60 minutes. The BCBA wants the simplest count-based measure of how much clapping occurs. What should the RBT record?

• a.Frequency, tallying each clap as it occurs throughout the session✓
• b.Duration, timing how long each individual clap lasts in seconds
• c.Latency, timing how long after arrival the first clap occurs
• d.Partial interval, marking each interval in which a clap occurs

(a) is correct: clapping is a discrete response with a clear onset and offset that occurs at a manageable rate, so a simple tally captures it accurately, and because sessions are always 60 minutes the counts are already comparable. (b) is plausible because duration is a continuous measure, but a clap lasts a fraction of a second, so timing it adds nothing the count does not already say. (c) is plausible because latency is easy to collect, but it describes only the first clap of the session. (d) is plausible because interval recording is common, but it is an estimate and it would lose the exact count that this behavior easily supports.

During a session the RBT records 4 seconds between the instruction 'Touch red' and Priya's first touch, then 3 seconds between the next instruction and her touch. A new RBT reads the sheet and calls these interresponse times. Why is that label wrong?

• a.Interresponse time is measured only when responses occur in bouts
• b.Interresponse time is reported as a rate rather than as raw seconds
• c.Interresponse time is measured from the instruction to a response
• d.Interresponse time is measured between responses, not from a stimulus✓

(d) is correct: the timings begin at an instruction, which makes them latencies; interresponse time begins at the end of one response and ends at the start of the next, with no stimulus involved. (a) is plausible because IRT is often used with bouts of high-rate behavior, but it is not restricted to them, so that is not what makes the label wrong. (b) is wrong because IRT is reported in units of time, exactly as these values are; the units are not the problem. (c) is wrong because it describes neither measure accurately — it mixes the start of latency with the end of duration.

Theo engaged in three separate episodes of pacing, lasting 2, 5, and 8 minutes, during a 60-minute session. The BCBA asked for the total duration of pacing. What should the RBT record?

• a.3 episodes, because that is the number of times pacing occurred
• b.15 minutes, because the three episode times are added together✓
• c.5 minutes, because that is the average length of a pacing episode
• d.25 percent, because pacing filled that share of the whole session

(b) is correct: total duration is the sum of the individual episode times, and 2 + 5 + 8 = 15 minutes. (a) is plausible because three episodes is also a true statement about the session, but a count of episodes is frequency, not duration. (c) is plausible and arithmetically correct as a mean — 15 divided by 3 is 5 minutes per episode — but the BCBA asked for the total, not the average. (d) is also arithmetically correct as a share of the session, since 15 divided by 60 is 0.25, but it is a percentage of session time rather than the total duration requested.

Sam's out-of-seat behavior happens only two or three times a session, but each episode can last most of the period. The team wants a measure showing how much of the session Sam spends out of his seat. Which measure fits best?

• a.Frequency, since counting the episodes shows how often Sam leaves his seat
• b.Duration, since timing the episodes shows how long Sam is out of his seat✓
• c.Latency, since timing the first episode shows when Sam first leaves his seat
• d.Rate, since dividing episodes by session minutes standardizes the count

(b) is correct: the question is about how much of the session is consumed by the behavior, and duration recording accumulates the time of each episode to answer it. (a) is plausible because frequency is the easiest measure to collect, but two episodes could mean four minutes or forty. (c) is plausible because latency is informative when the team cares about how quickly a behavior starts, but it describes only the beginning of the session. (d) is plausible because rate standardizes counts, but it standardizes the wrong dimension here — it still cannot distinguish a brief episode from a long one.

Over three weeks, the average time between the instruction 'Line up' and Ivy walking to the line dropped from 42 seconds to 9 seconds. The RBT is asked what this shows. Which statement is accurate?

• a.Latency to comply has decreased, so Ivy is responding more quickly✓
• b.Duration of compliance has decreased, so Ivy is lining up for less time
• c.Rate of compliance has decreased, so Ivy is lining up less frequently
• d.Interresponse time has decreased, so Ivy is lining up more often per hour

(a) is correct: the value being timed starts at an instruction and ends when the response begins, which is latency, and a smaller latency means a faster response — improvement for a compliance target. (b) is plausible because both are timed measures, but duration would time how long Ivy stands in line, which was never recorded. (c) is plausible because a falling number sounds like a decrease in behavior, but rate is a count per unit time and no counts were reported. (d) is plausible because IRT is also a shrinking time value in this pattern, but IRT is measured between two responses, and here the timing starts at the adult's instruction.

An RBT records 30 vocal protests in a 60-minute session and 20 vocal protests in a 25-minute session. Asked which session had more protesting, the RBT compares the rates. Which conclusion follows?

• a.The first session, because 30 protests is the larger raw count
• b.The first session, because it produced 0.8 protests per minute
• c.The second session, because it produced 0.8 protests per minute✓
• d.The second session, because 20 protests in 25 minutes is 1.25 per minute

(c) is correct: rate is count divided by time, so the first session is 30 ÷ 60 = 0.5 protests per minute and the second is 20 ÷ 25 = 0.8 protests per minute; the second session had the higher rate. (a) is plausible because 30 is the larger number, but the first session was more than twice as long, which is exactly why raw counts mislead. (b) attaches the correct rate figure to the wrong session — 0.8 belongs to the second session, not the first. (d) picks the right session with inverted arithmetic: 25 ÷ 20 = 1.25 gives minutes per protest, not protests per minute.

How do frequency and rate differ as measures of behavior?

• a.Frequency is a timed measure, and rate is a simple count of responses
• b.Frequency counts intervals scored, and rate counts responses per hour
• c.Frequency applies to continuous behavior, and rate applies to discrete
• d.Frequency is a count of responses, and rate is that count per unit time✓

(d) is correct: frequency is the raw number of responses observed, and rate divides that number by the length of the observation, which is what makes sessions of different lengths comparable. (a) reverses the two — neither is a timed measure in the way duration or latency is. (b) is plausible because rate is often expressed per hour, but frequency counts responses, not intervals; counting intervals is a discontinuous procedure. (c) is plausible because measures are indeed matched to behavior types, but both frequency and rate require discrete, countable responses.

An RBT observes a learner for ten consecutive 10-second intervals and marks an interval if vocal stereotypy occurs at any point during it. Four intervals are marked, including one in which stereotypy lasted only about one second. Which measurement procedure was used, and how should the result be reported?

• a.Whole-interval recording; reported as 40% of intervals.
• b.Partial-interval recording; reported as 40% of intervals.✓
• c.Momentary time sampling; reported as 4 occurrences.
• d.Frequency recording; reported as 4 responses.

(b) is correct: scoring an interval when the behavior occurs at any point during it is partial-interval recording — the one-second occurrence still counts — and interval data are reported as the percentage of intervals scored (4/10 = 40%). (a) is wrong because whole-interval recording scores only intervals where the behavior lasted the entire interval, so the one-second interval would not have been marked. (c) is wrong twice over: momentary time sampling looks only at the exact moment each interval ends, and its results are still reported as a percentage of intervals, not a count. (d) is wrong because frequency recording counts discrete responses continuously rather than marking intervals; an interval method estimates occurrence and never yields a response count.

During a 15-minute work period divided into 30-second intervals, an RBT marks an interval only if Lena stays in her chair for the entire interval. Which procedure is this, and what does it tend to do to the estimate?

• a.Partial-interval recording, which tends to overestimate the behavior
• b.Partial-interval recording, which tends to underestimate the behavior
• c.Whole-interval recording, which tends to underestimate the behavior✓
• d.Momentary time sampling, which tends to overestimate the behavior

(c) is correct: requiring the behavior to occupy the entire interval is whole-interval recording, and because any brief break inside an interval loses the whole interval, it systematically underestimates how much the behavior occurred. (a) names the wrong procedure but the right bias for it — partial-interval does overestimate, which is why the pairing is tempting. (b) names partial-interval with the wrong bias attached. (d) is plausible because momentary time sampling is also a discontinuous method, but it scores only the instant an interval ends and is not systematically biased upward the way partial-interval is.

An RBT sets a vibrating timer for every 2 minutes during a 20-minute activity and marks whether Owen is engaged with materials at the instant the timer vibrates. Which procedure is described?

• a.Momentary time sampling, scoring only at the end of each interval✓
• b.Whole-interval recording, scoring engagement across each interval
• c.Partial-interval recording, scoring any engagement in an interval
• d.Duration recording, scoring the total time Owen is engaged

(a) is correct: momentary time sampling scores the behavior only at a specified moment — here the instant the timer signals — and ignores everything that happens between those moments. (b) is plausible because the RBT is using intervals, but whole-interval recording requires watching the full interval and scoring only if engagement lasted throughout. (c) is plausible for the same reason, but partial-interval requires watching the full interval and scoring any occurrence within it. (d) is wrong because no time is being accumulated; the RBT is making a yes-or-no decision at set moments.

A BCBA reviews partial-interval data showing hand-flapping in 80% of intervals, then watches video of the same session and estimates that flapping actually occupied about half the time. What best explains the discrepancy?

• a.Partial-interval data underestimate behavior occurring in short bursts
• b.Partial-interval data convert occurrences into a rate per unit of time
• c.Partial-interval data score only the moment at which an interval ends
• d.Partial-interval data overestimate behavior by scoring brief occurrences✓

(d) is correct: a single brief occurrence scores an entire interval, so a behavior that touches many intervals briefly is credited with far more of the observation than it actually occupied. (a) states the opposite bias; underestimation is the characteristic problem of whole-interval recording. (b) is plausible because both are summary measures, but interval data are reported as a percentage of intervals and are never a rate. (c) describes momentary time sampling rather than partial-interval, though it is an easy swap because all three are discontinuous methods.

A team wants to measure Ruth's on-task behavior with an estimate that will not make her look better than she actually is. The BCBA asks the RBT which discontinuous procedure gives the most conservative picture of a behavior targeted for increase. Which should the RBT name?

• a.Partial-interval recording, because any occurrence marks an interval
• b.Whole-interval recording, because the behavior must fill the interval✓
• c.Momentary time sampling, because only the end moment is scored
• d.Event recording, because every separate on-task response is counted

(b) is correct: whole-interval recording underestimates, so for a behavior the team wants to increase it produces the conservative estimate the BCBA asked for. (a) is plausible because partial-interval is the most commonly used interval method, but it overestimates, which would flatter Ruth's on-task levels. (c) is plausible because momentary sampling is efficient and widely used, but it is not systematically conservative — it can land above or below the true value. (d) is wrong because event recording is a continuous measure, not a discontinuous one, and on-task behavior has no clean discrete response to count.

An RBT scores partial-interval data across forty 15-second intervals and marks 14 of them. How should this be summarized on the data sheet?

• a.35% of intervals scored, because 14 of the 40 intervals were marked✓
• b.14 occurrences, because each marked interval stands for one response
• c.65% of intervals, because 26 of the 40 intervals went unmarked
• d.1.4 per minute, because 14 marks fall across ten total minutes

(a) is correct: interval data are summarized as the percentage of intervals scored, and 14 ÷ 40 = 0.35, or 35%. (b) is plausible because a marked interval feels like an occurrence, but partial-interval recording only shows that the behavior happened at least once in that interval; the true count could be much higher. (c) inverts the calculation by reporting the unmarked intervals — 26 ÷ 40 = 0.65 — which describes the absence of behavior. (d) does the arithmetic correctly, since forty 15-second intervals is ten minutes and 14 ÷ 10 = 1.4, but interval data are an estimate of occurrence and cannot be reported as a rate.

Jonah hits his desk repeatedly, sometimes ten times within a single 10-second interval. The RBT uses partial-interval recording and reports 6 of 30 intervals scored. What is the main limitation of this summary?

• a.It cannot show whether the hitting occurred early or late in the session
• b.It cannot be graphed, because interval data are not plotted on line graphs
• c.It cannot show how many hits occurred, since an interval is scored once✓
• d.It cannot be converted to a percentage without knowing interval length

(c) is correct: partial-interval recording marks an interval the same way whether the behavior happened once or ten times, so a high-rate behavior like this loses its count entirely. (a) is plausible because interval data really do compress the session, but the sheet itself preserves which intervals were scored, so timing information is recoverable. (b) is wrong because percentage-of-intervals is routinely plotted on a line graph. (d) is wrong because the percentage comes from marked intervals divided by total intervals — 6 ÷ 30 = 20% — and interval length is not needed for that step.

An RBT runs a group activity with four learners and cannot watch any one learner continuously. The BCBA asks for a practical estimate of each learner's engagement. Which approach fits the situation?

• a.Continuous duration recording on each learner, timed one at a time
• b.Frequency recording of every engagement response for each learner
• c.Latency recording from each instruction to each learner's response
• d.Momentary time sampling of each learner at the end of each interval✓

(d) is correct: momentary time sampling asks only for a yes-or-no judgment at set moments, so one RBT can cycle through four learners and still produce a usable estimate of engagement. (a) is plausible because duration is the most precise measure of engagement, but continuous timing of four learners at once is not achievable by one observer. (b) is plausible because counting feels simple, but engagement has no clean discrete response to count, and four simultaneous tallies are unmanageable. (c) is plausible when the concern is how quickly learners start, but latency describes response initiation rather than how much engagement occurs.

Which statement correctly contrasts the discontinuous measurement procedures?

• a.Partial interval scores a whole interval; whole interval scores any part
• b.Momentary sampling scores any occurrence; partial interval scores the end
• c.Partial interval scores any occurrence; whole interval scores the full one✓
• d.Whole interval scores the end moment; momentary sampling scores any part

(c) is correct: partial-interval recording marks an interval if the behavior occurs at any point within it, while whole-interval recording marks it only if the behavior lasted the entire interval; momentary time sampling scores only the instant the interval ends. (a) reverses partial and whole, which is the single most common confusion on this topic. (b) swaps momentary sampling and partial interval. (d) mislabels whole interval as an end-moment method and momentary sampling as an any-occurrence method, mixing all three rules together.

An RBT uses momentary time sampling every 5 minutes and scores zero intervals for thumb-sucking, yet the caregiver insists it happened several times that hour. Which explanation is most likely?

• a.The sampling counted occurrences, so the session tally was recorded wrong
• b.The sampling looked only at set moments, so behavior between them is missed✓
• c.The sampling required the behavior to last the whole five-minute interval
• d.The sampling was scored as a rate, so brief occurrences round down to zero

(b) is correct: momentary time sampling observes only at the end of each interval, so a behavior that occurs several times but never at one of those instants produces a score of zero. (a) is plausible because a recording error is always possible, but momentary sampling does not produce a tally of occurrences in the first place. (c) describes whole-interval recording; that rule would also produce zeros, but it is not the procedure the RBT used. (d) is wrong because interval data are summarized as a percentage of intervals, and no rounding step is involved.

A data sheet reads: 'Mark + if screaming occurs at any point in the interval.' A new RBT marks + only when screaming lasted the whole interval. What is the effect of this error?

• a.The recorded data will understate how often screaming actually occurred✓
• b.The recorded data will overstate how often screaming actually occurred
• c.The recorded data will be unaffected, because both rules score intervals
• d.The recorded data will convert the interval measure into a duration one

(a) is correct: the RBT applied a whole-interval rule to a partial-interval sheet, so every interval containing a brief scream went unmarked and the totals fall below the true level. (b) states the opposite direction; overstatement would result from applying a partial-interval rule to a whole-interval sheet. (c) is plausible because both procedures mark intervals, but the scoring rule is what determines the number, so the totals are not interchangeable. (d) is wrong because no time is being accumulated — the sheet still yields marked intervals, just the wrong ones.

A BCBA's protocol specifies 10-second intervals, but the RBT finds these hard to manage and switches to 60-second intervals without telling anyone. Why does this matter for the data?

• a.Longer intervals turn the data into a continuous rather than discontinuous measure
• b.Longer intervals make it impossible to express the results as a percentage
• c.Longer intervals remove the need to define the behavior in observable terms
• d.Longer intervals inflate partial-interval estimates and break comparability✓

(d) is correct: with a longer interval the behavior has more chances to occur somewhere inside it, so the percentage of scored intervals rises even if behavior did not change, and the new sessions can no longer be compared with the earlier ones on the same graph. (a) is wrong because the procedure is still discontinuous no matter how long the interval is. (b) is wrong because a percentage can be calculated from any interval length. (c) is wrong because the definition of the behavior is independent of interval length and is still required.

Two RBTs measure the same 30-minute session. One marks an interval only when reading lasted the entire minute; the other marks an interval when reading was occurring at the moment the minute ended. Their totals differ. Which pair of procedures explains this?

• a.The first used momentary time sampling; the second used partial interval
• b.The first used whole interval; the second used momentary time sampling✓
• c.The first used partial interval; the second used whole-interval recording
• d.The first used whole interval; the second used partial-interval recording

(b) is correct: requiring the behavior to fill the interval is whole-interval recording, and scoring the instant the interval ends is momentary time sampling, so the two observers applied different rules to the same behavior. (a) reverses the assignment, putting the end-moment rule on the first observer. (c) misnames the first observer's rule as partial interval, which would have scored any occurrence. (d) correctly identifies the first observer but calls the second one partial interval, which would have required watching the entire interval rather than only its final instant.

An RBT completes whole-interval recording across sixty 30-second intervals and marks 21 of them. The supervisor asks for the summary that belongs on the graph. What should the RBT plot?

• a.21 responses, plotted as a count of on-task episodes for the session
• b.10.5 minutes, plotted as the total time the behavior was occurring
• c.35% of intervals, plotted as the percentage of intervals scored✓
• d.0.7 per minute, plotted as the rate of the behavior in the session

(c) is correct: interval data are graphed as the percentage of intervals scored, and 21 ÷ 60 = 0.35, or 35%. (a) is plausible because 21 marks look like 21 events, but an interval procedure does not produce a response count. (b) does the arithmetic correctly — 21 intervals of 30 seconds is 10.5 minutes — but whole-interval recording only estimates occurrence, so reporting it as measured duration overstates what the method can support. (d) also computes cleanly, since 21 ÷ 30 minutes is 0.7, but a rate requires a count of responses, which interval recording never produced.

Elena's program targets independently completed math problems. The RBT collects her worksheet at the end of the session and counts the problems she finished without help. Which recording procedure is this?

• a.Permanent product recording, measuring the outcome the behavior produced✓
• b.Direct observation recording, measuring behavior as it occurs in real time
• c.Partial-interval recording, measuring intervals in which work occurred
• d.Latency recording, measuring the time before Elena started each problem

(a) is correct: the RBT is measuring a lasting result the behavior left behind — the completed problems — rather than the responding itself, which is the defining feature of permanent product recording. (b) is plausible because the RBT was present in the room, but the measure itself was taken from the worksheet after the fact, not from watching each response. (c) is wrong because no intervals were timed or scored. (d) is wrong because nothing was timed; the RBT produced a count of completed items.

A group home logs 'bed made' each morning as a permanent product of Marcus's routine. This week staff noticed that a roommate sometimes makes the bed for him. What is the problem with continuing to use the product measure alone?

• a.Permanent products cannot be summarized as a percentage across days
• b.Permanent products require the behavior to have a clear onset and end
• c.Permanent products must be recorded during the session, not after it
• d.Permanent products cannot confirm who produced the result being scored✓

(d) is correct: a product shows that an outcome exists but not who created it, so if a roommate may have made the bed, the data no longer measure Marcus's behavior. (a) is wrong because product data are routinely summarized as a percentage of days or trials. (b) is plausible because onset and offset matter a great deal for duration measures, but a permanent product does not require them at all. (c) reverses the method's central advantage — measuring after the behavior is exactly what permanent product recording allows.

An RBT supports Nia during a 40-minute cooking group and cannot watch her continuously while also assisting other learners. Nia's target is the number of vegetables she chops. What is the main advantage of scoring this as a permanent product?

• a.It records the exact moment each chopping response occurred in the session
• b.It allows the count to be taken after the activity rather than during it✓
• c.It removes the need for an observable definition of the target behavior
• d.It converts the count into a rate that controls for the size of the group

(b) is correct: the chopped vegetables remain after the activity, so the RBT can count them once the group ends instead of tracking every response while also supporting other learners. (a) is the opposite of what a product measure offers — timing information is precisely what is lost. (c) is wrong because a product still needs a clear definition of what counts as a chopped vegetable. (d) is wrong because a rate requires dividing by observation time and has nothing to do with group size.

A BCBA wants to know both how many words Dax writes and how long he stays seated while writing. Which combination should the RBT use?

• a.Permanent product for both measures, scored from the finished worksheet
• b.Direct observation for both measures, scored continuously during work
• c.Permanent product for the word count and direct observation for seat time✓
• d.Direct observation for the word count and permanent product for seat time

(c) is correct: the written words are a lasting product that can be counted afterward, while sitting leaves no product and must be timed as it happens. (a) is plausible because the worksheet is right there, but the paper cannot show how long Dax remained in his chair. (b) would work but wastes the efficiency the worksheet offers; counting words live while also timing seat behavior adds unnecessary load. (d) reverses the two, assigning a product measure to a behavior that leaves no product at all.

Kira's target behavior is greeting peers with a wave. The RBT proposes scoring it as a permanent product at the end of the day. Why will this not work?

• a.Waving occurs too often to be counted accurately by a single observer
• b.Waving must be measured with latency because it follows a peer greeting
• c.Waving is a social behavior, so only caregivers may report on its occurrence
• d.Waving leaves no lasting physical result that can be counted later✓

(d) is correct: permanent product recording requires an outcome that persists after the behavior, and a wave leaves nothing behind to count once the moment passes. (a) is plausible because high-rate behaviors do strain observers, but waving at peers is not typically high rate, and frequency of occurrence is not what rules out a product measure. (b) is wrong because latency is one option among several, not a requirement. (c) is wrong because an RBT observing the setting can record the behavior directly.

What distinguishes permanent product recording from direct observation?

• a.The behavior's effect on the environment is measured after it occurs✓
• b.The behavior is scored only at the moment each interval comes to an end
• c.The behavior is counted by a second observer to check for agreement
• d.The behavior is timed from the delivery of the instruction to its onset

(a) is correct: permanent product recording measures a durable outcome the behavior produced, which means the observer does not have to be watching when the behavior happens. (b) describes momentary time sampling, a discontinuous procedure performed while observing. (c) describes an agreement check, which can be added to any measurement procedure and is not what defines a product measure. (d) describes latency recording, which requires watching the behavior begin in real time.

An RBT is updating Leo's line graph. Today is the first session after the BCBA started a new reinforcement procedure. How should the graph show this?

• a.Circle today's data point and write the procedure's name beside it
• b.Start a second graph so the two procedures are never shown together
• c.Draw a vertical phase-change line before today's point and label it✓
• d.Connect today's point to the last one and note the change in the legend

(c) is correct: a vertical phase-change line placed between the last session of the old condition and the first session of the new one, with a condition label above it, is the standard way a line graph shows that something changed. (a) is plausible because a circled point does draw attention, but it does not mark where one condition ended and another began. (b) is plausible because separate conditions are separate, but splitting the graph destroys the visual comparison the graph exists to provide. (d) is wrong because a legend identifies data paths, not conditions, and connecting across the change hides the boundary.

After drawing the phase-change line on Leo's graph, the RBT is unsure whether to connect the last baseline point to the first treatment point. What is the convention?

• a.Connect them, because the data path should be unbroken across sessions
• b.Leave them unconnected, because data paths break at a phase change✓
• c.Connect them with a dotted line to show that a change was introduced
• d.Leave them unconnected and also remove the baseline points from view

(b) is correct: data paths are broken at a phase-change line so that the reader compares conditions rather than reading a single continuous trend across them. (a) is plausible because within a condition the path is indeed unbroken, but that rule stops at the phase line. (c) is plausible because a dotted line signals something unusual, but the convention is to leave the gap rather than to bridge it with a different line style. (d) gets the break right but then discards baseline, which is the comparison the graph depends on.

An RBT is setting up a new line graph for the percentage of independent responses. Where do the variables belong?

• a.Sessions on the vertical axis and percentage on the horizontal axis
• b.Percentage on both axes so the graph shows change in two directions
• c.Sessions on the horizontal axis and the date on the vertical axis
• d.Percentage on the vertical axis and the sessions on the horizontal axis✓

(d) is correct: the behavior measure goes on the vertical axis and the passage of sessions or time goes on the horizontal axis, which is what makes trend readable from left to right. (a) reverses the axes, a common setup error that makes the trend read vertically and confuses anyone reviewing the graph. (b) is wrong because one axis must carry time. (c) puts time on both axes and leaves the behavior measure off the graph entirely.

While transferring paper data to the electronic graph, an RBT notices that yesterday's entry of 8 was typed as 3. The session has ended and the paper data sheet is intact. What should the RBT do?

• a.Correct the entry to 8 and note the correction as the workplace requires✓
• b.Leave the entry as 3, because a graphed data point cannot be changed
• c.Stop graphing and ask the supervisor to re-enter the whole week of data
• d.Enter the average of the two numbers to reflect the uncertainty

(a) is correct: the original data sheet shows what was actually recorded, so the RBT corrects the transcription error and documents the correction the way the employer requires — this is routine data entry within the RBT's role. (b) is plausible because data integrity rules discourage altering records, but those rules protect the original observation, and here the original supports the correction. (c) is unnecessary escalation: a single known transcription error does not require the supervisor to redo a week of entries, and delaying leaves a wrong graph in place. (d) invents a value that was never observed, which is a fabrication rather than a correction.

Nate's graph plots percentage of correct trials by session. The point for session 9 sits halfway between the gridlines labeled 60 and 80. What value should the RBT report for session 9?

• a.60 percent, because a point is read from the gridline below it
• b.70 percent, because the point falls midway between 60 and 80✓
• c.80 percent, because a point is read from the gridline above it
• d.20 percent, because that is the distance between the two gridlines

(b) is correct: a point positioned halfway between gridlines is read as the midpoint of their values, and the midpoint of 60 and 80 is 70. (a) is plausible because rounding down to the nearest labeled line feels conservative, but it discards half the interval and misstates the value. (c) makes the same error in the other direction by rounding up. (d) reports the spacing between gridlines rather than the value of the point, which describes the axis rather than the datum.

A BCBA asks the RBT to display both aggression and self-injury for Rosa on a single graph. How should the RBT set this up?

• a.Plot the two behaviors added together so one path shows total problems
• b.Plot only aggression, since two behaviors cannot share the same axes
• c.Plot two separate data paths with different markers and add a legend✓
• d.Plot the two behaviors on alternating sessions so the paths never cross

(c) is correct: two behaviors are shown as two data paths distinguished by different markers, with a legend identifying each, so the reader can see how each behavior changed. (a) is plausible because a single total looks simpler, but combining them hides the possibility that one rose while the other fell. (b) is wrong because a graph can carry more than one data path. (d) is wrong because sessions are not assigned to behaviors; both behaviors were measured in the same sessions, and crossing paths are normal.

An RBT finishes a session, sets the data sheet aside, and plans to graph the whole week on Friday from memory of the general pattern. Why is this a problem?

• a.Graphs must be drawn by the supervising BCBA rather than by the RBT
• b.Graphs must be updated daily, because weekly points cannot be plotted
• c.Graphs must show raw counts, and weekly entry produces percentages
• d.Graphs built from memory rather than the sheet can misstate the data✓

(d) is correct: the sheet holds what was actually observed, and reconstructing points from an impression of the week introduces errors into the record the team uses to make decisions. (a) is wrong because entering data and updating graphs is a task RBTs perform. (b) is plausible because timely entry matters, but the flaw here is graphing from memory rather than the interval between sessions and entry. (c) is wrong because graphs can display counts, rates, or percentages; the timing of entry does not change the measure.

An RBT's note says, 'Beckett was frustrated during math and acted out.' The supervisor asks for the same event in observable, measurable terms. Which rewrite is best?

• a.Beckett became upset by the math worksheet and refused to cooperate
• b.Beckett pushed the worksheet away and shouted twice during math work✓
• c.Beckett showed frustration behaviors while doing the math sheet
• d.Beckett did not want to do math and expressed this by acting out

(b) is correct: pushing the worksheet away and shouting are actions any observer could see, hear, and count, which is what makes the note usable as data. (a) is plausible because it names a specific activity, but 'upset' and 'refused to cooperate' are inferences about Beckett's state and intent. (c) is plausible because it sounds technical, but 'frustration behaviors' names a category without saying what Beckett actually did. (d) restates the same inference about wanting and adds the vague phrase 'acting out.'