ABAT - Applied Behavior Analysis Measurement and Data Collection Questions and Answers 2 — Questions and Answers

Question 1: Partial interval recording is likely to OVERESTIMATE the occurrence of a behavior because:

It records only the first occurrence in each interval regardless of duration
A behavior is scored as occurring in an interval if it happens even once, inflating perceived frequency (Correct answer)
It measures only the duration of behavior
It counts behaviors only during specific moments

Correct answer: A behavior is scored as occurring in an interval if it happens even once, inflating perceived frequency

Partial interval recording scores the entire interval as positive if the behavior occurs at any point, so even very brief occurrences are counted as a full interval.

Partial interval recording involves dividing the observation period into equal intervals and scoring whether the target behavior occurred at any point within each interval. Because any occurrence — even a very brief one — scores the entire interval as a positive, this method tends to overestimate the proportion of time the behavior occurs. Conversely, whole interval recording (scoring the interval only if the behavior persists throughout) tends to underestimate behavior. Momentary time sampling tends to be more accurate for higher-frequency behaviors.

Question 2: Which measurement system directly captures the total duration that a behavior occurs during an observation session?

Frequency recording
Duration recording (Correct answer)
Latency recording
ABC narrative recording

Correct answer: Duration recording

Duration recording measures the total elapsed time from the onset to the offset of each behavioral episode, summed across the session.

Duration recording captures the total length of time a behavior occurs. A timer is started when the behavior begins and stopped when it ends; cumulative duration is summed across the observation period. This is the most appropriate measure for behaviors where the total time is meaningful (e.g., time on-task, duration of tantrum, time engaged in stereotypy). Duration can be reported as total duration per session or as a percentage of observation time. Duration recording requires a timer and clear operational definitions of onset and offset.

Question 3: Interobserver agreement (IOA) is calculated to:

Determine the function of a behavior
Establish the reliability and consistency of data collection across observers (Correct answer)
Assess treatment effectiveness
Identify the maintaining reinforcer

Correct answer: Establish the reliability and consistency of data collection across observers

IOA assesses whether two independent observers collect consistent data on the same behavior, supporting the reliability of the measurement system.

Interobserver agreement (IOA), also called interrater reliability, is the degree to which two independent observers record the same data on the same behavior at the same time. IOA is calculated to demonstrate that the operational definition of the behavior is clear enough that observers reliably agree on its occurrence. A minimum of 80% IOA is generally accepted as adequate reliability, though higher standards (90%+) are often recommended. IOA should be collected regularly (BACB recommends at least 20% of sessions) to ensure measurement integrity.

Question 4: A behavior analyst wants to measure how quickly a learner begins a task after the instruction is given. The appropriate measurement system is:

Frequency recording
Latency recording (Correct answer)
Magnitude recording
Whole interval recording

Correct answer: Latency recording

Latency recording measures the time between the onset of a discriminative stimulus (instruction) and the initiation of the behavior.

Latency recording captures the time elapsed from the presentation of an antecedent stimulus (e.g., an instruction) to the onset of the target behavior. It is used when the speed of initiation is a clinically important dimension — for example, measuring how long it takes a student to begin work after instruction, or how quickly a child responds to their name. Latency data can reveal whether a learner is complying with prompts quickly or delaying, which may reflect avoidance or escape motivation.

Question 5: In a reversal (ABAB) design, what is the purpose of the withdrawal phase (second A phase)?

To introduce a new intervention
To demonstrate experimental control by showing behavior returns toward baseline when the intervention is removed (Correct answer)
To collect normative comparison data
To provide a rest period for the learner

Correct answer: To demonstrate experimental control by showing behavior returns toward baseline when the intervention is removed

The second A phase tests whether the behavior change was due to the intervention by removing it and observing whether behavior reverts toward baseline levels.

In an ABAB (reversal) design, experimental control is established by demonstrating that: (1) the target behavior changes when the intervention (B) is introduced, (2) behavior reverses toward baseline when the intervention is withdrawn (second A), and (3) behavior improves again when the intervention is reintroduced (second B). This pattern provides strong evidence of a functional relation between the intervention and the behavior. The withdrawal phase also has ethical implications — the behavior analyst must ensure that temporarily withdrawing a beneficial intervention is ethically justifiable.

Question 6: A multiple baseline design provides experimental control by:

Withdrawing and reintroducing the intervention across phases
Staggering the introduction of the intervention across behaviors, settings, or participants (Correct answer)
Comparing groups receiving different treatments simultaneously
Randomly assigning reinforcement schedules to different behaviors

Correct answer: Staggering the introduction of the intervention across behaviors, settings, or participants

The multiple baseline design demonstrates experimental control by showing behavior changes only when the intervention is introduced, with changes staggered across baselines.

The multiple baseline design (MBD) is the most commonly used single-case experimental design in ABA. It does not require withdrawal of a potentially beneficial treatment to establish control. Instead, the intervention is introduced at different time points across multiple baselines (different behaviors, settings, or individuals). Experimental control is demonstrated if behavior changes in each baseline only when the intervention is applied to it, while untreated baselines remain stable. MBDs are especially appropriate when reversal is unethical or when the target behavior is not expected to reverse.

ABAT - Applied Behavior Analysis Measurement and Data Collection Questions and Answers 2

Partial interval recording is likely to OVERESTIMATE the occurrence of a behavior because: