BCBA Exam (Board Certified Behavior Analyst) — All Questions

Which is the best operational definition of hitting for a behavior plan?

• a.Acting out physically in the classroom
• b.Showing anger physically when a request is denied
• c.Striking a person with an open or closed hand✓
• d.Being aggressive toward peers during free play

An operational definition describes the observable topography of the behavior so that two observers can agree on each occurrence. Striking a person with an open or closed hand is observable and countable. Being aggressive, showing anger and acting out are labels or inferred states; observers could disagree about what counts, so they are not operational definitions.

A teacher counts the number of correctly completed math worksheets left in a student's tray at the end of each day. This is

• a.a permanent-product measure✓
• b.a discontinuous measure
• c.an indirect measure
• d.an interobserver agreement measure

Measuring the lasting outcome a behavior leaves in the environment, here completed worksheets, is a permanent-product (product) measure, one of the three types the TCO distinguishes from direct and indirect measures. Indirect measures rely on reports such as interviews or rating scales. Discontinuous measures sample behavior during observation intervals, and interobserver agreement compares two observers rather than measuring behavior.

During a 30-minute observation, a student calls out 18 times. What is the rate of calling out?

• a.6 per minute
• b.0.06 per minute
• c.0.6 per minute✓
• d.1.67 per minute

Rate is count divided by observation time: 18 responses / 30 minutes = 0.6 responses per minute (check: 0.6 x 30 = 18). 1.67 is the inverted ratio (30/18). 6 and 0.06 misplace the decimal point.

A feeding therapist wants to know whether a child is eating too fast, so she records the time between two successive bites. She is measuring

• a.interresponse time✓
• b.response latency
• c.rate per session
• d.episode duration

Interresponse time (IRT) is the time elapsed between two successive responses, which captures how closely bites follow one another. Latency is measured from a stimulus to a response, not between responses. Duration is how long a single response lasts, and rate per session is a count over time that does not show the spacing between individual bites.

In whole-interval recording, an interval is scored only if the behavior occurs throughout the entire interval. Compared with the true proportion of time the behavior occurred, this method tends to

• a.underestimate it✓
• b.match it exactly
• c.double it
• d.overestimate it

Because any interval with even a brief gap in the behavior is scored as a non-occurrence, whole-interval recording can only miss behavior, never add it, so it tends to underestimate total duration. Partial-interval recording, which scores an interval if the behavior occurs at any point, is the method that tends to overestimate. Neither method matches the true value exactly except by chance, and nothing about the rule produces a doubling.

A teacher of 22 students wants to check on-task behavior without stopping instruction. A vibrating timer prompts her every 5 minutes, and she notes whether each target student is on task at that instant. This is

• a.momentary time sampling✓
• b.event recording
• c.partial-interval recording
• d.whole-interval recording

Recording whether the behavior is occurring at the moment an interval ends is momentary time sampling, which suits observers who cannot watch continuously. Partial- and whole-interval recording require watching throughout each interval. Event recording counts every occurrence, which is not possible while teaching a full class.

Two observers count a student's call-outs in the same session. Observer 1 records 16 and Observer 2 records 20. What is the total count IOA?

• a.125%
• b.80%✓
• c.90%
• d.20%

Total count IOA divides the smaller count by the larger and multiplies by 100: 16 / 20 = 0.80, so 80% (check: 20 x 0.8 = 16). Dividing the larger by the smaller gives 125%, which is not possible for an agreement index. 20% is the disagreement, and 90% has no basis in these counts.

A family's concern is how long a child's tantrums last, but staff record only how many tantrums occur each day. The main problem with this measurement is its

• a.sensitivity to rate
• b.reliability
• c.interobserver agreement
• d.validity✓

A measure is valid when it captures the dimension of behavior relevant to the question. Counting tantrums can be perfectly consistent (reliable) and have high agreement between observers while still missing the dimension of concern, duration. A count already is sensitive to rate, which is not the issue.

On a cumulative record, a flat horizontal segment indicates that during that period

• a.the reinforcer was delivered
• b.responding was at its highest rate
• c.no responses occurred✓
• d.responding declined to half its rate

A cumulative record adds each response to the running total, so the slope shows response rate. A flat segment means the total did not increase, that is, no responses occurred. The highest rate appears as the steepest slope, and a decrease in rate appears as a shallower but still rising slope, because a cumulative record cannot go down.

A BCBA plans to start an intervention, but the baseline data points vary widely from session to session with no clear pattern. The best next step is to

• a.average the baseline points and start treatment
• b.extend baseline until a stable pattern emerges✓
• c.switch to comparing the means of two groups
• d.start treatment now to reduce the variability

Single-case logic depends on a stable baseline that allows prediction of future behavior; with high variability, a treatment effect cannot be distinguished from ongoing fluctuation. Starting treatment immediately, or averaging the points, hides the variability instead of resolving it. Group means do not address the problem of predicting this individual's behavior.

A consultant needs to show a school board the average rate of problem behavior in each of four separate conditions from a completed study, as one summary figure. Which graph best suits this purpose?

• a.A cumulative record
• b.A bar graph✓
• c.A scatterplot of sessions by time
• d.An equal-interval line graph of sessions

Bar graphs display discrete, separate categories and summary values such as means per condition, which is the stated purpose. A cumulative record and a session-by-session line graph show change over time, which is valuable for ongoing decisions but is not a summary of four conditions. A scatterplot shows the relation between variables or the timing of responses.