Area: 6 Measurement of Behavior

Question 1

What are the 3 dimensional quantitites of behavior?

Answer 1

1. Repeatability (countability): behavior can be counted
2. Temporal extent: behavior occurs during some amount of time (e.g., duration)
3. Temporal locus: behavior occurs at a certain point in time with respect to other event (e.g., when behavior occurs)

Question 2

What are examples of repeatability? (3)

Answer 2

1. Count: tally (counting time should always be noted), observation time should be referenced and constant
2. Rate/Frequency: number of responses per unit of time (MOST sensitive measure of change and preferred measure)

• Always reference the counting time
• Calculate correct and incorrect rates of response when assessing skill development
• Take into account the varied complexity of responses
• Do not use rate to measure continuous behaviors that occur for extended periods of time

3. Celeration: rate/per unit of time, measures acceleration and deceleration

Question 3

What are examples of temporal extent?

Answer 3

1. Duration: an appropriate measure for behaviors that occur at a very high rate (e.g., rocking) and task-orientated continous bheaviors that occur for an extended time (e.g., cooperative play)

• T_otal duration per session_: cumulative amount of time, use when increasing the endurance of behavior is the goal
• Duration per occurrence: duration of each time behavior occurs

Question 4

What are examples of measures based on temporal locus? (2)

Answer 4

1. Response (Latency): a measure of the elapsed time between the onset of a stimulus and the initiation of a subsequent response.
   * decreasing latencies can reveal a person’s increasing mastery of some skills
2. Interresponse time (IRT): amount of time that elapses between 2 consecutive instances of a response class.

• important measure when the time between responses, or pacing of behavior, is the focus
• correlated with rate of responding
• important when implementing and evaluating DRL

Question 5

Use percentage of occurrence:

Answer 5

…a ratio formed by combining the same dimensional quantities (e.g. number/number, duration/duration, latency/latency)

5/10= 50%

Question 6

Use trials to criterion (learn units to criterion)

Answer 6

• Used to measure the “cost” of a treatment or instructional method
• Used to compare the relative efficiency of 2 or more treatments or instructional methods
• Used as a DV
• Used to meaure a learner’s increasing competence in acquiring a related class of concepts

Question 7

Describe whole-interval recording

Answer 7

…divide the observation period into a series of equal time intervals. At the end of each interval, they record whether the target behavior occurred throughout the entire interval

• Data obtained with whole-interval recording usually underestimate the overall percentage of the observation period in which the behavior actually occured
• provide only an estimate of the actual occurrence of the behavior

Question 8

Describe partial-interval recording:

Answer 8

…the observer records whether the behavior occurred at any time during the interval

• Ofter overestimates the total duration, it is likely to underestimate the rate of a high-frequency behavior

Question 9

Describe *momentary time sampling: *

Answer 9

…at the end of each interval, record whether the target behavior is occurring at that specific moment

Question 10

What are characteristics of good definitions of behaviors? (5)

Answer 10

1. Objectivity (observable/recordable)
2. Clear/unambiguous
3. Include boundaries
4. Technological (others can replicate it)
5. Discriminates between what is and is not an instance of behavior

Question 11

What are the advantages of sampling techniques? (4)

What are the disadvantages?

Answer 11

+ can provide an estimate of the frequency and duration of the behavior

+ provides an estimate of performance across time intervals

+ indicates when a behavior is likely to occur or not occur

+ permits point-to-point reliability checks

• requires the undivided attention of an observer
• not appropriate for low rates of responding

Question 12

Reliability describes the extent to which a “measurement procedure yields the ___ ___ when brought into repeated contact with the same state of nature”

Answer 12

Reliability describes the extent to which a “measurement procedure yields the same value when brought into repeated contact with the same state of nature”

Question 13

______ trumps _______

Answer 13

Accuracy trumps reliability

Question 14

Measurement that is ___, ___, and ____ yields the most trustworthy and useful data for science and science-based practice.

Answer 14

Measurement that is valid, accurate**, and **reliable yields the most trustworthy and useful data for science and science-based practice.

Question 15

Validity (of measurement) is the extent to which ___ obtained from measurement are ___ relevant to the target behavior of interest and to the reason(s) for measuring it.

Answer 15

Validity (of measurement) is the extent to which data** obtained from measurement are **directly relevant to the target behavior of interest and to the reason(s) for measuring it.

Question 16

The procedure that measures behavior after it has occurred by measuring its effect on the environment is called:

Answer 16

Permanent product

Question 17

Marco is learning to assemble computer components as part of his new job at a factory. Speed and accuracy are critical skills for him to learn in assembling his component. What is the most appropriate method for measuring his behavior?

Answer 17

Permanent product

Question 18

Given that one has the necessary resources, which of the following discontinuous response measures would be the best to use if the goal is to produce a decrease in stereotypic behavior which is occurring during most of the client’s school hours?

Answer 18

Partial interval recording with 20 sec. intervals during six 10 min. observation periods per school day.

Question 19

A behavior analyst is unable to obtain adequate interobserver agreement when measuring Carla’s duration of appropriate play behavior during recess. Which discontinuous response measure would yield the most conservative estimate of play behavior in terms of making proper treatment decisions, yet be likely to yield high interobserver agreement? 6-4

Answer 19

Whole interval recording of 10 sec. intervals during four 3 min. observation

Question 20

Which response measure would be the best and most conservative discontinuous response measure for behavior targeted for decrease that occurs at very high rate but varies in duration?

Answer 20

Partial interval recording

Question 21

Of the intervals below, which best represents the size range that is acceptable for partial interval recording?

Answer 21

10 to 20 seconds intervals.

Question 22

Missy hit her head, which was covered by a helmet, 120 times during a 1 min. observation session. The head-hitting was constant and steady during the entire interval. What is the BEST estimate of the average IRT?

Answer 22

0.5 seconds

Question 23

Elbert is working on a project to increase cooperative play in a pre-school program. He has an assistant doing data collection. Of the following, which recording system would be the most appropriate for the assistant to conduct? 6-4

Answer 23

Whole interval with 15 second intervals

Question 24

Chata suggests that staff collect some IRT data. The data turns out like this: They begin recording at 11:00 am; there is aggression at 1 pm, 2:00 pm, and 9:00 pm. Recording ends at 10:00 pm. What is the average IRT?

Answer 24

4 hours

Question 25

Matt is consulting on a case in which aggressive behaviors tend to occur right after lunch. What observation times should be used? Observation sessions should be

Answer 25

Conducted right after lunch.

Question 26

Behaviors appropriate for partial-interval data collection are those behaviors of varying ___ and behaviors ___ or ___ a clear onset and offset. Partial interval data collection approximates the frequency (or event) and duration (where duration varies) measures.

Answer 26

Behaviors appropriate for partial-interval data collection are those behaviors of varying duration** and behaviors **with** or **without a clear onset and offset. Partial interval data collection approximates the frequency (or event) and duration (where duration varies) measures.