Measurement Flashcards
Prepare for data collection
●Review data from the last session
●Determine targets to be run in upcoming session based on the data review
●Gather materials needed for upcoming session
●Set up for the session
Continuous measurement procedures
●Frequency
●Rate
●Duration
●Latency
●Interresponse time
Continuous measurement procedures
- What are they?
- What do they measure?
Used to measure all instances of a behavior.
- Record every time the behavior occurs.
- Most used
- Most accurate
- Records all instances of behavior
- The start and stop of a behavior is clear
Frequency
A count for each time the behavior occurs.
- Number of times a behavior occurs.
-When:
Frequency is best to use when the observation length is the same
●The observation length is consistent from day to day
●The behavior is discrete and short in duration
Rate
Frequency + Time component
- frequency / Time
- When:
The observation length, or session, is not consistently the same.
●The observation length varies from day to day
●The behavior lasts for more than a few seconds
Duration
Total length of time a behavior occurs.
- Measures how long a behavior occurs
- Often combined with frequency, it offers more information.
When:
●There is a clear beginning and end to the behavior
●The length of the behavior is a concern
●Behavior occurs at a high frequency
Latency
The amount of time from the onset of the stimulus to the start of the response.
- Time between the presentation of a stimulus and a response.
- Often used to increase or decrease a behavior.
When:
●There is a clear beginning and end to the behavior
●It is important to know the “time to respond”
IRT (Interresponse Time)
The amount of time between responses.
- Time between the presentation of two consecutive responses
When:
●There is a clear pause between two consecutive responses
●Determining if a behavior should be counted as more than one instance.
Discontinuous Measurement
Measures a sample of behavior that occurs by breaking the session down into small increments of time.
- Estimates the occurrence of a behavior
* Less accurate that continuous data collection methods
* Easier to collect in busy environments
●Partial interval
●Whole interval
●Momentary time sampling
Partial Interval
Behavior is recorded if it occurs at least once during the defined interval.
- A time period is broken into intervals.
- Records if the behavior occurred at least once during the interval,
- Data are entered as a percentage.
* Overestimate the occurrence of the behavior
When:
● The behavior happens quickly
● The behavior does not last long
● Behavior does not have a clear stop and start
● Such a high rate of behavior it would be difficult to record each instance.
● You want to decrease behavior
● It is not possible to observe continuously
+ provides an estimate of frequency and duration of a behavior.
+ provides information about where behaviors are occurring across observational sessions.
- Requires an observer’s undivided attention.
Whole Interval
Behavior occurs during the entire defined interval.
- Measure ongoing behaviors that you know will continue across intervals.
* Underestimates the occurrence of the behavior
when:
● The behavior is not easily counted
● The behavior does not have a clear beginning and end
● The behavior occurs at a high frequency
● You want to increase a behavior
+ Provides an estimate of the duration of a behavior and provides information about where behaviors are occurring or not occurring within an observational session.
- Requires an observer’s undivided attention. Can be challenging when attention is divided.
Momentary Time Sampling
Measures behaviors that occur at the end of an interval.
- Records if the behavior occurs only at the end of the interval.
* Tends to underestimate and overestimate the occurrence of the behavior.
When:
● The behavior is not easily counted
● The behavior does not have a clear beginning and end
● It is not possible to observe continuously
