Definition of Measurement
 Is the process of applying quantitative labels to
observed properties of events using a standard set of
rules
 Designed to know what is included and excluded
Why is Measurement Necessary?
 Without measurement, science is only guesswork and
opinion
 Is a philosophy
 ABA
 Measure behavior to answer questions
 Is the basis for talking about behavior
 Allows you to determine if a behavior has changed
Practical Applications
 Also used to measure behavior and evaluate effects of
interventions
 Used before, during, and after treatment
 Used to guide decision making
 Used to
 Discontinue ineffective treatment
 Continue or increase effective treatment
Benefits of Measurement
 Optimize effectiveness
 Verify legitimacy of treatments
 Identify and end use of pseudoscience
 Accountability
 Meet ethical standards
Features of Measurement
 Has dimensions or distinct features that can be
measured
 Three fundamental properties
 Countability: behavior can be counted
 Repeatability: Get the same result over and over
 Temporal Extent: Duration of the Behavior
 Temporal Locus: When does the behavior occur or stop
Counting
 Number of responses emitted during an observation
period
 Periods can vary
 Reported as frequency count
 Problem
 May not provide sufficient information for analysis

Duration may be more important variable
Repeatability / Reliability
 Do you get the same result over and over?
 When observing or examining a behavior , should get
similar results during different observation periods.
 If not, may have multiple stimuli controlling or
influencing the behavior.
Rate/Frequency
 Is a ratio of count per observation period
 More meaningful than count alone
 Include counting time for reference
 Number of Responses / 10 Minute Period
 Can also examine rate of correct vs. incorrect
responses
 Correct responses / Total responses
 # Widgets made correctly / Time or Total Widgets
made
 Is helpful in skill development
 Usually reported as number / standard unit of time
Guidelines When Using Rate
 Take complexity of response into account
 May need to break out into different responses
 Is a very useful measure for free operants
 Not appropriate
 For responses within discrete trials (end of a class
period)
 For continuous behavior over extended period
 Both can be adapted if needed.
Celeration
 Is a measure of the change in rate of responding per
unit of time
 Behavior studying / 15 minutes
 Examine stimuli influencing the behaivor
 Reported using Standard Celeration Chart
 Captures behavior acceleration and deceleration
 Can be used to identify stimuli that are influencing
behavior
Duration of Responding
 The amount of time a behavior occurs
 Total duration of session
 Duration of each occurrence
 Reported in standard time units
 Counting plus duration measures provide different
pictures of same behavior
 May be critical in changing a behavior
Response Latency
 Measure of elapsed time between onset of stimulus
and initiation of response
 Studying and acting out
 Typically reported using mean, median, and range
Interresponse Time
 Amount of time that elapses between two consecutive
instances of a response class
 Direct measure of behavior
 Is related to rate of responding
 Reported using mean, median, and range
Derivative Measures / Percentages
 Are ratios formed by combining the same dimensional
qualities
 Expresses a proportional quantity
 E.g., Correct to incorrect responses
 Proportion of observation intervals when behavior
occurred
Considerations for Using
Percentage
 Often is used incorrectly
 Most accurate with divisor of 100 or more
 May be misleading. Can indicate a behavior change
when an outlier observation impacts the percentage
 Has limited use because has no dimensional quantity
 Often sets artificial limits on behavior change
 May not be practical
Trials-to-Criterion
 Measure of the number of response opportunities
needed to achieve a predetermined level of
performance
 Can be combined or used to determine trials-tocriterion (e.g., rate)
 Typically are calculated post facto
 Used to compare effectiveness
Topography
 Is the physical form or shape of a behavior
 What does the behavior look like
 Anger: Facial expression vs. hitting someone
 Can have measurable dimension
 Must be careful not to interpret.
 Look at defined behvior
 Malleable by consequences
 Is not a fundamental quality of behavior
Magnitude
 Is the force or intensity with which a response is
emitted
 Important parameter for some response classes
 E.g., voice volume
 Is not usually a fundamental quality of behavior
 Try to measure with objective systems
 Automatic Measuring Meter for Loudness
Procedures for Measuring Behavior

1.
2.
3.
Typically involve one or a combination of these
three:
Event recording
Duration
Time sampling methods
Event Recording
 Uses procedures for detecting and recording the
number of times a behavior is observed
 Devices include:
 Clickers, wrist counters, digital counters, masking tape,
paper clips, match books, etc
Issues
 Easy to do
 Behavior must have discrete beginning and ending
 Rate must not be too high
 Inappropriate for behaviors with long duration
Problems
 Need good observers
 In applied settings (bars, restaurants, parks) may miss
behaviors
 Solutions
 Work in teams

Have another team observe the first team
 Recording devices – Cameras

Allows teams to examine and recheck behavior.
Time Recording
 Procedures to measure duration, response
latency, and/or interresponse time
 Duration Techniques
 Computer systems, stopwatch, wall clocks, tape
recorder
 Response latency and interresponse time
techniques
 Precise recording of duration between events of
interest
 Can use similar techniques
Time Sampling
Whole-Interval Recording Method
 Used to measure continuous behavior
 Brief intervals (5-15 seconds)
 Record if behavior occurs throughout the interval,
 Problems:
 Risk of underestimation
 Reported as percentage of intervals when behavior
occurs
Partial-Interval Recording
 Record if behavior occurs at any time during interval
 Yes it occurs – No it does not
 Multiple occurrences are scored as one
 Does not capture duration
 Allows recording of multiple behaviors
 Reported as percentage of intervals when behavior
occurred
Momentary Time Sampling
 Records whether the behavior is occurring at the end
of the interval
 Does not require undivided attention
 Misses much behavior
 Best for continuous behavior
 Reported as percentage of intervals when behavior
occurred
Planned Activity Check
 At end of interval, measure number of students
engaged in target activity
 Variation of momentary time sampling
 Measures behavior of individuals within a group
Guidelines for Time Sampling
 Use a timing device to signal beginning and end of
observation
 Increase accuracy
 Not distracted by watching a stopwatch
 Record a response for every interval (e.g., yes or no)
 Prevents losing your place with blank intervals
 Can use technology to help you
 Cameras etc.
Artifactual Variability
 Artifact is something that appears to exist because of
the way it is examined or measured
 Time sampling provides estimate of actual occurrences
 Different procedures produce different results
 Differences produce variability in data
Measuring Behavior by Permanent
Product
 Measures its effects on the environment
 Ex post facto
 All previous procedures can be applied to permanent
product measurement
 Products can be a natural or contrived
Advantages of Permanent Product
Recording
 Practitioner free to do other tasks
 Possible measurement of otherwise inaccessible
behavior
 More accurate, complete, continuous
 Easier data collection (IOA, treatment integrity)
 Allows for the measurement of complex behavior
Determining Appropriate Use
 Is real-time measurement needed?
 Are moment to moment decisions required
 Can behavior be measured as a permanent product?
 Each occurrence must produce same product
 Product can only be produced by target behavior
Other Issues
 Reactivity effects
 Observing behavior can change the behavior
 Usually decreases over time
 What are costs to obtain and measure the permanent
product?
 Availability, cost and effort of generating the product
Computer-Assisted Measurement
 Data collection and analysis software combined
 Multiple systems available
 Sophisticated and easy to use
 Laptops, hand-held computers, PDAs
 Simultaneous recording of multiple behaviors across
multiple dimensions
Conclusions
 Measuring behavior is important
 Allows you to determine if changes are occurring
 Extremely important for working in applied settings