Unit 6 Flashcards
External validity
The degree to which a study’s finding’s have generality to other subjects, settings, and/or behaviors.
Approaches to functional behavioral assessment
- FBA interview
- ABC narrative recording
- Rating scales (motivational assessment scale)
- Scatterplot analysis
- Descriptive analysis
- Brief FBA
- Antecedent A-B analysis
- Functional A-B-C analysis
- Progressive analyses
Threats to external validity
Differences between the conditions of assessment and the client’s natural environment in the functional properties of motivating and reinforcing events
Most valid functional behavioral assessment approach
Combination of methodologies
Internal validity
Establishes that problem behavior is a function of specific motivating and maintaining events
Threats to internal validity
Events that co-occur with the independent variable that could account for the findings
Reasons for conducting a descriptive analysis
- Identify idiosyncratic forms of motivating and reinforcing events
- Design experimental FA conditions that more closely represent natural conditions
- Estimate natural schedules of reinforcement maintaining problem behavior
- Interpret the results of experimental FA methods
Approaches to descriptive analysis
• Scatterplot analysis
• Coding behaviors without reference to specific
motivational or reinforcing environmental events
• Direct recording of behavior in relation to
possible motivating and reinforcing events
Touchette, MadDonald, & Langer (1985)
Results showed that a scatter plot is a more practical way to collect data by direct-care staff across an entire day or shift and produced the same conclusions as more labor intensive methods of data collection
Bijou, Peterson, & Ault (1968)
• First published descriptive assessment
• Developed to empirically study child development
• Real-time measurement of child behavior and
environmental events
• Important for operational definitions of behavior
Mace & Lalli (1991)
- Combined descriptive and experimental analyses in the study of bizarre speech
- Found that bizarre speech was functionally related to attention
Ndoro, Hanley, Tiger, & Heal (2006)
- Used conditional probabilities to assess student and teacher interactions
- Found that teachers were more likely to give “Do” directives than “Don’t” directives and that there was a greater probability of teacher attention given noncompliance than compliance
Lalli, Browder, Mace, & Brown (1993)
Possibly the first descriptive analysis research to employ a fully comprehensive approach to functional behavioral assessment
More rigorous and accurate method of data collection (according to Dr. Mace)
Count within interval recording (rate)
Data collection method that yields the best estimate of duration (according to Dr. Mace)
Momentary time sampling
Data collection method that yields the best estimate of duration (according to Dr. Mace)
Momentary time sampling
Best inter-observer agreement measure
Exact agreement on an interval by interval basis
p (target behavior | MO)
Represents the probability of behavior when an antecedent event is present
(Reflects the power of the MO to evoke behavior)
p (subsequent event | MO-TB)
Represents the probability of a consequent event when a motivating operation and target behavior are present
(Estimates natural schedule of reinforcement)
Best practice to set temporal gaps between events
Use natural lines of fracture (“windows”)
Sasso et al. (1992)
Conducted a conventional analysis and ABC analysis, and a classroom analysis and found the
same outcomes across analyses
Lerman & Iwata (1993)
Results indicated that the descriptive analysis was useful in identifying the extent to which SIB was related to social versus nonsocial contingencies, but was limited in its ability to distinguish between positive and negative reinforcement (i.e. attention versus escape).
Thompson & Iwata
Found that social consequences arranged by the experimental analysis did match naturally occurring events in the environment
Reasons to use a line graph display of data
• Tells the story over observational samples
• Provides a more “fine grain” way to analyze the
data
• Shows the variability in conditional probabilities
over repeated samples
Convenient, practical, less labor-intensive methods of data collection
• Count of episodes of a target behavior across the day
• Count during large time periods across the day
• Rating scales with defined, numeric anchors
across the day or during shorter periods
• Rating scales without anchors
• Occurred or did not occur rating across the day
or during periods
Rigorous and accurate methods of data
collection
• Partial interval recording
• Whole interval recording
collection
• Count within interval recording • Momentary time sampling
Reasons for combining descriptive and experimental methods
- DAs can help identify idiosyncratic forms of motivating and reinforcing events
- This information can inform the design of experimental conditions that better represent natural conditions
- Natural schedules of reinforcement can be calculated to be used to design EAs
- Comparing DA and EA results can help estimate how valid the FA findings are
