Kazdin Chapters 5 & 6 Flashcards
Accuracy
Extent to which observations correctly reflect the true performance of the clien
Intraobserver consistency
Consistency to which one observer scores the same data when looking at the same session at different points in time
Interobserver agreement
The consistency of two or more observers obtaining the same score when looking at the same behaviors at the same time
Why is it harder to obtain accuracy data for behavioral observations?
There is less likely to be a permanent product to assess
When should you check interobserver agreement?
In each phase (a few times)
When is a frequency ratio used?
When behavior can be freely formed or take on any value (“continuous”). Ex. A child cursing at the dinner table, there is no set amount
Main consideration when using frequency ratios
It doesn’t tell you about interobserver agreement Ex. Mom observed 18 swears, dad heard 20
When is point-by-point agreement ratio used?
When the behavior is discrete and observers agree on it’s occurrence
Problems w/ point-by-point agreement ratio
What counts as an agreement? Frequency of a behavior can influence agreement
Chance level of agrement
Level of agreement that would be expected by randomly marking occurrences for given number of intervals
How to handle chance levels of agreement due to high base rate?
Calculate instances of occurrence and nonoccurence , plot agreement data, use correlational statistics
Reactivity
Changes people make when their behavior is being monitored or evaluated
Observer drift
Tendency for observers to change the manner in which they apply definitions of behavior over time
Observer expectancies and feedback
Expectancies of observers regarding client’s behavior and feedback observers receive from the experimenter in relation to that bevhaior
General Requirements of Single-Case Designs
Repeated, Ongoing Assessment
Baseline Assessment
Stability of Performance (Trend/Variability)
Two functions of baseline assessment
Descriptive - describes existing level of performance
Predictive - serves as basis for predicting level of performance for immediate future
ABAB Design Definition
Examines the effects of an intervention by alternating the baseline (A), when no intervention is in effect with the intervention condition (B phase) and are repeated to complete four phases
Additional function of second A phase in ABAB design in addition to description/prediction
Test prediction from prior phases
Logic of ABAB design
Making and testing predictions about performance under different conditions
Methodological Issues when Same Subjects Receive Two or More Interventions
Order/sequence effects - is it more effective because it came first or second?
Carryover effects
Potential Solutions to Comparing Two/More interventions
- Washout period (ABAC design)
- Counterbalance - give them in a different order to different subjects
