occurrence/nonoccurrence agreement
- More conservative measure than interval agreement
* Calculated as interval agreement
functional relation
- An intervention reliably produces a behavior change
* Systematic manipulation of IV results in consistent effect on DV
baseline
- Background or context for viewing effects of a second type of condition
- Used to establish initial pattern of behavior
- Assessment of the DV prior to the introduction or change of the IV
dependent variable
- Behavior targeted
* Changes as a function of IV
kappa coefficient
- Consistency in recording categorical data reported conservatively
- Well suited for categorical and interval datak= (Po - Pc)/(1 - Pc)
- Po = proportion of agreements on occurrences and nonoccurrences
- Pc = proportion of expected agreements on the basis of chance
replication
• Repeated predictions and verifications within the same study and across studies
total agreement
- Total number of responses by each observer
- Index of overall occurrence of behavior(smaller total)/(larger total)× 100
• Less conservative than interval agreement
independent variable
- Intervention
* Particular aspect of environment manipulated by experimenter to determine its effect on behavior
experimental control
- A predictable change in behavior (DV) reliably produced by systematic manipulation of some aspect of the person’s environment (IV)
- Established by demonstrating functional relation
- Requires at least 2 manipulations
external validity
- Extent to which an experiment demonstrates that the intervention accounts for change
- Extent to which results can be generalized
causal inference
• Intervention is responsible for behavioral change
testing effect
• Exposure to test (e.g., dependent measurement)
hypothesis
- Prediction
* If IV has no effect, data path of DV will be unchanged
interval agreement
• Point-by-point
• Agreement on occurrence of behavior for each interval
agreements/(agreements + disagreements)× 100
operationalized
- Specified so that it is replicable
* Research questions stated in observable and measurable terms
internal validity
- Extent to which observed changes in the DV are due to the IV (and not due to extraneous variables)
- Ruling out alternative explanations for results
- Shows that researcher has established experimental control
selection bias
• Sample does not represent population (not relevant to SCD)
interobserver reliability
- Consistency of measurement
- Applicable to DV and IV
- Agreement between observers is quantifiable
treatment fidelity
- Procedural validity; procedural integrity
- Verification that IV was implemented accurately
- Accuracy in implementing experimental treatment
threats to validity
• Methodological issues that are likely to rival the explanation that the intervention explained the effect
threats to internal validity
- History
- Maturation
- Testing
- Instrumentation
- Statistical regression
- Selection bias
- Attrition (mortality)
- Diffusion of treatment
threats to external validity
- Generalizability of findings
- Reactive assessment
- Pretest sensitization
- Multiple treatment interference
