SCD data collection, data analysis, and designs Flashcards
Importance of detailed behavioral definitions and technological descriptions of the importance of behavior-change programs
Baer, Wolf, & Risley (1968)
Alternating-treatments design – Different interventions are alternated during the intervention phases
Ulman & Sultzer-Azaroff (1975)
Simultaneous-treatment design – Different conditions are administered in the same phase, usually on the same day
Hersen & Barlow (1976)
SCR has been used to document interventions that are functionally related to change in socially important outcomes
Wolf (1978)
Six features of SCR visual analysis: (1) level, (2) trend, (3) variability, (4) immediacy of effect, (5) overlap, and (6) consistency of data patterns across similar phases
Parsonson & Baer (1978)
Four types of experimental validity – Internal, external, construct, and statistical conclusion
Cook & Campbell (1979)
Adapted ATD – Compare instructional practices with non-reversible behaviors (functional, developmental, academic)
Sindelar, Rosenberg, & Wilson (1985)
- Documentation of functional relationship requires compelling demonstration of an effect
- Compromised when: Long latency between manipulation of independent variable and change in dependent variable, mean changes across conditions are small, trends do not conform to those predicted
Parsonson & Baer (1992)
- SCR employs within- and between-subjects comparisons to control for major threats to internal validity
- Requires systematic replication to enhance external validity
Martella, Nelson, & Marchand-Martella (1999)
- SCR is experimental and its purpose is to document functional relationships between independent and dependent variables
- Involves replication of the intervention in the experiment – Introduction and withdrawal, iterative manipulation, staggered introduction
- SPED is a field that emphasizes individual student as unit of concern, active intervention – responders and nonresponders, and practical procedures for behavioral intervention and experimental effects in educational conditions, testing of conceptual theory, cost effective (problem-solving discipline)
- Individual participant is unit of analysis – Each participant serves as their own control; compares performance prior to intervention with during and/or after intervention
- Visual analysis examples: Level – mean performance during a condition, trend – rate of increase or decrease of the best-fit straight line for the dependent variable within a condition, variability – degree to which performance fluctuates around a mean or slope during a phase, immediacy of effects, overlap, magnitude of changes, consistency
Horner et al. (2005)
Knowing precisely why and how change occurs can be important for maximizing the impact of the intervention and extending the intervention to other settings
Kadzin (2007)
Regression-based estimators are probably best justified for estimating effect size with sensitivity analyses
Shadish, Rindskopf, & Hedges (2008)
Propose statistical analysis may add to confidence of visual analysis of data, quantify strengths of outcomes, and increase objectivity of analysis
Campbell & Herzinger (2010)
Responsibility of researchers to report fidelity of implementation of each step of a behavior-change program by condition
Gast (2010)
• Visual analysis – Four steps
Steps:
• 1. Documentation of predictable baseline pattern
• 2. Examining data within each phase of the study to assess within-phase patterns
• 3. Compare data from each phase with the data in the adjacent phase to determine effect
• 4. Integrate all information from the phases of the study to determine whether there are at least 3 demonstrations of an effect at different points in time
• Immediacy of effect – Change in level between the last three data points in one phase and the first three data points of the next
• Overlap – Proportion of data from one phase that overlaps with data from the previous phase
• Consistency of data in similar phases – Looking at data from all phases within the same condition and examining the extent to which there is consistency in the data patterns from phases with the same conditions
• No agreed upon methods or standards for effect size estimation
Kratochwill et al. (2010)
- Threads to validity – Methodological issues that are likely to rival the explanation that it was the intervention that explained the effect
- Internal validity – Extent to which an experiment rules out alternative explanations of the results
- Threats to internal validity – Factors or influences other than the independent variable that could explain the results : History (any event occurring at time of experiment that could influence results), maturation, instrumentation, testing (effects of repeated assessment), statistical regression, diffusion of treatment
- External validity – Extent to which the results of an experiment can be generalized or extend beyond the conditions of the experiment
- Threats to external validity – Characteristics of the experiment that may limit the generality of the results: Generality across subjects, responses or measures, settings, time, behavior-change agents; reactive experimental arrangements, multiple-treatment interference
- Construct validity – Explanation of the causal relation between the intervention and outcome (Is the reason for the relation between the intervention and behavior change due to the construct given by the investigator?)
- Threats to construct validity – Attention and contact accorded the client; special stimulus conditions, settings, and contexts
- Data-evaluation validity – Aspects of data can interfere with drawing valid inferences
- Threats to data-evaluation validity – Excessive variability, unreliability of measures, trends, insufficient, mixed data patterns
- Reliability – Consistency of measure or measurement procedure
- Interrater reliability – Extent to which different assessors, raters, or observers agree on the scores they provide when assessing, coding, or classifying subjects’ performance (percent agreement, pearson product-moment correlations, kappa)
- General requirements – Continuous assessment, baseline assessment, stability in performance (trend in the data, variability in the data)
Kazdin (1982, 2011)
- Graphically represented to analyze trend, level, and stability
- All conditions remain constant within exception of the introduction of one variable in the intervention condition
- Researchers strive for IOA between 80 and 100%
- Implementation fidelity gives credence to interpretations of data using visual analysis
- Primary goal of visual analysis is to identify if a functional relation exists between the introduction of an intervention and change in a socially desirable behavior, as well as replicate effects across multiple participants
- Allows researchers to analyze each participant’s behavior through repeated measurement and evaluation, observe abrupt and subtle changes over time
- Generality – Provide detailed descriptions of participants preintervention behaviors to increase likelihood of understanding for whom and under what conditions interventions may be effective
- Agreement for results of visual analysis
Lane & Gast (2013)
Visual analysis currently provides the best option to analyze effects of single-subject studies
Cook et al. (2014)
Three demonstrations is the minimum required for the design to be considered experimental in nature (causal attributions can be made and functional relations can be demonstrated)
Gast, Ledford, & Severini (2018)
MB and MP differ in one way – frequency in which pre-intervention data are collection; MB require plan for continuous measurements of all targets prior to intervention; MP collects data intermittently prior to introduction of intervention
Gast, Lloyd, & Ledford (2018)