Week 3 Flashcards
A-B-A-B Design
- Baseline and treatment phases are implemented twice
Multiple Base Design
- Multiple Baseline Across Subjects Design
- Multiple Baseline Across Behaviors Design
- Multiple Baseline Across Settings design
Multiple Baseline Across Subjects Design
Same target behavior but with 2 or more subjects are observed
Multiple Baseline Across Behaviors Design
2+ or more behaviors but with same subject are observed
Multiple Baseline Across Settigns Design
2+ or more settings but with same subject as well as same behavior are observed
Alternating Treatments Design
Baseline and two treatments conditions which are then conducted in rapid succession. You then see which intervention/treatment was the most effective
Changing Criterion Design
[Think of it as Successive Goals]
- Includes a baseline and treatment phase
- Successive goals are identified during the treatment phase
- Success is determined by whether the subject is able to change behavior to meet the goal
Ex: Subject has until next week Thursday to play with friends. Subject is able to change his bad attitude into a good one and was able to gain a friend thus being successful
Evaluation of Design
- Evaluate The Design
If it:
meets design standards or meets with reservations
Then:
Evaluate the evidence
If it:
Has strong evidence or moderate evidence
Then:
Evaluate the effect size estimation though social validity assessment
Evidence Standard (Strong)
Strong:
If your baseline has:
- Documentation of research question “problem”
- Documentation of predictable pattern (≥ 5 data points)
In Each Phase of the Analysis
- Documentation of predictable pattern (≥ 5 data points)
Basic Effects
- Documentation of predicted change in the DV when the IV is manipulated
Experimental Control
- Three demonstrations of basic effect, each at a different point in time
- No demonstrations of intervention failure
Basic Effects
- If you have documentation of the predicted change in the DV when the IV is manipulated
Experimental Control
- Three demonstrations of basic effect, each at a different point in time
- No demonstrations of intervention failure
Evidence Standard (Moderate)
- All of “Strong” criteria, with these exceptions:
- If you only have 3-4 data points per phase
- If you have three demonstrations of effect but in additional demonstrations you fail to-document an effect
Evidence Standard (No Evidence)
- Misnomer
- Evidence does not meet moderate level
Visual Analysis within Single-case design
Documenting Experimental Control
- Three demonstrations of basic effect at three different points in time
To assess basic effect
Visual Analysis include simultaneous assessment of:
- Level, Trend, Variability, Immediacy of Effect, Overlap across adjacent planes, Consistency of data patterns in similar phases
Visual Analysis of Single Case Designs
Baseline
- Document the “problem” requiring intervention
- Typically ≥ 5 data points
- Documentation of a pattern of responding that allows prediction into the future
Each Phase
- Documentation of a clear pattern of responding
- Typically ≥ 5 data points
Adjacent Phases
- Do data document a “basic effect”
Whole Study
- Do the phases document experimental control (e.g. at least three demonstrations of a basic effect, each at a different point in time (at least 3)
Assessing within phase “pattern” and Between phase “basic effect”
Within Phase “pattern”:
- Level
- Trend
- Variability
Between Phase “basic effect”
- Level
- Trend
- Variability
- Overlap
- Immediacy of Effect
- Consistency across similar phases
- Other: vertical analysis; intercept gap
Basic Effect
A predicted change in the dependent variable when the independent variable is actively manipulated
Level
- The mean of the data within a phase
- Also can be used to assess the level of the last 3-5 data points within a phase
Trend
The slope of the best-fit straight line describing data within a phase
Variability
The level deviation of data around the slope of the vest fit straight line (range, standard deviation)
Overlap
The percentage of data from one phase (typically the intervention phase) that overlaps with the range of data from the previous phase (typically the baseline phase)
Immediacy of Effect
The magnitude of change (in level, trend or variability) between the last 3-5 data points in one phase and the first 3-5 data points in the next phase
Consistency of Data Pattern in Similar Phases
The extent to which phases with similar conditions are associated with data with similar data patterns
Assessing change in Variability
- The greater the variability the more data points are needed within a phase to document a predictable pattern
- Within phase shift in variability requires more data points to establish a stable pattern
- Outliers require more data points within a phase, and a phase should not end with an outlier
Assessing Overlap
Overlap (using range) becomes less relevant with increased trend
- Instead use overlap projected by confidence intervals around best fit straight line
Immediacy of Effect 2
Variability at end of first phase and beginning of second phase reduce the impact of “immediacy of effect”
