How do you know it worked

1
How do you know it worked

• Single Subject Designs in Teaching

2
How do we know if our teaching is successful or
change is due to chance?

• Functional Relationship a cause and effect. The
  target behavior changes as a result of the
  intervention
• Functional Relationship exists between the two
  variables when the interventions have been
  systematically replicated on or more times

3

• Variable any number of factors involved in
  research. (factors related to participants,
  conditions, interventions)
• GOAL to control for the presence of absence of
  variables that may effect the outcomes

4
Variables

• Independent intervention being used
• Dependent behavior targeted for change
• Confounding Those variables in the environment
  that are not controlled but may influence the
  dependent variable

5
Components of a SSD

• Baseline Measures
• A measure of the behavior under the conditions
  that currently exist.
• Provide a measure of the behavior if no
  intervention occurs.

6
Components of a SSD

• Baseline Measures
• Why do we want a baseline to be as stable as
  possible?
• What are two measures of stability?
• Variability
• Trend

7
What to consider when trying to intervene?

• Too much variability makes it difficult to draw
  conclusions
• Good operational definition of the dependent
  variable
• Naturally occurring variability

8

• Trends in the data points
• No trend
• Ascending trend
• Descending trend

9
Components of a SSD

• Intervention Measures
• Repeated measures of the behavior under treatment
  conditions
• Experimental Control insures that changes in the
  behavior are in fact due to the intervention and
  not other confounding variablesa functional
  relationship exists

10
Teaching designs

• A functional relationship is not established
  (lack of experimental control)
• Less confident assumptions can be drawn
• Provide sufficient indication of behavior change

11
Research Designs

• Allows for experimental control and the existence
  of a functional relationship

12
AB Designs
Advantages Disadvantages
Simple to use
Compares student performance Cannot make confident assumptions
Good to graph acquisition obj.

• Referred to as the Teaching design
• Consists of two phases
• Data collected during intervention are compared
  to those collected during baseline

13
(No Transcript)
14
Reversal Designs

• Used to study the effectiveness of a single
  intervention (independent variable)
• Consists of 4 phases
• Should not be used
• When dependent variable is dangerous
• When dependent variable is not reversible

What problems does this pose?
Advantages Disadvantages
Experimental control Must withdraw intervention
Simple
15
Reversal Designs

• Repeatedly compares baseline data to intervention
  data
• Dependent on the replication of baseline and
  intervention effects
• Confounding variables?

16
Changing Criterion Design
Advantages Disadvantages
Functional relationship without withdrawing intervention Must be used on behaviors that require small incremental changes
Progress oriented
Starting with small change

• Evaluates the one independent variable on one
  dependent variable
• Experimental control is demonstrated by
  incrementally increasing or decreasing the
  dependent variable
• Consists of two phases

17
Changing Criterion Design

• Implementation
• Collect baseline data
• Determine interim criterion for performance
• Mean of the stable portion of baseline
• Half the mean of the baseline
• Highest or lowest baseline
• Professional estimate

18
Changing Criterion Design

• Demonstrating Functional Relationship
• Alter the number of sessions
• Continue with a sub-phase until a stable rate
• Vary the increase
• Require a change in the opposite direction

19
Multiple Baseline Designs

• Analysis of 1 independent variable on more than 1
  dependent variables
• Across behaviors
• Across settings
• Across individuals
• Consists of 2 phases

20
Multiple Baseline Designs
Advantages Disadvantages
Establish a functional relationship (without reversing the intervention Must implement across settings, people, etc. -- cumbersome
Allows the examination of one intervention on many kids Collect baseline over extended period of time

• Cannot be used with a behavior that calls for
  immediate action
• When behaviors are not independent

21
Multiple Baseline Designs

• Implementation
• Baseline is collected on all conditions at the
  same time
• Begin intervention in first condition when stable
  baseline is reached
• Begin intervention in second condition when
  change has occurred in the first condition

22
Multiple Baseline Designs

• Extended Baselines
• Not appropriate for some behaviors
• Kids may learn error response
• Kids may become frustrated
• No instruction being delivered

23
Alternating Treatments Designs

• Allows the comparison of the effectiveness of
  more than one intervention on a single dependent
  variable

Advantages Disadvantages
Efficient for teachers Must implement a replication phase for functional relationship
Multiple treatment spillover
24
Alternating Treatments Designs

• Implementation
• Each condition equal number of times
• Schedule of interventions should be
  counterbalanced (to avoid order effects)
• Distinctive discriminative stimulus should
  immediately precede the condition

25
Changing Condition Design

• Implementation
• Interventions are introduced sequentially.
• Functional relationship only if a return to
  baseline occurs before C condition

26
Advantages Disadvantages
Compare interventions No functional relationship can be established

• Used to study the effectiveness of two or more
  treatments on the behavior of a student. ABC
  design

Cummulative effects
27
Analysis of Results

• Visual Inspection
• Mean of data points
• Levels of performance
• Trend in performance