Chapter 03 - Graphing Behavior and Measuring Change

Graph

a visual representation of the occurrence of a behavior over time 

Helps visualize data and findings to assess levels of behavior before, during, and after treatment

Y-axis and X-axis, labels for Y and X axis, numbers on Y and X axis, data points, phase lines, phase labels

Six essential components of a behavior modification graph

Y axis label

behavior and dimension of behavior recorded

X axis label

unit of time during which the behavior was recorded

Data points

Plotted points indicating the level of the behavior that occurred at each particular time period

Phase lines

Vertical line on a graph that indicates a change in treatment (no-treatment to treatment, treatment to no-treatment, one treatment to another treatment)

Phase labels

Labels indicating the different phases of a treatment plan

Level

how high or low the data are on the y-axis - observing the _____ of behavior in baseline and treatment phases will confirm whether the treatment intervention produces the desired behavior change

Trend

exists when the behavior is increasing across a phase (increasing trend) or decreasing across the phase (decreasing trend), when no increase/decrease occurs across a phase then we say a behavior is stable

Variability

refers to how high and low the data points are away fom the mean level in the phase

Low variability

data points are all close to the mean level in the phase

High variability

data points are far above and far below the mean

Functional relationship

relationship between a behavior and an environmental event(s) in which the occurence of the behavior is controlled by the occurrence of the environmental event - demonstrated in research by altering treatment (IV) to see if any changes in behavior (DV) are demonstrated

• Target behavior changes when an IV is manipulated while all other variables are held constant 

• IV manipulation is replicated one or more times and the behavior changes each time

Functional relationship is established if

Research design

used to determine whether a treatment (IV) was responsible for the observed change in target behavior (DV) and to rule out any extraneous variables that potentially caused the behavior change

A-B design

simplest type of graph used in behavior modification, with only a baseline phase and treatment phase

• Not a real research design because there is not establishment of a functional relationship due to treatment not being replicated 

  • Possibility for third variable to be the reason for behavioor change 

A-B-A-B reversal design

extension of A-B design in which baseline and treatment phases are implimented twice

A-B-C-A-C design

variation of A-B-A-B design in which the first treatment intervention was not effective, so an alternate treatment is implemented instead

• May not be ethical to remove treatment in second baseline if the behavior is dangerous (such as self-injourous behavior) 

• Must be fairly cerain that the level of behavior will reverse when treatment is withdrawn, otherwise a functional relationship will not be demonstrated

Considerations with A-B-A-B designs

Multiple baseline across subjects design

baseline and treatment phase that is replicated for the same target behavior of two or more subjects

Graphs - Multiple baseline across subjects design

Graphs - Changing criterion design

Graphs - Alternating Treatments Design

Graphs - AB Design

Graphs - ABAB Reversal Design

Multiple baseline across behaviors design

baseline and treatment phase replicated for two or more behaviors in the same subject

Multiple baseline across settings design

baseline and treatment phase is replicated in two or more settings w/ the same behavior in the same subject

• there is interest in the same target behavior exhibited by multiple subjects 

• there’s more than one target behavior of the same subject 

• you are measuring a subjects behavior across two or more settings

Multiple baseline design may be used when

Alternating treatments design (ATD)

Baseline and treatment phases are conducted in rapid succession, typically in alternating days or sessions - baseline and treatment phases can be compared w/ each other within the same time period, there can also be multiple treatment conditions

The larger the discrepancy between the two conditions the stronger proof there is of a functional relationship between behavior and conditions

How is a functional relationship demonstrated in an alternating treatments design?

Changing Criterion Design

sequential goal/criterion levels are speficied for the behavior during treatment; a functional relationship is demonstrated if the behavior meets the criterion goal specified within treatment