Chapter 3 – Graphing Behaviour and Measuring Change
Graph: a visual representation of the occurrence of a behaviour over time.
Using a graph allows them to document changes in behaviour during treatment
and make decisions about the continued use of the treatment.
Components of a Graph
Time is indicated on the horizontal axis (x-axis or the abscissa) and the behaviour is
indicated on the vertical axis (y-axis or the ordinate).
Six components are necessary for a graph:
The y-axis and x-axis
Labels for y-axis and x-axis
The numbers on the axis.
Phase lines: vertical line on a graph that indicates a change in treatment
The purpose of research designs is to determine whether the treatment (independent
variable) was responsible for the observed change (dependent variable) and to rule out
the possibility of a third variable causing the variable to change.
Involves both treatment implementation and replication
Functional relationship: when it can be shown that a behaviour modification procedure
causes a target to change, it shows that there is a relationship between the procedure
and the target behaviour.
This is established if (a) the target behaviour changes when an independent
variable is manipulated (a procedure is implemented) while all other variables
held constant and (b) the process is replicated or repeated more times and the
behaviour changes each time.
A = baseline B = treatment
Comparing baseline and treatment to determine whether the behaviour changed in the
expected way after treatment. This type of graph cannot show a functional relationship because the treatment is
not replicated. This means it’s not a true research design. It does not rule out the
third variable option.
Because of this, it is rarely used by BMod researchers. It is most often used in
applied, nonresearch situations.
A-B-A-B Reversal Design
This is an extension of the A-B design. In this case, the baseline and treatment phases
are implemented twice. Called a ‘reversal design’ because it reverses back to baseline
after the first treatment and then replicates.
If that first treatment does not work, the graph may change to an A-B-C-A-C
design, with C being a different treatment.
Some issues must be considered before doing a reversal design:
Is it ethical to remove the treatment? It may not be if it’s treating for a dangerous
behaviour (self-injurious behaviour).
You must be fairly certain that the behaviour will reverse when you return to