Professional Documents
Culture Documents
Naurah Nazhifah
Table of contents
01 Introduction to Single-subject Research
02 Research Design
03 Procedure
04 Journal Analysis
Single-subject Research
1. Size limited to one or a small group
A phase is a series of observations made under the same conditions. In the terminology of
single-subject research, observations made in the absence of treatment are called baseline
observations, and a series of baseline observations is called a baseline phase (A). Similarly,
observations made during treatment are called treatment observations, and a series of
treatment observations is called a treatment phase (B). The purpose of a phase within a
single-subject experiment is to establish a clear picture of the participant’s behavior under the
specific conditions that define the phase.
Before doing the treatment
1. Reliability through multiple measurements
This letter sequence indicates that the researcher first made a series of baseline observations, and
then implemented a treatment (B) while continuing to make observations. Next, the researcher
tried a modification of the treatment (perhaps treatment B was not effective), followed by
withdrawal of all treatment (back to baseline). Then, the original treatment (B) was administered
in combination with a new treatment (C), and finally, treatment C was administered by itself.
Phase [Level]
One way to define a pattern within a phase is in terms of the level
of behavior. The term level simply refers to the magnitude of the
participant’s responses. If all of the observations within a phase
indicate approximately the same magnitude, or level, of behavior,
then the data have demonstrated a consistent or stable level of
behavior within the phase.
Phase [Trend]
An alternative way to define a pattern within a phase is in
terms of a trend. The term trend refers to a consistent
increase (or a consistent decrease) in the magnitude of
behavior across the series of observations that make up
the phase.
Dealing with unstable data
1. wait and keep making observations
As the letters indicate, the ABAB design consists of four phases: a baseline phase (A), followed by
treatment (B), then a return to baseline (A), and finally a repetition of the treatment phase (B). The
goal of the ABAB design is to demonstrate that the treatment causes a change in behavior by showing
that:
1. The pattern of behavior in each treatment phase is clearly different from the pattern in each baseline
phase. This demonstration is necessary to establish a relationship between the treatment and the
behavior.
2. The changes in behavior from baseline to treatment and from treatment to baseline are the same for
each of the phase-change points in the experiment. This demonstration is necessary to establish a
causal relationship between treatment and behavior.
The ABAB
design
The first phase change (baseline to treatment)
shows a clear change in the pattern of behavior.
The second phase change (treatment to baseline)
shows the participant’s behavior returning to the
same level observed during the initial baseline
phase. The final phase change (baseline back to
treatment) shows the same treatment effect that was
observed in the initial phase change.
The ABAB design
The graph shows results from an ABAB study in which the treatment clearly is
not effective.
The limitation of ABAB design
The credibility of this causal interpretation depends in large part on the reversal (return to baseline)
that is a component of the design. Therefore, the ABAB design is not appropriate for evaluating
treatments that are expected to have a permanent or long-lasting effect. Another concern is
withdrawing treatment in the middle of the experiment can create some practical and ethical problems
that can limit the application and success of this specific design.
Variations on the ABAB design
Multiple-baseline design
The multiple-baseline design provides an alternative technique that eliminates the need for a return to
baseline and therefore is particularly well suited for evaluating treatments with long-lasting or
permanent effects. A multiple-baseline design requires only one phase change—from baseline to
treatment—and establishes the credibility of the treatment effect by replicating the phase change for a
second participant or for a second behavior.
The key to this design is that the different behaviors are independent (one does not influence another)
and can be treated separately by focusing a treatment on one behavior at a time.
Multiple-baseline design