TataKelola dan KamSiber Kecerdasan Buatan v022.pdf
Research on Single Subject Design
1. SINGLE SUBJECT
RESEARCH
PREPARED FOR:
DR JOHAN @ EDDY LUARAN
PREPARED BY:
AFZA ARRMIZA BINTI RAZIF [2013697158]
HANIFAH BINTI RAMLEE
IZYAN NADHIRAH BINTI WAHID
MAIZATUL AKMAL BINTI ZULKIFLI
2. DEFINITION
• A research of collecting data from a few individual.
For example: researcher who wish to study children who
suffer from multiple disability for example who are both deaf
and blind may have only a small number of children available
to them.
• Number of available is six or less. It would make little sense
to form two groups of three each in such an instance.
• Each child would probably need to be observed in great
detail.
3. SINGLE SUBJECT DESIGN
•
Adaptations of the basic time-series design shown the
previous chapter .
• Data are collected and analyzed for only one subject at a
time.
• Commonly used to study the changes in behavior an
individual exhibits to an exposure to an intervention or
treatment of some sort.
4. CHARACTERISTIC
The Graphing Of Single-Subject Design
• Primarily use line graphs to present their data and to illustrate
the effects of a particular intervention or treatment.
• Vertical axis usually display the dependent (outcome)
variable.
• Horizontal axis used to indicates sequence of time.
5. • Figure 14.1 presents an illustration of such a graph. The
dependent (outcome) variable is displayed on the vertical
axis (the ordinate, or y-axis).
• The horizontal axis (the abscissa, or x-axis) is used to
indicates sequence of time, such as sessions, days, week,
trials or months.
6. The A-B Design
• The basic approach of researchers using an A-B design is to
collect data on the same subject, operating as his or her own
control, under two conditions or phases.
• The first condition is the pretreatment condition called
baseline period is identified as A. The baseline is extremely
important in single-subject research since it is the best
estimate of what would have occurred if the intervention were
not applied.
• Once the baseline condition has been established, a
treatment or intervention condition, identified as B.
7. Figure 14.2 A-B Design
As you can see, five measures were taken before the
intervention and five more during the intervention. Looking at
the data in figure 14.2, the intervention appears to have been
effective. The amount of responsiveness after the
intervention increased markedly.
8. The A-B-A Design
A-B-A design will able researcher to simply adding another
baseline period.
This may improves the design considerably.
9.
10. The A-B-A-B Design
• Two baseline period are combined with two treatment
periods.
• This further strengthens any conclusion about the
effectiveness of the treatment.
• It permits the effectiveness of the treatment to be
demonstrated twice.
• In fact, the second treatment can be extended indefinitely if a
researcher so desires.
11.
12. The B-A-B Design
• In such cases, a B-A-B design may be used which involves
a treatment followed by a baseline and also followed by a
return to the treatment.
• It usually according to times when an individual’s behaviour
is so severe or disturbing.
13.
14. The A-B-C-B Design
• A further modification of the A-B-A design.
• C refers to a variation of the intervention in the B condition.
• In the first two condition, the baseline and intervention data
are collected.
• During the C condition, the intervention is changed to control
for any extra attention the subject may have received during
the B phase.
• The C condition, therefore, might be praise given no matter
how the subject responds.
15.
16. Multiple-Baseline Design
• Alternative for A-B-A-B design
• Used when it is not possible or ethical to withdraw a
treatment and return to the baseline condition.
• When using this design, researchers do more than collect
data on one behavior for one subject in one setting.
• They collect on several behaviors for one subject, obtaining
a baseline for each during the same period of time.
17.
18. Threat to Internal Validity in SingleSubject Research
It involve the :
The length of the baseline and intervention conditions
The Number of Variables Changed When Moving from One
Condition to Another
The Degree and Speed of Change
The Return to Baseline Level
Independence of Behaviour
Number of Baselines
19. CONDITION
LENGTH
Refers to how long the
baseline and intervention
conditions are in effect.
The number of data points
gathered during a conditions.
Minimum three of data point
to establish a clear pattern or
trend.
As a hypothesis,in a certain
period or condition of
length,the researcher need
to gathered enough data as it
will establish the clear
pattern
21. NUMBER OF
VARIABLES
CHANGED WHEN
MOVING FROM ONE
CONDITION TO
ANOTHER
Only one variable should be
changed at a time when
moving from one condition
to another
When analyzing a single –
subject design,it always
important to determined
whether only one variable
at a time has been changed
22. DEGREE AND
SPEED OF
CHANGE
In single-subject
research,the stability is
important
The data change at the
time the intervention
condition is implemented
influenced the stability of
baseline
(when the independent
variable is introduced or
removed)
25. RETURN TO
BASELINE
LEVEL
The subject’s behaviour did
not return to the original
baseline level suggest that
the one or more extraneous
variable may have
produced the effects
observed during the
intervention condition
27. NUMBER OF
BASELINES
In order to have a multiple –
baseline design,researcher
must have at least two
based line
Baseline begin at same
time,the intervention occur
in different time
More baseline and
intervention will lead to
invalid conclusion
28. NUMBER OF
BASELINES
The greater number of
baselines,the greater the
probability that the intervention
is the cause of any change in
behaviour
The more baselines,there
are,the longer the later
behaviours must remain in
baseline
The fewer the number of
baseline,the less likely we can
conclude that is the intervention
rather than some other variable
that causes any change in
behaviour
29.
CONTROL OF
THREATS TO
INTERNAL
VALIDITY IN
SINGLE-SUBJECT
RESEARCH
Single-Subject designs are
most effective in controlling
for :
Subject characteristics
Mortality
Testing
History