presentation edu 702
TRANSCRIPT
QUANTITATIVE RESEARCH METHODOLOGIES
SINGLE-SUBJECT RESEARCH
ESSENTIAL CHARACTERISTICS
Data are collected and analyzed for only one subject at one time.
Most commonly used to study the changes in behavior
THE GRAPHING OF SINGLE-SUBJECT DESIGNS
Researchers primarily use line graphs to; Present their data Illustrate the effects of a particular intervention or treatment.
THE A-B DESIGN
Collecting data on the same subject, operating as his or her own control under two conditions or phases.
First condition (A); pretreatment condition
Second condition (B); treatment or intervention condition, introduced and maintained for a period of time.
figure 14.2 A-B Design.jpg
DISADVANTAGE OF A-B DESIGN
Researcher does not know whether any behavior change occurred because of the treatment.
THE A-B-A DESIGN
Sometimes called reversal design Researchers simply add another
baseline period To get stronger evidence for the
effectiveness of the intervention Reduce threats to internal validity
figure 14.3 A-B-A Design.jpg
DISADVANTAGE OF A-B-A DESIGN
It involves leaving the subjects in the A condition.
Studies ended without some degree of final improvement.
THE A-B-A-B DESIGN
Two baseline periods are combined with two treatment periods.
Permits the effectiveness of the treatment. Avoids leaving subject without an
intervention. Demonstrate that dependent variable
changes as independent variable is applied.
figure 14.4 A-B-A-B Design.jpg
DISADVANTAGES OF A-B-A-B DESIGN
The possibility of data-collector bias
An instrumentation effect; need for an extensive number of data collections period
THE B-A-B DESIGN
Involves a treatment followed by a baseline by a return to the treatment.
Appropriate when there is a lack of behavior.
figure 14.5 B-A-B Design.jpg
THE A-B-C-B DESIGN
Further modification of the A-B-A design.
The intervention is changed to control for any extra attention the subject may have received.
Contingent (or selective) praise is critical for improved responsiveness.
figure 14.6 A-B-C-B Design.jpg
MULTIPLE-BASELINE DESIGNS
Researcher do more than collect data on one behavior.
They collect on several behaviors for one subject.
Researchers systematically apply the treatment at different time for each behavior.
figure 14.8 Multiple-Baseline Design.jpg Treatment is applied to change behaviors
accordingly.
THREATS TO INTERNAL VALIDITY
1. Condition length
2. Number of variables changed when
moving from one condition to another.
3. Degree and speed of change.
4. Return to baseline level
5. Independence of behavior
6. Number of baseline
THINGS TO PONDER
Which do you think is easier to conduct: single subject or group comparison research? Why?
What sorts of behaviors might require only a few data points to establish a baseline?