“Single subject research (also known as single case experiments) is popular in the fields of special education and counseling. This research design is useful when the researcher is attempting to change the behavior of an individual or a small group of individuals and wishes to document that change. Unlike true experiments where the researcher randomly assigns participants to a control and treatment group, in single subject research the participant serves as both the control and treatment group. The researcher uses line graphs to show the effects of a particular intervention or treatment. An important factor of single subject research is that only one variable is changed at a time. Single subject research designs are “weak when it comes to external validity….Studies involving single-subject designs that show a particular treatment to be effective in changing behavior must rely on replication–across individuals rather than groups–if such results are be found worthy of generalization” (Fraenkel & Wallen, 2006, p. 318).
Suppose a researcher wished to investigate the effect of praise on reducing disruptive behavior over many days. First she would need to establish a baseline of how frequently the disruptions occurred. She would measure how many disruptions occurred each day for several days. In the example below, the target student was disruptive seven times on the first day, six times on the second day, and seven times on the third day. Note how the sequence of time is depicted on the x-axis (horizontal axis) and the dependent variable (outcome variable) is depicted on the y-axis (vertical axis).
Once a baseline of behavior has been established (when a consistent pattern emerges with at least three data points), the intervention begins. The researcher continues to plot the frequency of behavior while implementing the intervention of praise.
In this example, we can see that the frequency of disruptions decreased once praise began. The design in this example is known as an A-B design. The baseline period is referred to as A and the intervention period is identified as B.
Another design is the A-B-A design. An A-B-A design (also known as a reversal design) involves discontinuing the intervention and returning to a nontreatment condition.
Sometimes an individual’s behavior is so severe that the researcher cannot wait to establish a baseline and must begin with an intervention. In this case, a B-A-B design is used. The intervention is implemented immediately (before establishing a baseline). This is followed by a measurement without the intervention and then a repeat of the intervention.
Multiple-Baseline Design
Sometimes, a researcher may be interested in addressing several issues for one student or a single issue for several students. In this case, a multiple-baseline design is used.
“In a multiple baseline across subjects design, the researcher introduces the intervention to different persons at different times. The significance of this is that if a behavior changes only after the intervention is presented, and this behavior change is seen successively in each subject’s data, the effects can more likely be credited to the intervention itself as opposed to other variables. Multiple-baseline designs do not require the intervention to be withdrawn. Instead, each subject’s own data are compared between intervention and nonintervention behaviors, resulting in each subject acting as his or her own control (Kazdin, 1982). An added benefit of this design, and all single-case designs, is the immediacy of the data. Instead of waiting until postintervention to take measures on the behavior, single-case research prescribes continuous data collection and visual monitoring of that data displayed graphically, allowing for immediate instructional decision-making. Students, therefore, do not linger in an intervention that is not working for them, making the graphic display of single-case research combined with differentiated instruction responsive to the needs of students.” (Geisler, Hessler, Gardner, & Lovelace, 2009)
Regardless of the research design, the line graphs used to illustrate the data contain a set of common elements.
There are three basic types of behavioral recordings: frequency recording, duration recording, and interval recording (although many other variations are sometimes used for certain purposes). The recording procedure that you choose will depend on the kind of behavior that you plan to measure. Frequently recording or momentary time sampling recording (a type of durational recording) are most commonly used.
Frequency recording is a simple counting of how many times a behavior occurs during a designated period of time. Those designated periods might be a minute, an hour, a day, or a week. It is most useful with behaviors that are discrete and short in duration. For example, a data collector might record how many times a target subject talked out of turn during a 15-minute class discussion. A researcher trying to reverse a student’s underachievement might record how many homeworks were not submitted over a week.
In momentary time sampling recording, the data collector periodically looks at the target subject at predetermined (NOT spontaneously selected) intervals and records whether a given behavior is occurring. If a researcher were interested in on-task behaviors in a classroom, the data collector might select a 15-minute period and recorded whether the target subject was on or off task at 30 second intervals. At 30 seconds, 1 minute, 90 seconds, 2 minutes, etc. the data collector would mark whether the target subject was on or off tasks and then tally the number of times the target subject was off task over the 15-minute period. The score could range from 0 (never on task) to 60 (off task every time point the data collector checked during the 15 minutes.
With duration recording, you record how long a behavior occurred during predetermined intervals. This is helpful with behaviors that are continuous (last for a given period of time), rather than separate incidents. For example, you might study a dog’s barking behavior. You could count the number of distinct, separate barks (frequency recording) or you could record the length of each barking sequence (duration recording). You might observe the dog for 5-minute intervals. In the first interval, the dog might bark for 20 seconds, stop barking for 2 minutes, bark for 5 seconds, stop barking for 40 seconds, bark for 95 seconds, and then not bark for 20 seconds. If you were frequency recording, you would record 3 for the number of times the dog barked over 5 minutes. If you were duration recording, you would record 115 second for the amount of time the dog barked over 5 minutes.
This material was adapted from material posted at https://www.behavioradvisor.com/BehRecord.html and http://www.specialconnections.ku.edu/?q=assessment/data_based_decision_making/teacher_tools/duration_recording
Del Siegle, Ph.d.
Posted 2/20/2024
I hope this helps you better understand single subject research.
I have created a PowerPoint on Single Subject Research.ppt, which also available below as a video.
I have also created instructions for creating single-subject research design graphs with Excel.
References
Fraenkel, J. R., & Wallen, N. E. (2006). How to design and evaluate research in education (6th ed.). Boston, MA: McGraw Hill.
Geisler, J. L., Hessler, T., Gardner, R., III, & Lovelace, T. S. (2009). Differentiated writing interventions for high-achieving urban African American elementary students. Journal of Advanced Academics, 20, 214–247.
Del Siegle, Ph.D.
University of Connecticut
del.siegle@uconn.edu
www.delsiegle.info
Revised 02/02/2024