Friday, 3 October 2014



Traditionally, single-subject experimental research design has been used most widely in psychology-related fields and in special education, however it is slowly expending into other areas of inquiry. 
According to Neuman and McCormick, single-subject design is an experimental technique where one subject or a small number of subjects is studied intensively. Unlike much traditional group-data analysis, these designs allow for the study of response changes in single individuals. Thus although there may be any number of subjects in an investigation, the designation single-subject means that each subject's behaviors and outcomes are analyzed individually, not averaged with other members of an experimental or control group. In this respect, the method has something in common with case study research. Unlike much case study research, however, single-subject experimental studies allow the researcher to describe cause-and-effect relationships between independent and dependent variables, [such as inexperimental and quasi-experimental design]. In most cases, single-subject experimental studies are conducted in the context in which the behavior is practiced (i.e., classroom), rather than in contrived laboratory settings. Here, the emphasis is on examining the fundamental relationship between an independent variable (the intervention) and a dependent variable (the outcome measure) for a particular individual. Typically, the dependent variable (or variables) focuses on behaviors that are measurable and practically important for students success. Consequently, whether or not the intervention is inferred to be successful is based on its educational (or social) relevance and importance rather than on statistically significant standards.
Although single-subject experiments (also sometimes referred to as N=1 studies) are often confused with experimental and quasi-experimental designs by beginning scholars, they have several important distinguishing factors, the most obvious of which is the lack of a control group. Instead of utilizing a control group to create the hypothetical counterfactual, single-subject experimental research designs rely on baseline logic in order to make causal claims on the effect of a particular intervention. This means that data collected  under no-treatment conditions for a particular individual--referred to as a baseline--is compared with that same individual's performance under treatment conditions.

A.    Definition of Single Subject Experiment
Single Subject Research Designs (also referred to as single-case experimental designs) are designs that can be applied when the sample size is one or when a number of individuals are considered as one group. These designs are typically used to study the behavioral change an individual exhibits as a result of some treatment. In single-subject designs, each participant serves as her or his own control, similar to a time-series design. Basically, the participant is exposed to a non-treatment and a treatment phase and performance is measured during each phase.

B.     Characteristic of Single Subject Experiment
James H. McMillan has summarized five characteristics of single-subject research.
1.    Reliable measurement: Since these designs involve multiple measures of behavior, it is important for the instrumentation to be reliable. Conditions for data collection, such as time of day and location, should be standardized, and observers need to be trained. Consistency in measurement is especially crucial in the transition before and after the treatment.
2.    Repeated measurement: The same behavior is measured over and over again. This step is different from most experiments, in which the dependent variable is measured only once. Repeated measures are needed to obtain a clear pattern or consistency in the behavior over time. They control for the normal variation of behavior that is expected within short time intervals. This aspect of single-subject designs is similar to time series studies, which investigate groups rather than individuals and do not provide for a return to conditions that were present before the treatment was implemented.
3.    Description of conditions: A clear, detailed description of the conditions of measurement and the nature of the treatment is needed to strengthen internal and external validity.
4.    Baseline and treatment conditions: Each single-subject study involves at least one baseline and one treatment condition. The baseline refers to a period of time in which the target behavior (dependent variable) is observed and recorded as it occurs without a special or new intervention. The baseline behavior provides the frame of reference against which future behavior is compared. The term baseline can also refer to a period of time following a treatment in which conditions match what was present in the original baseline. The treatment condition is a period of time during which the experimental manipulation is introduced and the target behavior continues to be observed and recorded. Both the baseline and treatment phases of the study need to be long enough to achieve stability in the target behavior.
5.    Single-variable rule: During a single-subject study, only one variable should be changed from baseline to treatment conditions. In some studies two variables are changed together during the same treatment condition. This is an interaction in single-subject research.

C.    Types of Single Subject Research
There are three commonly accepted types of single subject research designs.

1.      A-B-A-B Withdrawal (Reversal) Designs

The A-B-A-B withdrawal procedure uses a baseline (control) phase referred to as the A of the experiment and an intervention (treatment) phase of the experiment known as the B. In the A-B-A design, the baseline is required to establish the student’s pre-intervention performance level. This non-intervention period is initiated until the behavior in question demonstrates stability. The intervention phase B of the study is initiated and data continues to be collected. In the A-B-A design, a third phase is subsequently instituted in which the experimental intervention is withdrawn with a return to the baseline (control) state A. Finally, the B phase is reinstated to demonstrate that the effects are a function of the intervention and not some other variable. This is done to determine if a causal relationship exists between the intervention (B) and improved student performance. Demonstrating the effect across additional participants further strengthens the causal relationship.
Example: A-B-A-B Withdrawal Design
A teacher is working to improve a student’s ability to remain in seat during work periods uses an A-B-A-B Design.
  • Phase A: The teacher records the number of times per day Bob is out of his seat during work periods.
  • Phase B: The teacher implements reinforcement program for student remaining in seat for a week and counts the frequency of out of seat behavior.
  • Phase A: The teacher suspends the reinforcement program and the records the number of times Bob is out of his seat.
  • Phase B: The teacher re-implements reinforcement program for student remaining in seat and counts the frequency of out of seat behavior.
In this experiment the teacher is able to show the reinforcement program is effective in meeting her goals for Bob.

2.      Multiple-Baseline Designs

The first example looked at the effects of a single targeted student behavior, out of seat, in a single setting.
The second example, a multiple-baseline design, is useful for studying the effects of a teaching practice in which two or more behaviors, people, or settings can be tracked on a single experiment. The multiple-baseline can examine one of these three variables (subject, behavior, or setting) while keeping the other two variables constant. In the A-B-A-B design all three parameters were kept constant, a single behavior for a single subject in a single setting. What if a teacher wants to know the impact of fluency training for three students in her/his classroom? The multiple-baseline design is well suited to answering this type of question. Another advantage of the multiple baseline design is it is not necessary to return a participant to baseline condition.  The multiple baseline design can demonstrate causal relation without having to terminate intervention.  In the case of academic skills this offers the teacher an important tool as it is often the case that once the skill have been learned then a return to baseline will not produce a change in performance.
Example: Multiple-Baseline:
A pre-intervention reading fluency baseline is established for all three students (Bill, Gloria, and Bob). The data is usually collected concurrently across all three participants.  After three to five trials and assuming a stable baseline is established, Bill is introduced to the experimental condition, a reading fluency  program. The program is introduced while the other two subjects continue with the baseline condition.
After three to five more trials, Gloria goes into the experimental phase. The third student, Bob, continues with the baseline condition.  After three to five more days, Bob starts the experimental phase while the other two participants continue in the intervention phase.
In this example, the dependent variable is the number of words read correctly during a one-minute time period, and the independent variable is the three students. 
This design allows the teacher to know that the intervention is effective for Bill and Gloria but is not working for Bob.

3.      Alternating Treatment Design

The alternating treatment design is used to compare the effects of two treatments on one subject. Two interventions or practices are introduced and alternated at each of the training sessions. The teacher determines randomly which of the two experimental interventions to select for each session. This design offers the teacher a method for comparison of the two interventions is better suited to improving the student’s performance. 
Example: Alternating Treatment Design
A teacher is interested in increasing her student’s rate of reading. She has two fluency programs that she believes may work with Bob. She implements the two different programs on alternate days over 19 days. After the completion of the experiment, she is confident that intervention #1 will achieve better results for improving Bob’s reading skills.

D.    Control Strategies in Single-subject Research
  • A-B Design (Stable Baseline)
  • A-B-A Design
    • Baseline, treatment, withdrawl
    • Problems:
      • Treatment may not be reversible
      • May not be ethical to leave the subject in the untreated condition
  • A-B-A-B Design (Repeated Treatments Design)
  • A-B-A-B-BC-B-BC (Interaction Design)
    • For examining the effects of 2 variables
    • Obeys the cardinal rule of single-subject designs:  Change only one thing at a time!
  • Multiple Baselines Design
    • Introduce the treatment at different times for different behaviors
    • Really has multiple DVs and related IV's; example:
      • IV1 = reward behavior A
      • DV1 = frequency of behavior A
      • IV2 = reward behavior B
      • DV2 = frequency of behavior B
  • Changing Criterion - changing the criterion for measuring a change in behavior as learning takes place (because some learning is irreversible)

                       1.     Reliable measurement of the target behavior
                       2.     Target behavior is clearly defined in operational terms
                       3.     Sufficient measurements are made during each time frame to establish stability
                       4.     Full descriptions of the procedures, subjects, and settings are provided
                       5.     Use of one (1) standard treatment
                       6.     Control of experimenter and/or observer effects
                       7.     Results should have practical significance

F.     Validity and Reliability in Single Subject Experiment
1.    Internal validity = The validity of inferences about cause-effect relationships is dependent on the degree to which the researcher demonstrates that:
·         A treatment preceded an observed effect.
·         The treatment is related to the effect.
·         There are no other plausible explanations for the effect. 
Single-subject experimental designs are considered to be strong with respect to this type of validity. This is due to their use of continuous measurement of responses over time, subjects as their own controls, and dual reliability assessments.
2.      Construct validity = Constructs are category labels for theoretical entities. Construct inferences are judgments about the degree to which a sampling component is, in fact, representative of the actual construct it seeks to address. It is important, then, that operational definitions for each dependent variable are precise, exclusive and clear. Moreover, it in imperative that measurement/data collection tools reflect the dependent variable(s) researchers are trying to measure.
3.      Reliability of data collection and analysis procedures = Given that "standardized assessments are often not used, reliability of dependent measures is crucial in these studies. Reliability requires that the definition of a dependent variable is targeted and specific to insure appropriate and concise coding of what may constitute a response. Ambiguity cannot exist".
4.      External validity = Is concerned with inferences about the degree to which experimental effects can be generalized across variations in units, treatments, assessment instruments, and settings. The ability to generalize in single-subject experimental research designs is "directly related to the number of replications performed and to the specificity of methodology in the original study"(Neuman & McCormick, 2000, p. 193).
5.      Ecological Validity = Single-subject experiments are highly compatible with authentic classroom instruction, as they can be conducted in the context of instruction, are often targeted to individual learners needs, and can provide answers to instructional questions in a relatively short period of time. Furthermore, as with formative experiments, they often make some attempt to bridge research and practice.  

G.     Advantages and Disadvantages of Single Subject Experiment
1.    Advantages:
·         Group means could conceal patterns that appear in individuals' data
·         Big effects - only clinically significant effects are likely to be found
·         Ethical and practical advantages (eg; can not withhold treatment; too few subjects)
·         Flexibility
2.    Disadvantages:
·         Can not examine any between-subject effects
·         Can not detect small effects
·         May be less generalizable

Single Subject Research Designs (also referred to as single-case experimental designs) are designs that can be applied when the sample size is one or when a number of individuals are considered as one group.
There are five characteristics of single-subject research:
v  Reliable measurement
v  Repeated measurement
v  Description of conditions
v  Baseline and treatment conditions
v  Single-variable rule
There are three commonly types of single subject research designs:
v  A-B-A-B Withdrawal (Reversal) Designs
v  Multiple-Baseline Designs
v  Alternating Treatments Design
Control Strategies in Single-subject Research consists of:
v  A-B Design (Stable Baseline)
v  A-B-A Design
v  A-B-A-B Design (Repeated Treatments Design)
v  A-B-A-B-BC-B-BC (Interaction Design)
v  Multiple Baselines Design
v  Changing Criterion 
The advantages of Single Subject Experiment:
v  Group means could conceal patterns that appear in individuals' data
v  Big effects - only clinically significant effects are likely to be found
v  Ethical and practical advantages (eg; can not withhold treatment; too few subjects)
v  Flexibility
The disadvantages of Single Suject Experiment:
v  Can not examine any between-subject effects
v  Can not detect small effects
v  May be less generalizable



Post a Comment