National Center for Technology Innovation
 

Single Subject Research


Single subject research is a study which aims to examine whether an intervention has the intended effect on an individual, or on many individuals viewed as one group. The two most common single subject research designs are the A-B-A-B design, and multiple baseline design. Each of these designs has two main components: (1) a focus on the individual and (2) a design in which each individual is used as his or her own control observation. The focus on the individual differs from other research designs, such as experimental and quasi-experimental designs, which look at the average effect of an intervention within or between groups of people. In single subject research, researchers often use more than one individual, but results are examined by using each individual as his or her own control, rather than averaging results of different groups. Comparisons are made on the behavior of one individual to that same individual at a different point in time.

Single subject research has an important role to play in identifying and documenting solutions for individuals with disabilities. The field needs much more evidence on what works for whom, under what conditions, for which tasks, etc. Although individuals with disabilities—even those with the same diagnosis—often experience unique needs, solutions may be adaptable in different environments, and knowledge sharing can inform others working on assistive solutions.

  • A-B-A-B design
  • The A-B-A-B design is a time series design, where A is a baseline observation and B is an observation using the intervention. A baseline observation in single subject research is defined as an observation where the intervention is not present – a “business as usual” time period. Because this research design has a baseline observation after a treatment observation, this type of research design is not applicable to programs or interventions where a subject is taught a skill, since that would require unlearning the intervention. The A-B-A-B design is meant to test the effectiveness of specific methods, tools, or technologies. Furthermore, the outcome the researcher is trying to impact, the dependent variable, must be quantifiable. By having the outcome measure quantifiable, researchers are able to graph and analyze the data experimentally.

    A communication device is an example of a tool that can be tested in the A-B-A-B design. A subject without the device could attempt to communicate to a researcher or caregiver, which would be the first observation, A. Then this same individual could attempt to use the device for communication, B. This same process repeated again would give two baseline and two treatment observations. Of course, the individual and the caregiver would need to be trained to use the device or software, but the absence of it when trying to achieve the same task constitutes the business as usual condition.

    Chart showing number of events 0 – 20 on the Y axis and Time on the X axis. The chart is divided vertically into four sections labeled Baseline, Treatment, Baseline, Treatment.

  • Multiple baseline design
  • Because single subject designs focus on studying individuals rather than groups, they can be particularly vulnerable to threats to internal validity. Internal validity addresses how valid it is to make causal inferences about the intervention in the study. For more, see section on validity.

    Particular internal validity threats in the A-B-A-B single subject research design are maturation (the natural growth in the study participant’s ability over time) and test-retest (a study participant doing better on each administration of a test due to their experience taking the test). The multiple baseline design helps to control for these threats to internal validity by having a study participant give multiple baseline observations before using the intervention. Further, if multiple individuals are tested with the treatment given at different time points for different individuals, researchers can have a better understanding of whether or not the treatment is effective. Unlike A-B-A-B single subject research designs, multiple baseline single subject research studies can be used to study programs or interventions where a subject is taught a skill.

    Chart showing number of events 0 – 20 on the Y axis and Time on the X axis. The chart is divided vertically into five sections; the first three from the left are labeled Baseline, the last two are labeled Treatment.

  • What are the benefits of conducting single subject research?
  • Because single subject research uses each participant as his/her own control subject, researchers can get a better understanding of individual differences rather than the difference of the average between groups. For this reason, single subject research is often considered the best research design when measuring behavioral change. When done correctly and carefully, single subject research can show a causal effect between the intervention and the outcome.

    The flexibility, simplicity, and low cost of this study design are also beneficial. While it is important to carefully plan any analyses, the design of single subject research is far more flexible, as you may change the order of the delivery of baseline or treatment observations depending on the individual participant responses. Single subject research can also be much easier to plan, since it can be much smaller in scale than forms of experimental and quasi-experimental research. Because of this, it may also be much cheaper than other research designs.

    All studies attempt to maximize both internal and external validity. Internal validity addresses how valid it is make causal inferences about the intervention in the study. External validity addresses how generalizable those inferences are to the general population. Single subject research may produce results that have strong internal validity, when all internal validity threats are addressed. However, due to the small number of study participants, single subject research tends to have poor external validity, limiting the ability to generalize the findings to a wider audience (For more, see section on validity).

  • When should I conduct single subject research?
  • Single subject research studies attempt to examine the effects of an intervention on an individual. The general form of a research question that a single subject study can answer is “What is the effect of [specific program/intervention] on [specific individuals]?” Many other forms of research may be more appropriate for your needs, depending on your research question. For example, if one asked “How useful is x?” or “What is the market demand of y?”, single subject research would not be helpful.

    The main limitations of conducting single subject research are issues concerning validity. While the A-B-A-B study design does have some issues with internal validity (mentioned above), the main limitation of single subject research is external validity. The limitation of not being able to generalize from single subject research is a challenge to AT developers. Individuals with disabilities often present unique challenges that are met with unique solutions. Testing the efficacy of these solutions is important, but they may not be applicable to other individuals. Nevertheless, rich description of the need and the solution, in addition to the analyses of the intervention, may be helpful to others in the field.

  • What are the resources needed to conduct single subject research?
  • While conducting single subject research is much easier than conducting quasi-experimental or experimental research, it can still be difficult to implement. It is necessary to understand all of the possible threats to the study’s validity, as well as the statistical methods needed to run accurate analyses. It is recommended that organizations use outside consultants or research organizations to run single subject research. Not only does this allow the single subject research to be conducted by experts in study design, implementation, and analysis, but it also protects the results of the study from a perceived bias of the organization. For example, while a company with a product that is for children with learning disabilities might conduct a valid and bias-free study that shows that their product does in fact improve literacy in children with learning disabilities, school districts might not trust this result unless it was conducted and analyzed by an unaffiliated organization.

    Single subject research tends to take less time and money to conduct than other forms of experimental research. This is largely due to (1) the nature of the intervention, testing specific methods, tools, or technologies, rather than programs or interventions, and (2) the small number of study participants.

    Elements of single subject research

  • Study participant recruitment
  • First, it is important that study participants are members of the population in which you hope your intervention will be effective. Second, study participants must agree to be in the study. Often, this involves getting parent/guardian approval by signing consent forms which describe the study and any risks and benefits that study participants may be exposed to. Due to the small number of study participants needed to perform single subject research, this process is far easier than with other forms of research. However, just getting individuals to agree to participate is not enough. Any studies that are funded by government agencies must have the study, data collection items, and even the consent forms approved by an Institutional Review Board before study participant recruitment begins. For more, see section on Institutional Review Boards (IRB).

  • What sorts of data should I collect?
  • In order to analyze data in single subject research, it is important that the data you collect—usually measures of a behavior—be quantitative in nature. Some examples of quantitative data include test scores and observations (how many times each person did x).

  • How should I analyze this data?
  • The main form of analysis in single subject research is graphic analysis. In graphic analysis, you graph the data to see how it changes over time in both baseline and treatment conditions. For examples of graph templates, see A-B-A-B design, and Multiple baseline design.

    For more resources on graphic analysis and single subject design see:

    Wikipedia: Single-subject Research

    Center for the Study of Ethics in the Professions: Single Subject Research Design

    Examples and additional resources

  • Real world example
  • Instant Messaging for AAC Users: Cool Heads and Collegiality Promote Assistive Technology Innovation. This NCTI 2008 Tech in the Works research award found unexpected discoveries when researchers turned an AAC platform into an instant messaging device, such as the critical need for chat specific vocabularies. The researcher and developer worked with individual AAC users in a single subject design that informed further technology development.

    Final report of the study: http://www.nationaltechcenter.org/documents/point_and_chat_final_report.pdf.

    Scaffolding Early Math Concepts with Stages Math: Number Sense. This 2009 NCTI Tech in the Works research award studied how regular classrooms could meet the needs of diverse young learners by using Stages Math.  The study utilized a single subject design and principles of universal design for learning.

    See the final report (coming soon).

  • Published articles using single subject research
  • Autism Spectrum Disorders & Augmentative and Alternative Communication. (January 2010). reSearch: A collection of research reviews on rehabilitation topics from NARIC and other information resources,5(1).

    The entire collection focuses on Autism Spectrum Disorders & Augmentative and Alternative Communication research, a field that relies on single subject design to personalize and customize devices for individual needs.