Dr. Jennifer Ledford is an assistant professor of special education at Vanderbilt University in Nashville, Tennessee. Dr. Ledford, a member of Council for Exceptional Children Division for Research (CEC DR), co-chaired the recent policy and position statement and user guide focused on single-case design. Wendy Rodgers, Tim Lewis, Sam Odom, Rob O’Neil, Ilene Schwartz, Kimberly Vannest and Katie Zimmerman also assisted with the development of the position statement and guide. Learn more about CEC DR and view the Policy and Position Statement and User Guide .

NCII: What is single-case design, and what are the essential elements of single-case design research?

Ledford: Single-case designs are experimental research designs that are appropriate for assessing the impact of one or more interventions on the behaviors (e.g., academic, adaptive, social, problematic) of individuals.

“Single case” can be a confusing term—some people confuse it with the term “case study” or they assume that only one participant can be involved. But, single case means that each “case” (or student) serves as their own control. In these designs, behaviors are measured repeatedly over time, under at least two different conditions, for the same participants. We draw conclusions about the extent to which data are consistently different across the two conditions, for each participant. So, the data collected during a baseline (no intervention) condition for one student serve as the comparison for data collected during that student’s intervention condition. We call this baseline logic, and all single-case designs are built around this concept.

Of course, just one comparison between a baseline and an intervention condition (what we would call an A-B design) would not be very compelling—anything within or outside of the study could cause a behavior change that occurred only once. Instead, there are established rules about ordering conditions to reduce the likelihood that factors external to the intervention (called threats to internal validity) have an impact on outcomes. In the User Guide, we outline condition ordering and important controls for threats to internal validity. We identify several critical factors, including a sufficient number of replications and data points, as well as adequate reliability and fidelity data. The User Guide also includes graphical examples of common single-case designs.

NCII: Why did the Council for Exceptional Children’s (CEC’s) Division for Research pursue a policy statement on single case designs?

Ledford: Single-case designs are widely used to assess interventions for individuals with disabilities, and they are appropriate for establishing causal relations. Because they are both useful and practical research designs, it seemed important for the Division for Research of the CEC to establish a Policy and Position Statement to confirm CEC’s advocacy for the appropriate use of single-case designs to help establish evidence-based practices for individuals with disabilities. We hope that the Policy and Position Statement and the User Guide can be used by practitioners and educational leaders to guide and inform their use of practices that have been evaluated using single-case research design.

NCII: Why is single-case design research important, particularly for students who require intensive intervention?

Ledford: As I mentioned before, single-case designs are widely used in special education and related fields. They are especially critical in a few areas—for example, for populations that are low incidence or very heterogeneous. They are also critically important in terms of intensive interventions because they are extremely well-suited for answering questions about individualized intervention procedures. So, while studies using group comparison paradigms (such as randomized controlled trials) are valuable for determining average effectiveness for a population, single-case design studies answer the question of whether a particular intervention is effective for each participating individual. Moreover, if a specific intervention is not effective for improving behavior, then the procedural plan can be modified or changed, and you can experimentally evaluate the second or modified intervention compared with the originally planned one. As you might expect, these characteristics result in single-case designs being well-aligned with the goals of intensive intervention—that is, we strive to identify effective, individualized interventions that will result in behavioral improvements for each student, making data-based modifications as needed.