Single-Case Experimental Design

Within-Series Designs

The within-series designs are the most commonly used single-case experimental designs and are characterized by the evaluation of data points across time and within phases or conditions (e.g., treatment vs. no-treatment conditions). Specifically, data are gathered over time and grouped into phases, with each phase consisting of a certain number of consecutive data points. Changes are then assessed as they occur across time. Although there are a number of withinseries designs (e.g., simple phase-change, changingcriterion, parametric, and periodic interventions), the simple phase-change design and the changingcriterion design are the most common.

The most basic form of the simple phase-change design is the A/B design in which A and B represent different phases—A represents a baseline phase (no treatment) and B represents an intervention phase (treatment). The simple phase-change design can be extended to include replication (e.g., A/B/A/B) and the investigation of additional interventions (e.g., A/B/A/ C/A or A/B/A/C/B+C). Intervention effects within the simplephase change design are evaluated by investigating patterns of response across time and within phases.

The changing-criterion design is another example of a within-series design. It also typically consists of two phases: a baseline phase and an intervention phase. The intervention phase is characterized by systematic changes in the required level of performance in response to the dependent variable (the variable that is expected to change as a result of the intervention/ independent variable). Intervention effects are evaluated by the dependent variable's response to the preset criteria. As an example of a changing-criterion design, [Page 502]consider a student with selective mutism, an anxietybased disorder characterized by the absence of speaking in particular social situations. The behavior of interest could be the number of times the child speaks to the teacher during class. Whereas the initial criterion could be one time per day, the subsequent criterion would consistently increase according to the student's performance and expected changes. The confidence by which one can draw conclusions about the effects of the intervention can be improved by scheduling a period of time when the criterion is not changed and demonstrating that the behavior did not change during the scheduled criterion. Moreover, the researcher can schedule an intervention withdrawal and replication, a procedure characteristic of the simple phase-change variation of a within-series design.

The parametric and periodic intervention designs are less common forms of the within-series designs. Briefly, the parametric intervention design involves varying one treatment (e.g., by intensity) and monitoring the effects of the variations. The periodic intervention design takes into account that treatments are not always present outside of scheduled treatment sessions. As such, if it is not feasible for a treatment to be implemented consistently in applied settings, the periodic intervention design allows for the implementation of a treatment and the monitoring of effects as frequently as treatment sessions.

...