Single-Subject Research Design

Description

Single-subject research (SSR) methodology is commonly associated with behavioral psychology, specifically applied behavior analysis, and is a research methodology [Page 1521]and intervention evaluation approach that is used by general and special educators, speech and language therapists, social workers, therapeutic recreation practitioners, clinical psychologists, and occupational therapists. Although used in basic and applied research, SSR methodology use in education and clinical practice has increased considerably in recent years as a means for objectively evaluating the effects of educational and treatment programs. In this age of accountability, SSR methodology is appropriate for use by professionals interested in the formative evaluation of their interventions.

SSR design, also referred to as single-case experimental design, N of 1 design, single N design, and quasiexperimental design, refers to a quantitative research methodology in which each participant serves as his or her own control. That is, participants in SSR design investigations have their behavior (dependent variable) repeatedly measured across days (or sessions) under at least two adjacent experimental conditions: baseline and intervention. After a stable level of performance is established during the baseline condition, the intervention condition (independent variable) is introduced. A “within-subject” analysis of the data is conducted, in which a participant's performance under the intervention condition is compared to his or her performance under the baseline condition. This strategy of directly and repeatedly measuring each participant's behavior under two clearly described adjacent conditions is known as baseline logic and is the cornerstone and underlying premise of all SSR designs. This research approach is not to be confused with a traditional casestudy approach, in which informal observations are made, anecdotal information is gathered, and conclusions are drawn. A case-study approach does not permit ruling out alternative explanations. Unlike case-study methodology, SSR methodology strives for experimental control by collecting data within the context of an experimental design that evaluates and attempts to control threats to internal validity.

The question of internal validity (reliability of effect) is of paramount importance to scientific research: “Is the independent variable and only the independent variable responsible for observed changes in the dependent variable?” The extent to which researchers address threats to internal validity will determine the level of confidence one has that alternative explanations are unlikely. Specifically, researchers must attempt to control for the influence outside or extraexperimental events may have on the outcome of the study (history threat), as well as changes that may occur simply through the passage of time (maturation threat). Repeated measurement of a behavior also may influence the behavior under study (testing threat), as may participants' knowledge that they are subjects under investigation (Hawthorne effect). Changes in the measurement system during the course of an investigation (instrumentation threat) will jeopardize experimental control, as may data variability (instability threat) or the introduction of the intervention prior to the establishment of a stable baseline level and trend. If experimental procedures are not implemented as purported (procedural threat), questions will arise as to what actually was studied. Precautions also must be taken to reduce carryover effect, or the influence that procedures used in one condition may have on participants' behavior in an adjacent condition (multiple treatment interference threat). Although applied research methodologies cannot control for all possible sources of confounding, researchers must attempt to limit the number of alternative explanations by careful selection of a measurement system and research design.

...