The percentage of non-overlapping data (PND) is a statistical metric used in single-subject research to summarize the effectiveness of an intervention. It is calculated by identifying the percentage of responses in a treatment condition that are more extreme than the single most extreme response recorded in the corresponding control or baseline condition. For example, in a study by Hall and his colleagues, all measurements of Robbie's study time during the first treatment condition were greater than the single highest measurement recorded during his initial baseline phase. This resulted in a PND of $$100\%$$, indicating a very strong treatment effect. Generally, a higher percentage of non-overlapping data signifies a more robust treatment effect.

Percentage of Non-Overlapping Data

The percentage of non-overlapping data (PND) is a statistical metric used to summarize data in single-subject research. It is defined as the percentage of responses in the treatment condition that are more extreme than the most extreme response in a relevant control condition. A greater percentage of non-overlapping data indicates a stronger treatment effect.

In single-subject research, how are formal statistical procedures, such as the t test or analysis of variance, typically utilized?

To apply formal statistical procedures in single-subject research, a researcher might calculate the mean and standard deviation of a single participant's responses under different conditions and then use an inferential test to formally compare those conditions.

Although visual inspection remains the primary analytic tool, formal statistical procedures are increasingly utilized as supplementary methods in single-subject research. One statistical approach mirrors group research by calculating the mean and standard deviation for each participant's responses under different conditions, followed by the application of inferential tests like the $$t$$ test or analysis of variance to formally compare these conditions.

Google

Single-subject research, frequently referred to as small-$$n$$ design (where $$n$$ represents the statistical symbol for sample size), is a quantitative methodology that involves studying a small number of participants and focusing closely on each individual's behavior. Despite its name, this approach typically investigates between two and ten participants rather than just one. Serving as a crucial alternative to group research, particularly in applied areas of psychology, single-subject research seeks to comprehend objective behavior by utilizing strict experimental manipulation and control to collect highly structured data that is analyzed using quantitative techniques.

Single-Subject Research

https://kpu.pressbooks.pub/psychmethods4e/

KPU Research Methods in Psychology - 4th American Edition

 Such research usually begins by establishing a baseline. After a reliable baseline has been established, an intervention is introduced. The effects of this intervention are then determined by comparing the baseline level of behavior with the post intervention level. 

Process of Single-Case Designs

Single-case designs are often used to study the effectiveness of a therapeutic method. Single-case studies allow the experimenter to establish cause-effect relationships. More than that, they provide a method of studying clinical behavior (especially therapy methods) that does not require the withholding of treatment by assigning certain participants to control groups or waiting lists. Another practical reason for using single-case designs is that it is often extremely difficult in clinical settings to find enough participants for matching or researcher reading a tailored script to him while a random assignment to control groups.

Uses of Single-Case Designs 

While both single-subject and group research are quantitative methodologies, they differ significantly in their sample size and analytical focus. Single-subject research focuses on the detailed experimental analysis of a small number of individuals (typically two to ten). Conversely, group research—the more common approach in psychology—involves studying large numbers of participants and analyzing their behavior primarily through aggregated metrics like group means and standard deviations.

Comparison of Single-Subject and Group Research

Although both single-subject research and case studies focus in detail on a small number of participants, they differ fundamentally in methodology and data analysis. Case studies represent a qualitative approach focused on understanding people's subjective experience through observation, resulting in relatively unstructured data (such as detailed interviews) that is analyzed using narrative techniques. Conversely, single-subject research is a quantitative approach that seeks to understand objective behavior through strict experimental manipulation and control, yielding highly structured data that is analyzed quantitatively.

Comparison of Single-Subject Research and Case Studies

A primary assumption of single-subject research is the necessity of focusing intensively on individual behavior rather than aggregated data. This focus is critical because group research can obscure important individual differences; for example, an intervention that benefits half the participants but harms the other half might falsely appear to have no effect on average. By isolating individual responses, researchers can uncover these hidden variations and directly address the specific behaviors of particular individuals, such as designing a targeted intervention for a disruptive student.

Intensive Focus on Individuals in Single-Subject Research

The origins of single-subject research can be traced back to the inception of psychology as a scientific field. During the late 1800s, pioneer Wilhelm Wundt conducted intensive studies on a limited number of participants to investigate sensation and consciousness. Additional foundational examples of this focused methodological approach include Herman Ebbinghaus’s pioneering studies on memory and Ivan Pavlov’s discovery of classical conditioning, illustrating that the detailed examination of individual subjects is a deeply rooted tradition in psychological research.

Early History of Single-Subject Research

The experimental analysis of behavior is a prominent subfield of behaviorism that relies almost exclusively on single-subject research designs to investigate how external factors, such as rewards and punishments, systematically influence behavior over time. Developed in the mid-20th century by B. F. Skinner, who refined the techniques of single-subject methodology, this approach was traditionally conducted using nonhuman subjects like rats and pigeons to establish foundational principles of learning.

Experimental Analysis of Behavior

Although single-subject research is most closely associated with the behavioral perspective, its methodological framework is versatile and can address questions from any theoretical orientation. For example, researchers can use single-subject designs to test cognitive-based studying techniques on individual students. Furthermore, clinical practitioners operating from cognitive, humanistic, psychodynamic, or behavioral perspectives can effectively use this approach to document therapeutic change and measure individual clients' improvement over time.

Theoretical Perspectives in Single-Subject Research

The reversal design, also referred to as the ABA design, is the most fundamental single-subject research design. In this approach, a researcher first establishes a baseline of the dependent variable (Phase A), then introduces a treatment (Phase B), and finally removes the treatment to return the participant to the baseline condition (Phase A).

Reversal Design

An AB design is a basic single-subject research structure that involves only two phases: an initial baseline phase (A) followed by a single treatment phase (B). This methodology functions similarly to an interrupted time-series design applied to a single participant. A significant limitation of the AB design is its susceptibility to confounding variables; if the dependent variable changes following the intervention, it remains unclear whether the treatment or a coincidental extraneous variable caused the change, resulting in lower internal validity compared to reversal designs.

AB Design

Visual inspection is the primary method of data analysis in single-subject research, serving as an alternative to the inferential statistics typical of group research. This approach involves plotting an individual participant's data, carefully observing the resulting graphs, and making judgments about whether and to what extent the independent variable had a noticeable effect on the dependent variable.

Visual Inspection in Single-Subject Research

Statistical Analysis in Single-Subject Research

Single-subject research is guided by three foundational assumptions: focusing intensively on individual behavior to uncover unique differences, discovering causal relationships through rigorous experimental control, and prioritizing the study of effects that possess significant biological or social importance in real-world settings.

Assumptions of Single-Subject Research

An early demonstration of single-subject research in an applied setting was conducted by Vance Hall and his colleagues in 1968. To address disruptive behavior in a classroom, researchers first observed the students to establish a baseline of their usual activity. Following this, a treatment was implemented where the teacher exclusively provided positive attention when a student engaged in schoolwork, ignoring any disruptions. The intervention led to a significant increase in study time and a drop in misbehavior. When the positive reinforcement was paused, the students reverted to their prior disruptive habits, confirming that the teacher's focused attention was the direct cause of the behavioral improvement.

Vance Hall's Study on Teacher Attention

Single-subject research designs repeatedly measure behavior within one participant or unit across baseline and intervention phases so researchers can evaluate whether a treatment is associated with a clear behavior change.

Learn Before

Related

Learn After