Association for Behavior Analysis International

Please join us for opening remarks with conference co-chair, Dr. Timothy Slocum.

This session explores a comprehensive approach to analyzing single-case research by integrating visual analysis with strategies to control Type I error. The first presentation will examine the central role of visual analysis in single-case design, highlighting its flexibility, common applications, and both strengths and limitations. The second presentation will address how to mitigate the risk of drawing incorrect conclusions due to Type I error, outlining analytic methods that complement visual inspection and help ensure valid inferences. Together, these presentations aim to strengthen researchers’ ability to interpret single-case data with greater rigor and confidence by balancing clinical judgment with methodological safeguards.

Single-case researchers, reviewers, editors, and those who teach single-case designs

Behavior may change because of an intervention or may change because of something other than the intervention. If the change is due to something other than the intervention, but we conclude it was from the intervention we have made a Type I error. In this session we will focus on methods that have been developed to limit the chances of making a Type I error when analyzing single-case data. We will consider options that are available, as well as the circumstances in which each option does and does not work well. The aim is to help participants be able to select an approach that will complement their visual analysis, and is appropriate given their single-case research design and their data.

This session presents two complementary approaches to advancing the analysis of single-case data through quantitative models and statistical techniques. The first presentation explores theoretical frameworks (Shull’s bout-analytic model and response disequilibrium theory) that reconceptualize behavior and reinforcement in terms of temporally extended patterns. These models offer new analytic tools for understanding response organization and treatment effects, with strong potential for translational research. The second presentation addresses the quantification of intervention effects in single-case experimental designs (SCDs), focusing on selecting and interpreting appropriate effect size measures. Attendees will be introduced to classes of statistical techniques, including non-overlap, regression-based, and mean-based metrics, and guided through effect size calculation using an intuitive R Shiny application. Together, these talks offer a practical and conceptual foundation for improving the precision, utility, and interpretability of single-case research outcomes.

Single-case researchers, reviewers, editors, and those who teach single-case designs

Quantitative models of behavior provide theory-driven accounts of how behavior unfolds over time and is shaped by environmental contingencies. While such models have a longstanding role in the experimental analysis of behavior, their application in clinical settings remains limited. This presentation highlights two intriguing frameworks—Shull’s (2001) bout-analytic model and Timberlake and Allison’s (1974) response disequilibrium theory—and explores their translational potential for applied behavior analysis. Each model reconceptualizes what counts as a meaningful unit of behavior or reinforcement, shifting focus from discrete to temporally extended events. Shull’s bout model characterizes operant behavior as organized into bouts and pauses, with parameters that correspond to motivational and motoric processes. This structure offers a precise framework for analyzing response patterns and evaluating treatment effects. Response disequilibrium theory reframes reinforcer efficacy in terms of behavioral allocation and constraint, offering a dynamic and testable alternative to traditional notions of reinforcer preference and value. This presentation will provide brief primers on each model, review relevant translational and applied findings, and identify future directions for extending their applications in behavior analysis. Emphasis will be placed on how these frameworks can generate testable hypotheses and enhance conceptual foundations of applied research and practice.

Due to the increased interest in establishing an evidence base for intervention effectiveness using single-case experimental design (SCD) studies, there is a need for appropriate quantification of intervention effectiveness. However, analyzing SCD data is not always straightforward and can be challenging because the choice of an appropriate statistic depends on the research question and SCD data characteristics. During the presentation, a distinction is made between non-overlap statistics, regression-based statistics, mean based-statistics, and others that can be used to quantify intervention effectiveness. A step-by-step demonstration of effect size calculation will be provided using a point-and-click R Shiny application. This presentation is designed to help participants gain a working understanding of appropriate effect size selection and interpretation.

The accumulation of evidence to inform intervention practices requires both methodologically rigorous experimental designs and detailed reporting in published reports. Quality of reporting serves as a proxy for methodological quality including those aspects which we are unable to observe after the experiment concludes. In single-case research, as in other research traditions, researchers sometimes (or often) omit essential methodological details from published reports that researchers need to ascertain methodological quality. This session focuses on topics pertinent to the complete reporting (and not cherry picking) of methods and results in single-case intervention studies. First, we will present an overview of the rationale for enhanced methods and results reporting practices and existing reporting practices within single-case research (Machalicek). Second, we will present data related to statistical conclusion validity in the adapted alternating treatments design, discuss the importance of replicating differences between conditions to decrease the probability of Type I error, and provide recommendations for reporting replications (Petursdottir). Finally, we will present recommendations for the improved articulation of design, visual and statistical analysis decisions in published single-case reports (Vannest). Participants of this session will provide input to inform our collective understanding of the feasibility, acceptability, and potentially effects of our recommendations.

Single-case researchers, reviewers, editors, and those who teach single-case designs

Single Case Design is a well suited, rigorous methodological approach to understanding practices that work with children with significant and complex support needs. Considering SCD plus qualitative in a mixed methods approach can answer questions beyond relationships between practices and outcomes. Mixed Methods research answers research questions that consider who practices work for and the contexts and conditions that promote desired outcomes, and describes discrepant effects in our applied settings of school, homes, and communities (Love et al., 2023). In this panel, we will discuss values of mixed methods approaches that include SCD and approaches in doing so with illustrative examples.

Single-case researchers