Association for Behavior Analysis International

This presentation will review the origins of single-case research methods in basic laboratory research, how the designs have been used in basic and applied behavior analysis, and the strengths and weaknesses of past and current practices. In considering origins, we will discuss the essential characteristics of single-case research, the steady-state strategy, the importance of replication, the place of statistical inference, and the extent to which the experimenter’s judgment figures into the design and conduct of research. Historically, single-case research in behavior analysis has been associated with inductive and ideographic paradigms. As single-case methods have been extended to applied research in and outside of behavior analysis, they have become more aligned with deductive and nomothetic paradigms. There has also been an increase in the propagation of standards/quality indicators for single-case research and suggestions for changes that allow for the alignment of the outcomes of single-case research with those of other research paradigms (e.g., effect sizes, randomization). We will consider the ways in which the use of different paradigms has led to disparate approaches to single-case research. Understanding these differences is necessary for understanding the empirical contributions of each approach and for successfully prioritizing experimental standards that should be met for individual studies and groups of research.

Single-case researchers

Quality Indicators of Single-Case Design (QI-SCD) formalize expert consensus about research conventions intended to ensure internal, external, and social validity in single-case designs. They were developed to prospectively guide the development of research studies, as well as to standardize retrospective appraisals of the quality of evidence in support of a given practice or procedure. Although QI-SCD have proven useful to both endeavors, the unqualified prioritization of QI-SCD can sometimes lead to invalid methodology. It can also have a suppressive effect on the complexity of the research questions we ask as a field and can lead to the oversimplification of extant evidence. In this presentation, we will provide a brief overview of QI-SCDs, will match QI-SCDs to specific empirical objectives (i.e., internal, external, & social validity), and will give attendees practice identifying and justifying the prioritization of some QI-SCDs over others, by logically tethering them to the objective(s) of their research.

Single-case researchers

Scientific progress is driven, in part, by ongoing improvement in research methods. In the past decade, research methodologists have described substantial problems with replication of group comparison research in numerous disciplines and have identified a set of questionable research methods that appear to be important contributors to these problems. Much of the success of behavior analysis stems from the use of single-case experimental research (SCER). Heretofore, the concept of questionable research practices has not been systematically applied to SCER. This session will describe a systematic process to understand and identify potential questionable research practices and alternative improved research practices in SCER. This process included both conceptual development and bottom-up derivation of potential questionable and improved practices through sustained engagement with dozens of experienced SCER researchers as well as a broad survey of over a hundred researchers. Much of the session will be devoted to describing a set of recommended improvements in SCER procedures, data analysis, and reporting practices to further enhance its scientific validity and applied usefulness. These recommendations have immediate implications for conducting, reporting, reviewing, and applying SCER.

Single-case researchers

Across domains from health care to education, systematic reviews are used to integrate findings from multiple past studies to inform evidence-based practice recommendations. In areas where single-case designs (SCDs) are widely used, researchers have recognized the importance of incorporating such studies into systematic reviews, and this goal has driven many developments in methodology—as well as some controversies. This session will examine how systematic review methods can account for the unique features of SCD studies. We will first describe core organizing principles of systematic reviews and delineate several distinct types, including scoping reviews, evidence maps, and quantitative meta-analyses. We will then consider three fundamental challenges for systematic reviews focused on intervention efficacy: 1) identifying all relevant evidence; 2) determining how to align findings across studies using heterogeneous procedures; and 3) assessing risks of bias in primary SCD studies. Finally, we will highlight practices that can strengthen the quality of systematic reviews of single-case studies. Presenters will then lead a structured discussion inviting participants to consider a) critiques of published systematic reviews of single-case research; b) limitations of the presenters’ methodological practice recommendations; and c) how primary study practices might need modification to support their inclusion in systematic reviews.

Single-case researchers