Single-case design (SCD) is one type of experimental method for examining the relation between independent variables and dependent variables using a small number of participants (or subjects). In research using single-case designs, individual behavior is repeatedly measured in both the absence and presence of one or more independent variables. Single-case designs have an important advantage over the group comparison method (i.e., randomized controlled trials) in that independent variables can be changed flexibly and quickly. However, in current research in experimental psychology, the group comparison method is considered the gold standard for examining causal relations. Single-case designs have not always been evaluated fairly as an experimental method. However, recently, an increased interest in single-case designs has been observed in a variety of disciplines. Based on this increasing interest, the present article provides a brief history of single-case designs and reviews the present condition of single-case designs both inside and outside of Japan. Furthermore, conditions needed for the development and expanded use of single-case designs are described. The present article contends that the movement for evidence-based practice has had an impact on the development of various kinds of single-case-design standards.

View full abstract

The present article explains fundamental characteristics of single-case experimental designs and introduces their basic variations: reversal, changing criterion, multiple baseline, and alternating treatments. In order to evaluate the strength of the evidence yielded by those designs, the article explains how researchers can attempt to enhance the internal and external validity of the conclusions of experiments using those designs. Although group designs address, as a matter of internal validity, threats to validity such as participant selection, interactions with selection, and some interference between treatments, single-case designs must treat those threats as a matter of external validity in experimental replications. Some problems with conventional replications are described, the need for guidelines for reporting single-case experiments is suggested, and the formation of practice-based research networks is proposed.

View full abstract

In the present session, 2 reports were presented from the field of nursing care. The first report was about support for an elderly woman with dementia living in a nursing home; the second, about support for organizational development in a company providing visiting care. Both reports focused on the application of single-case designs for enhancing evidence-based collaboration among service providers. The strengths and the difficulties involved in introducing single-case designs in the nursing care field were examined. The discussion dealt with tasks and implications for more widespread use of single-case designs in human services settings.

View full abstract

Methods for reducing problem behaviors of elderly people with dementia who are living in nursing homes have not yet been verified. The aim of the present study was to assess the function of problem behavior using the three-term contingency and to intervene with problem behavior using a single-case research design (AB). The participant was a 90-year-old woman. In the present analysis, there was an intimate interplay in behavior between the participant and the staff. The staff who were reducing the problem behavior managed the participant according to the function of her behavior. For intervention, the method of desirable staff's management was also tried by other staff. The results indicated a stable decrease in the participant's problem behavior. The present study was required to use an AB design because of ethical concerns. Using an ABA or other design would be highly desirable. It is important to manage problem behavior on the basis of evidence.

View full abstract

In order to examine effects of introducing a team system for care workers in a company providing visiting care, the present study initially attempted to conduct an experiment with a single-case design. However, because it was anticipated that this would increase the care workers' burden, that experiment was not carried out. Instead, after the project of introducing the team system had finished, a retrospective analysis was conducted, which targeted 5 part-time care workers who had formed a team and examined their behavior of posting reports to the company's report system, assuming that effects of the formation of the team had been investigated in a quasi-experiment with an A-B design. The results showed an increase in the frequency of report posting by 2 of the care workers, and a change in the content of another care worker's reports after the introduction of the team system. This suggests that a retrospective analysis assuming an A-B design may be useful for delineating effects of an attempt at organizational development. However, it was also found that, in the field of visiting care, obtaining information necessary for experiments was difficult, which hampered rigorous verification of effects of the organizational development introduced.

View full abstract

The purpose of the present article is to provide a review of statistical methods for single-case research designs. This review includes the following: (a) features of single-case data, (b) a summary of related studies about the reliability of visual inspection, (c) a summary of statistical inference methods (i.e., statistical tests), (d) a summary of descriptive statistics (i.e., effect-size indices), (e) introduction of some resources for studying these methods, and (f) a review of the present status of these methods and a discussion of the review.

View full abstract

The purpose of the present study was to suggest a possible joint methodology combining qualitative inquiry and behavior analysis. These 2 methods have many methodological commonalities, such as the importance of utilizing only a few participants and the interaction between the researchers and the participants, emphasis on day-to-day behavioral interactions, and orientation towards knowledge- generation in a specific context. Through individual narrative interviews, it may be possible to examine the meaning of a target behavior or an intervention process from the viewpoints of practitioners or their clients. Also, for social validity, participant observation could be useful for examining the interaction between the participants and other individuals in a broader context, such as schools or hospitals. It is important to remember that the epistemological background of qualitative research is extremely different from that of behavior analysis, in which "truth" or "fact" is not considered static, but is viewed as dependent on the sociocultural context. Narratives or stories, on the other hand, are not a mere reflection of the "inner" state of the participant, but are the construction of reality between an interviewer and interviewee in a specific context. Thus, based on these differences, it is suggested that a study combining these methods should be conducted.

View full abstract

In the present article, I argue frankly about single-case designs in general, including the methods themselves and how they should be applied, rather than making comments on each individual presentation. My argument could be said, in turn, to be a comment from a non-member of the Japanese Association of Behavior Analysis. First, I state a strong sympathy for single-case designs, which usually examine causality between manipulated independent variables and measured dependent variables within each case, followed by a close examination of generalizability. Then I raise questions about studies done with single-case designs in relation to interpretations of sustainability or irreversibility and the generalization of effects, applications of statistical analysis, and other matters that should be mentioned in academic articles. Finally, I comment on recently published articles in the Japanese Journal ofBehavior Analysis, based on my critical and strict attitude, as well as my concern that we not abandon efforts to think carefully.

View full abstract

The present paper, presented at the Symposium held on July 26, 2013, at Telepia Hall in Nagoya, in commemoration of the 30th Anniversary of the Japanese Association for Behavior Analysis, describes a brief history of the adoption of the single-subject method by Skinner in his experiments during World War II, and summarizes the dissemination of the analysis of variance in agricultural testing in Japan immediately after the War. The importance of the social environment in the dissemination of research methods was discussed.

View full abstract