Randomization in single-case design experiments: Addressing threats to internal validity.

Abstract

We review how various forms of randomization can be applied in single-case experimental design (SCED) methodology to help control various threats to internal validity. Randomization strategies that can be added to various SCEDs include phase-order randomization, between-intervention case randomization, within-intervention case randomization, and intervention start-point randomization, along with two- and three-way combinations of each. Specific examples of how these forms of randomization can be applied in numerous variations of SCEDs wherein replication is a primary internal and external validity feature (e.g., intrasubject replication or ABAB, alternating treatment, multiple baseline) to increase the scientific credibility of these methodologies are discussed. We also provide examples of the utility of randomization to control validity threats in nonconventional designs where replication is not part of the design structure. Previous recommendations to adopt randomization have assumed implicit advantages of this strategy but without specific details of how randomization serves to control validity threats. We make explicit how each form of randomization controls for internal validity concerns that traditional replication alone may not address. Additional benefits of randomization in SCED experiments include improving the status of this methodology and increasing the likelihood of researchers including SCED intervention research in their literature syntheses. In addition, design randomization allows for various randomization statistical tests to be conducted, thereby increasing data-evaluation/statistical-conclusion validity. Implications for future SCED intervention research methodology are discussed, along with recommendations targeting the need for randomization standards in SCED research. (PsycInfo Database Record (c) 2025 APA, all rights reserved).