Systematic Review of Single-Case Designs to Identify Mediator Characteristics (Poster 2)

• In Event: Innovative Methodologies and Systematic Reviews in Analyzing Single-Case Experimental Design Data

Abstract

Mediation analysis in Single Case Experimental Designs (SCEDs) allows for evaluating mechanisms through which interventions achieve desired effects for a single individual. Prior to extending methodological work validating statistical approaches for mediation analysis using piecewise regression and hierarchical linear modeling, an overview of characteristics of mediators typically encountered is needed. This will inform design conditions to be embedded in future methodological studies and ensure that these conditions are realistic and representative for the field of SCED meta-analyses. In addition, a database will be created providing raw data, which will be used for demonstration/illustration purposes.
The systematic literature search procedure as outlined by Jamshidi et al. (2020) and used in Moeyaert et al. (2023) was replicated in current research by two independent coders. To be included, the study is required to be (a) available in English, (b) peer-reviewed with full text availability, (c) published in 2023, (d) related to the field of education, and (e) a meta-analysis including SCED studies. The following six online databases were used: PsycINFO, Web of Science, Science Direct, Medline PubMed, ERIC, and CINAHL. Next, for all identified SCED meta-analyses, the primary study articles will be systematically reviewed for mediator characteristics. Two sets of search strings were specified in all databases: (“single case” OR “single subject” OR “N of 1” OR “small N” OR “multiple baseline design” OR “alternating treatments design” OR “reversal design” OR “withdrawal design”) AND (“meta-analysis” OR “synthesis” OR “review”). The two independent researchers replicated Jamshidi’s search procedure for 2023 for all of the six databases and verified whether the same number of studies per database were retained. The same two independent researchers applied the inclusion criteria to identify the pool of eligible SCED meta-analyses that will be used for further analysis.
The initial search through the six online databases yielded470 records after duplicates across databases were removed. After applying the inclusion criteria, a total of 29 SCED meta-analyses published in the field of education in 2023 were identified. The next step is the systematic review of primary studies per SCED meta-analysis to identify mediator characteristics related to specific topics. An overview of eligible studies per SCED meta-analysis will be provided by August 2024. Next, raw data from eligible primary studies will be digitized for further analysis (e.g., measurement scale of the mediators, size of the direct effect, a-b-paths, and indirect effect), and a complete overview of SCED mediator characteristics will provided by September 2024.
This research project will significantly contribute to the field as SCED researchers are interested in examining mechanisms through which the intervention changes the outcome. Therefore, further methodological work is needed to identify appropriate statistical techniques that can be recommended to the field. Based on the results of the systematic review, recommendations are given for conditions to be included in future methodological work focusing on the analysis and meta-analysis of SCED mediators. In addition, an open-source database will be created containing the raw data for demonstrations and illustrative purposes.

Authors

• Mariola Moeyaert, University at Albany - SUNY

• Joelle Fingerhut, Marist College

• Milica Miocevic, McGill University

• Matthew Valente, University of South Florida

• Panpan Yang, Princeton University

• Yaosheng Lou, University at Albany - SUNY