ABSTRACT

Due to the increased interest in establishing an evidence base for interventions, along with the difficulties encountered in large scale experimental studies, there has been a substantial increase in the use of single-subject experimental designs (SSEDs). To enhance generalizability, researchers replicate across subjects. The research that is proposed focuses on the multilevel modeling framework for the (meta-)analysis of SSED studies, both using raw data and effect sizes. In a first part of the project, we will look at possible approaches for handling complexities that were not (fully) investigated in previous research, including the: (a) estimation of the variance components at the subject and study level, (b) heterogeneity in the covariance structure, (c) analysis of non-continuous outcome variables, (d) subject-specific external event effects, (e) response guided decisions, (f) complex functional forms, (g) mixed designs and (h) dependencies that arise due to combining multiple types of effects, using multiple outcomes, or observing in multiple settings. In the second part, we will study scenarios where multiple complexities happen simultaneously (e.g., some studies using count data, others using interval-scaled data; one or more studies outcomes exhibiting different trajectory forms and/or external events; some studies using MBDs, others use ATDs or other designs, etc.). The results of our study are intended to guide applied SSED researchers in setting up studies, doing the analyses and interpreting and reporting the results.