How to Analyze Single Case Studies in Education: An Illustration with Two Alternative Methods

• Activity Number: 475
• Type: Invited
• Date/Time: Wednesday, August 6, 2014 : 10:30 AM to 12:20 PM
• Sponsor: Social Statistics Section
• Author(s): Diep Thi Nguyen*+ and John M. Ferron
• Companies: University of South Florida and University of South Florida
• Keywords: single case study ; single subject research ; multiple baseline ; non-parametric approach

Abstract

In educational settings, single-case, or single subject, research is the intensive study of individual learners through repeated measurement of an outcome while altering the conditions under which the learner is being observed. The most common single case design is a multiple-baseline design, which is characterized by short interrupted time series data (typically 10 to 30 observations) on a small number of participants (commonly 4 to 8 persons) where the introduction of the intervention is temporally staggered across the participants. These features present some challenges for traditional analyses because small sample size at level 1 (short time series) or level 2 (few participants) makes it difficult to identify and estimate an appropriate model for the data. This paper illustrates two alternative methods for analyzing multiple baseline studies: a variation of multilevel modeling and a non-parametric approach. A reanalysis of a previously published multiple baseline study is presented to demonstrate the use of these two methods in single subject educational research.