Using Ssdforr: An R Package for Analyzing Single-Subject Design​

Background and Purpose: Single-subject research designs are often used as a method for social workers to assess their own practices. The notion of social workers evaluating their practices has historically been supported by the NASW Code of Ethics (National Association of Social Workers, 2008). More recently, CSWE’s Educational Policy and Accreditation Standards (EPAS) have promoted engaging in research-informed practice and practice-informed research as one of ten core social work competencies (Council on Social Work Education (CSWE), 2022). The Council has also promoted using data for the purposes of monitoring client progress and evaluating the effectiveness of interventions (Council on Social Work Education (CSWE), 2022).

Analysis of single-subject research typically occurs on two levels: visually and statistically (Auerbach & Zeitlin, 2021). Visual analysis is the most commonly used approach (Orme & Combs 2012); however, there are situations in which visual analysis alone may not be sufficient. Statistical analysis may be valuable if observed effects are small, the variation is large, the effects of the intervention do not appear immediately as phases change, there is no clear trend in the data within phases, or when there may be an issue of autocorrelation, which is nearly impossible to detect through visual inspection alone (Auerbach & Zeitlin, 2021).

Single-subject research typically compares measured outcomes over time under varying conditions, but across the same client unit (e.g., individual, family, or group). Therefore, it is often appropriate to apply different statistical procedures and analysis in single-subject designs than in group comparisons (Kratochwill et al., 2010). These consist of both parametric and non-parametric methods including analysis of autocorrelation, effect size, regression methods, standard-deviation band analysis, conservative dual criteria analysis, and others (Auerbach & Zeitlin, 2021).

There has been little in the way of software to help those interested in conducting single-subject research collect, analyze, and interpret their data. The application, SSDforR, has advanced capabilities that make it more suitable for advanced, publishable research in agencies or other settings (Auerbach & Zeitlin, 2013; Auerbach & Zeitlin, 2021). These include comprehensive visual and statistical functions also for cases that could be considered “single-subject,” whether those cases include individuals, couples, families, groups, or communities.

SSDforR is a comprehensive package of statistical tests that produces accompanying graphs for almost every function. These include, among others, tests of autocorrelation, trends, t-tests, effect sizes, chi-square analyses, conservative-dual criteria analysis and statistical process control (SPC) charts (Auerbach & Zeitlin, 2021). Participants will be shown where and how to download the software and functions and will be given the opportunity to walk through the use of these functions by tracking a hypothetical client through a practice evaluation study. This will include testing for autocorrelations and trends, visual analysis, using appropriate effect size; and using an appropriate tests of type I error to compare the baseline phase to the intervention phase.

The presenters will discuss their experiences using SSDforR in Master’s level courses. Participants will be given a host of resources to enable them to use this package with their students and in their own research.