An R package for single-case randomization tests

Abstract

Randomization tests are nonparametric statistical tests that obtain their validity by computationally mimicking the random assignment procedure that was used in the design phase of a study. Because randomization tests do not rely on a random sampling assumption, they can provide a better alternative than parametric statistical tests for analyzing data from single-case designs. In this article, an R package is described for use in designing single-case phase (AB, ABA, and ABAB) and alternation (completely randomized, alternating treatments, and randomized block) experiments, as well as for conducting statistical analyses on data gathered by means of such designs. The R code is presented in a step-by-step way, which at the same time clarifies the rationale behind single-case randomization tests.

CAML—Maximum likelihood consensus analysis

“…Although software written in R was very successful in recent years (e.g., Bulté & Onghena, 2008;Grassie et al, 2010;Nimon et al, 2008), the user interface of R might not appeal to everyone. We intend to address this limitation in future versions of CAML.…”

“…R is freely available and is supported by an active community that continuously provides software solutions to various statistical problems (e.g., Bulté & Onghena, 2008;Grassie, Luccio, & Di Blas, 2010;Nimon, Lewis, Kane, & Haynes, 2008). CAML makes use of the maximum likelihood method of parameter estimation.…”

Percentage of nonoverlapping corrected data

“…Therefore, we consider that the modifications balancing statistical properties improvements and low levels of calculus/interpretative complexity have to be encouraged. Furthermore, the present study followed the practice of offering data analysis programs for singlecase designs in freeware, such as R (see, e.g., Bulté & Onghena, 2008)-a practice that we feel ought to be promoted.…”

Randomization tests for multiple-baseline designs: An extension of the SCRT-R package

“…This software program, however, is Macintosh-based and has no IBM PC-compatible version (Koehler & Levin, 2000). We already created an R package, SCRT-R, to perform randomization tests on data from single-case phase and alternation designs (Bulté & Onghena, 2008). Here, we present an extension for multiple-baseline AB data.…”

“…The null hypothesis is rejected when this value is less than or equal to the predetermined significance level (Murray, Varnell, & Blitstein, 2004;Potvin & Roff, 1993;Strauss, 1982). A more comprehensive step-by-step explanation of the randomization test procedure can be found in Bulté and Onghena (2008).…”