Single Case Design Tools

In some situations where researchers would like to demonstrate causal effects, it is hard to obtain a sample size that would allow for a well-powered randomized controlled trial. Single case designs are experimental designs that can be used to demonstrate causal effects with only one participant or with only a few participants. The 'scdtb' package provides a suite of tools for analyzing data from studies that use single case designs. The nap() function can be used to compute the nonoverlap of all pairs as outlined by the What Works Clearinghouse (2022) < https://ies.ed.gov/ncee/wwc/Handbooks>. The package also offers the mixed_model_analysis() and cross_lagged() functions which implement mixed effects models and cross lagged analyses as described in Maric & van der Werff (2020) . The randomization_test() function implements randomization tests based on methods presented in Onghena (2020) . The scdtb() 'shiny' application can be used to upload single case design data and access various 'scdtb' tools for plotting and analysis.

scdtb

The goal of ‘scdtb’ is to provide tools for the analysis of single case design data.

Installation

You can install the released version of ‘scdtb’ from CRAN:

install.packages("scdtb")

To install the development version of ‘scdtb’ from GitHub, use the devtools package:

# install.packages("devtools")
devtools::install_github("mightymetrika/scdtb")

Analyze Your Data

Use the scdtb() function to launch a ‘shiny’ application which implements many of the tools in the ‘scdtb’ package.

The application has the following functionalities:

• Upload a CSV file to be used as the dataset for modeling.
• View the variables available in the uploaded dataset.
• Enter the roles played by the variables in the dataset.
• Create a raw plot of the data.
• Run a mixed effect model.
• Run a crossed lag correlation analysis.
• Run a non-overlap of all pairs computation.
• Run a randomization test.

library(scdtb)

# Launch application
scdtb()