scan - Single Case Data Analysis with R

The scan package (formerly SCDA) was designed for analysing and visualizing single-case AB data sets. These designs are commonly used for evaluating learning progress and curriculum-based measurements. Furthermore, the scan package comprises a module for generating random single and multiple-baseline datasets under a variety of circumstances for conducting Monte-Carlo studies on the power of single-case analysing techniques.

Some of the included analyzing procedures are:

  • Overlapping indices: Percentage of all non overlapping data, Nonoverlap of all pairs, percentage exceeding the trend, percentage exceeding the median, percentage non overlapping data
  • Piecewise regression model
  • Hierachical piecewise regression
  • Randomization tests
  • Standardized mean difference
  • Reliable change index
  • Outlier analyses
  • Data smoothing
  • Interpolation of missing values
  • Analyses of linear, cubic etc. trends


Please consider that the scan package is still under development. There may be - and surely are - several bugs in the code. So be careful using it.

A draft of a book on scan can be found at:

https://jazznbass.github.io/scan-Book/

 

scan is available from the rforge repository.

https://r-forge.r-project.org/projects/scan/

Install scan from within R with:

install.packages("scan", repos="http://R-Forge.R-project.org")

Make sure you installed all dependencies in advance.

If you encounter further problems, please feel free to write an email to: juergen.wilbertuni-potsdamde