Calculate Task Split Half Reliability Estimates

A series of functions to calculate the split half reliability of RT based tasks. The core function performs a Monte Carlo procedure to process a user defined number of random splits in order to provide a better reliability estimate. The current functions target the dot- probe task, however, can be modified for other tasks.

splithalf R package

This package contains several functions to calculate the split-half reliability of response time based cognitive tasks.

Currently the package contains the following functions;

splithalf() calculates the split-half reliability within the conditions specified. DPsplithalf() calculates the split-half reliability for the attention bias index from dot-probe data, in each condition specified. DPsplithalf.all() calculates the split-half reliability for the attention bias index from dot-probe data, and the split-half reliabilities of congruent and incongruent trials, in each condition specified. TSTsplithalf() calculates the split-half reliability for switch cost indices in a task switching paradigm, as well as for the repeat and switch trials separately.

Functions will be added as they are developed, including accuracy rate indices, task-switching paradigms. The central code of the splithalf functions can be modified as needed to fit with most tasks.

Examples for each function are provided in the function descriptions, and two extended examples are presented in the vignette.


splithalf v0.2.0 (current development version)


  • Added the TSTsplithalf function
  • Some minor bug fixes
  • Output for all functions now includes two-part alpha

splithalf v0.1.0 (Release data: 07/04/2017)


  • Submitted to CRAN

Reference manual

It appears you don't have a PDF plugin for this browser. You can click here to download the reference manual.


0.2.0 by Sam Parsons, 10 months ago

Report a bug at

Browse source code at

Authors: Sam Parsons [aut, cre]

Documentation:   PDF Manual  

GPL-3 license

Imports plyr, stats

Suggests testthat, knitr, rmarkdown, tools, ggplot2

See at CRAN