Reliability of cognitive measurements

Short summary:

It is a travesty that measurement reliability is not reported as standard practice within experimental psychopathology, and more generally in research that utilises ‘cognitive bias’ measures. Simply put, this needs to change. This project aims to promote the routine estimation and reporting of measurement reliability. I maintain, and continue to develop, the splithalf R package, allowing for robust (read: permutation/bootstrapped approaches) estimation of internal consitency reliability of cognitive bias measures.

You can find the package on CRAN (warning, currently it is several versions behind); or better, on github.

There is a fairly detailed documentation/tutorial website on

If you are interested in finding out how reliable your measurements are, but don’t know where to start, fire an email my way and we can talk.


Psychological science relies on behavioral measures to assess cognitive processing; however, the field has not yet developed a …

The relationship between measurement reliability and statistical power is a complex one. Where reliability is defined by classical test …