Power considerations for single-case neuropsychology

Activity: Academic talk or presentation typesOral presentation

Description

Case-control comparisons are widely used in neuropsychology for diagnosing and investigating cognitive-behavioural deficits and dissociations at the single-case level. The tests developed for this purpose by John Crawford and collaborators provide transparent control over Type I (false positive) error rates. But far less attention has been given to the ability to avoid Type II (false negative) errors; that is, to the statistical power of these tests. We aim to give an accessible overview of power in case-control comparisons, and we provide a package for R (singcar), which implements a range of relevant power calculations. We suggest that power calculations can usefully inform the design and interpretation of single-case studies in neuropsychology. However, there are hard limits on power, which may have important implications for the preregistration of single-case studies, and more generally for the inferential status of case-control comparisons in clinical and research contexts.
Period23 Apr 2021
Event titleBritish Neuropsychological Society Spring Meeting
Event typeConference
Degree of RecognitionNational