Meta-analysis of the known-group validity (Hedges g) of various cognitive tests, i.e. the capacity of these tests to discriminate patients from controls. The study data will be included upon publication. More information: [email protected] or [email protected]
The project contains two R scripts: a risk of bias assessment using ROBVIS, and the meta-analysis with a forest plot visualization. The random-effect model allows for study-to-study differences in the discrimination power, which is expected since different tests are used to target the same construct. The meta-analysis is separated in various subcomponents of the construct of interest (here social cognition). It also calculates separate summary indices for different clinical populations.