It is often assumed that neuropsychological measures are ecologically valid in 'normal' people, but this assumption has not yet been thoroughly evaluated. The aim of the present study was to evaluate the cross-sectional and longitudinal ecological validity of individual neuropsychological test scores and their composites in a large sample of neurologically intact people. Three neuropsychological composite measures were established, i.e. a "Memory Quotient", an "Executive functioning and Speed Quotient", and a "General Cognitive Quotient". The ecological validity of the individual neuropsychological measures and their composites was low to moderate. Multivariate models that included both neuropsychological and non-cognitive variables (i.e. demographic variables, depressive symptoms and anxiety) accounted for 4.6-21.4% of the variance in daily life functioning. The General Cognitive Quotient was the neuropsychological measure that was the most consistently related to daily life functioning.