The clock drawing test (CDT) is an easy to apply, well accepted and reliable test that is widely used to screen for visuo-constructional difficulties in the aged people. Yet, besides visual agnosia and constructional apraxia, executive control influences performances in this task. Execution modalities and rating schemes vary widely, as for the way stimuli are proposed, the time to which the clock is set, and the elements that are considered for scoring. The scoring system we have selected is called clock drawing interpretation scale (CDIS) contains 20 items in 3 areas: visuo-perceptual, attentional and numerical factors. Our aim is to know the meanings of the CDIS total and sub-areas score, investigating the relationships with specific neuropsychological tests, in elderly persons with moderate cognitive impairment. CDIS has been administered to 90 people, aged about 75 years, attending our rehabilitative day hospital or our memory clinic. A neuropsychological battery has been administered to a sample of 47 outpatients, selected by contingence. Respective mean MMSE for the 2 samples are 24 and 25.1 (adjusted for age and education). The score method shows good internal consistency, with Cronbach's a about 0.75, either for total score or for the 3 sub-scores. Both total score and all sub-scores share correlations with mini mental state examination (MMSE), geriatric depression scale (GDS 5-item form), digit-symbol test, phonetic fluency and constructional apraxia tests. Total score and Group A (visuo-perceptive items) correlate also with cancellation attentional matrices, trail making test A and B, Corsi's cubes and Raven's colored matrices. Group B (attention items) and Group C (numerical factors) lack some of these correlations (Pearson correlation coefficients between 0.264 and 0.629). Neither CDIS total score nor sub-scores (except for numerical factors) correlate with verbal learning and memory. CDIS total cores correlates also with the level of education (r=0.418; p=0.001), but not with age. In conclusion, the clock drawing test, scored by the 20-item CDIS, looks as a homogeneous and analytic test, which is focused on visuo-perceptive and executive skills, while it disregards verbal learning and memory, in elderly people with moderate cognitive impairment. Its 3-item groups show good internal consistency; they also plausibly correlate to specific neuropsychological tests. The almost overlapping results of total and visuo-perceptual sub-score come from the way the test is performed and evaluated. The correlation between CDIS and GDS confirms the relevance of mood towards executive functions.