Mild cognitive impairment (MCI) is one of the most important clinical entities in the field of cognitive neuroscience, the size of their numbers and the possible evolution of some of the subtypes of dementia. The aim of this study is to test a method and procedure for classification of MCI in terms of neuropsychological test scores and its convergence with established clinical criteria, and analyze the differences between the resulting groups. After applying the inclusion-exclusion criteria, was as-sessed using a battery of neuropsychological tests to 39 people aged be-tween 60 and 89 years. Diagnostic groups were established by psychometric criteria and a cluster analysis and satisfactory results were obtained by analyzing the coherence between the two criteria. There is a significant number of subjects diagnosed with MCI (43.6% and 53.9% as a function of each criterion), showing them lower levels of schooling than healthy controls. MCI prevention and the definition of subtypes is necessary, al-though the use of criteria: objective, scientifically valid and assess different cognitive domains, is an appropriate mechanism to be analysis and diagnosis.