We provide a set of normative data on the Mini‐Mental State Examination (MMSE) from a random sample of 906 normal, healthy subjects of both sexes, 20 to 79 years of age. Subjects were selected in six Italian cities and in the Republic of San Marino. Results show the significant influence of sociodemographic variables, such as age and education, on MMSE performance. We used Multiple Linear Regression Analyses to correct MMSE normative values for these demographic effects. The analyses allow us to specify cutoff scores for distinguishing “normal” performance from “borderline” and pathologic performance.