Our study aimed to assess whether education affects the positive predictive value of the MMSE when it is used as a screening test for dementia. The MMSE has been consistently found to correlate with education and, at the same time, education is regarded by some researchers as being of potential etiological significance for dementia. In the present study, results on the MMSE from a community-based study in Sweden were compared with two standards: the clinical diagnosis of dementia and the Katz index of daily living activities. For both these comparisons, education did not substantially change the predictive ability of the MMSE. Our data, when compared with previous studies, show that the influence of education on the predictive ability of the MMSE varies in different populations. We conclude that differential MMSE cut-off points for different educational levels or adjustment techniques for education are not universally applicable.