It measures the strength of the association between an exposure and a disease by dividing the incidence rate (number of cases in exposed/number of people exposed) by the incidence rate in unexposed individuals (number of cases in unexposed/number of people unexposed).