Objective: It is controversial if the clustering of certain metabolic abnormalities should be separately designated as the metabolic syndrome (MetSyn). We operationalized the „syndrome“ concept and tested whether MetSyn was compatible with these operational constructs. Research Design and Methods: The baseline cross section of the Multi-Ethnic Study of Atherosclerosis recruited a population-based cohort of 6781 persons, aged 45-84 years, from 6 communities in the United States. MetSyn components (waist circumference, blood pressure, fasting serum HDL-cholesterol, triglycerides, plasma glucose), HOMA insulin resistance (fasting glucosexinsulin) and intimal-medial thickness (IMT) in the common and internal carotid arteries by B-mode ultrasound were measured. Results: 1. Higher syndrome component count is associated with higher HOMA levels (trend p < 0.001). 2. Given the prevalence of individual components, the non-prevalence of any component or the co-prevalence of 4 or 5 components is greater than expected (chi(2) p<0.001). 3. After accounting for the additive association of each component, the current definition of MetSyn (co-prevalence of 3 or more components) does not have supra-additive association with thicker IMT in the common carotid (men: p = 0.075, women p = 0.949) or internal carotid artery (men: p = 0.106, women: p = 0.121). Conclusions: 1. MetSyn did not have supra-additive association with IMT, but 2. its components clustered greater than chance expectation and 3. higher component count was associated with greater insulin resistance. MetSyn was compatible with two of three „syndrome“ constructs tested.