ABSTRACTThis paper describes efforts to reduce measurement error in the assessment of cardiovascular reactivity by standardizing task requirements and by aggregating data across tasks and testing sessions. Using these methods, reliable measures of reactivity (.80 or greater) were obtained on five different measures of cardiovascular function (heart rate, systolic blood pressure, diastolic blood pressure, stroke volume, pre‐ejection period) in samples of college students and community volunteers. Methodological limitations may have hampered previous efforts in this area. Current findings are consistent with a dispositional model of cardiovascular reactivity, and they suggest productive future strategies for obtaining reliable assessment