High blood cholesterol increases the risk for heart disease and is a major modifiable risk factor (Morbidity and Mortality Weekly Report, 2000). Since the interest in cholesterol and its corresponding link to heart disease has emerged, the study of risk factors that are associated with elevated cholesterol and more specifically elevated low density lipoprotein cholesterol (LDL-C) has increased. The purpose of this study was to identify factors that may discriminate between high and low capillary cholesterol (below 200 and 200 and above), high and low total venous cholesterol (below 200 and 200 and above), and high and low venous cholesterol risk ratio (below 3.5 and 3.5 and above), how their order of importance varies and if there are different predictors for different types of cholesterol measurement. Variables included in the study were age, gender, non-essential body fat (beyond 16% for males and 25% for females; Brizzolara et al., 1999; MeArdle, Katch, & Katch, 1991; Everett, 1989; Stamford, 1987), diastolic blood pressure, self-reported aerobic activity, smoking status, and race/ethnicity. All benefits-eligible faculty and staff (9,291 employees) of a large southern university were invited to participate in a worksite health screening. The health promotion program screened 14% (n=1400) of benefits-eligible faculty and staff. The sample was reduced to 2.7% (n=270) after eliminating those participants who did not have total capillary, total venous and risk ratio cholesterol scores. A discriminant analysis was performed. Age was the highest discriminator between high and low total capillary cholesterol with a Wilk's Lambda of .912, followed by diastolic blood pressure (.889), and non-essential fat percentage (.880). Significant discriminators were also found with regards to venous cholesterol. Age was the highest discriminator with a Wilk's Lambda of .928, followed by diastolic blood pressure (.902), non-essential fat percentage (.895), smoking status (.890), and gender (.886). Two variables were found to be significant predictors of high and low venous risk ratio cholesterol. Gender was the highest discriminator with a Wilk’s Lambda of .909, followed by non-essential fat percentage (.838). All variables were significant to the .0001 level of significance. The results demonstrate different discriminant variables emerge when plotted against the different measures of cholesterol. It has been demonstrated that the three methods of cholesterol testing measure different aspects of risk (Bowden, Kingery, & Brizzolara, 2000), but they may also have different predictive risk factors as well.Keyword(s): assessment, health promotion, worksite