Total plasma fatty acids or those in cholesteryl ester and phospholipids are often used to reflect fatty acid intake in epidemiological studies, but their relative performance as biomarkers of intake has not been clearly evaluated within a single population. The assessment of fatty acids in plasma fractions is more labor intensive. Thus, their use as biomarkers of dietary intake needs to be justified. Dietary intake was assessed in 200 population-based controls from a case-control study of diet and heart disease in Costa Rica by a validated food frequency questionnaire (FFQ). Fatty acids in fasting whole plasma and plasma fractions (cholesteryl ester, phospholipid, and triglyceride + free fatty acid) were measured in the 200 controls by the same laboratory using gas chromatography with flame ionization detection (GC-FID). We compared the plasma and plasma fractions data with the FFQ and adipose fatty acid profile using partial Spearman correlations to assess utility as biomarkers of intake and exposure. We found that whole plasma was equally or more strongly correlated with the FFQ and adipose fatty acid profile than either cholesteryl ester or phospholipid in most of the established markers of dietary intake, including dairy (15:0 and 17:0) and seafood (eicosapentaenoic acid and docosahexaenoic acid). Of the three plasma fractions, only fatty acids in the plasma triglyceride + free fatty acid fraction had stronger correlations with dietary intake than whole plasma. In our study population, fatty acids measured in fasting whole plasma perform as good as or better than those measured in plasma fractions as biomarkers for dietary fatty acid intake. Thus, the fractionation of plasma to evaluate long-term fatty acid intake may not be warranted.