The aims of a national dietary study are several-fold. One purpose is to monitor the intake of foods and nutrients in the population and to compare intakes with dietary recommendations. It is, however, difficult to measure dietary fat intake and plasma biomarker fatty acid (FA) composition may be used as an objective measure of dietary fat intake. Thus, the relative ability of a diet record to capture habitual fat intake was validated against biomarker FA. Dietary fat intake was measured in a novel self-assisted web-based 4-d food record – the ‘Riksmaten’ method. Spearman rank correlations between dietary FA, certain food groups (fish-shellfish, dairy products, meat and sausages, and spreads) and the fat content of these food groups and biomarker FA were explored. Participants were 150 women and 129 men, aged 18–80 years, who took part in the Swedish National Dietary Survey, Riksmaten adults 2010–11. Blood samples were collected on average 20 d after the diet record and FA composition was measured in plasma phospholipids by GLC. Total n-3 FA (r 0·31), EPA (r 0·34) and DHA (r 0·42) were correlated between plasma and diet (all P ≤ 0·001). Adjustment for covariates attenuated the relationships. Linoleic acid was only marginally correlated (r 0·15; P = 0·06) in women. Plasma pentadecaenoic acid and heptadecaenoic acid were correlated with dairy product intake as previously reported. In conclusion, the Riksmaten method appears valid for the purpose of collecting data on dietary fat composition, at least in a healthy adult population.