This study assesses the validity of self-reported height and weight in a sample of 4,808 middle-aged men and women from the EPIC-Oxford cohort. The researchers found high Spearman rank correlations (r > 0.9, P < 0.0001) between self-reported and measured height, weight, and BMI. On average, men overestimated their height by 1.23 cm (95% CI 1.11–1.34) and women overestimated their height by 0.60 cm (0.51–0.70). Weight was underestimated by 1.85 kg (1.72–1.99) in men and 1.40 kg (1.31–1.49) in women. The extent of overestimation of height was greater in older individuals, shorter men, and heavier women, while the extent of underestimation of weight was greater in heavier individuals but did not vary with age or height. Using standard BMI categories, 22.4% of men and 18.0% of women were misclassified based on self-reported data. Correcting self-reported values using predictive equations derived from a 10% sample of subjects reduced misclassification to 15.2% in men and 13.8% in women. The study concludes that self-reported height and weight data are valid for identifying relationships in epidemiological studies, but measurements in a representative sample can improve the accuracy of estimates of height, weight, and BMI.This study assesses the validity of self-reported height and weight in a sample of 4,808 middle-aged men and women from the EPIC-Oxford cohort. The researchers found high Spearman rank correlations (r > 0.9, P < 0.0001) between self-reported and measured height, weight, and BMI. On average, men overestimated their height by 1.23 cm (95% CI 1.11–1.34) and women overestimated their height by 0.60 cm (0.51–0.70). Weight was underestimated by 1.85 kg (1.72–1.99) in men and 1.40 kg (1.31–1.49) in women. The extent of overestimation of height was greater in older individuals, shorter men, and heavier women, while the extent of underestimation of weight was greater in heavier individuals but did not vary with age or height. Using standard BMI categories, 22.4% of men and 18.0% of women were misclassified based on self-reported data. Correcting self-reported values using predictive equations derived from a 10% sample of subjects reduced misclassification to 15.2% in men and 13.8% in women. The study concludes that self-reported height and weight data are valid for identifying relationships in epidemiological studies, but measurements in a representative sample can improve the accuracy of estimates of height, weight, and BMI.