The objective of the present study was to determine the accuracy of two foot-to-foot (FF) bioelectrical impedance analysers (BIA) to assess body composition in overweight and obese adolescents, compared with dual-energy X-ray absorptiometry (DXA) and hand-to-foot (HF) BIA. Body composition was assessed in fifty-three overweight or obese adolescents (BMI 27·9 (sd 4·1) kg/m2; aged 13–16 years) by DXA (Hologic QDR-4500; Hologic Inc., Bedford, MA, USA) and BIA (HF (BIA 101, RJL System, Detroit, IL, USA) and FF (Body Fat Monitor Scale BF-625, Tanita Corporation of America Inc., Arlington Heights, IL, USA; Téfal Bodymaster Vision, Téfal, Rumilly, France)). Bland–Altman tests showed that, compared with DXA, FF-Tanita and FF-Téfal underestimated (P<0·05) fat mass (FM) less than HF-BIA (−1·7 (sd 3·1), −0·7 (sd 5·8) and −2·3 (sd 2·2) kg respectively, P<0·001). However, the limits of agreement between DXA and FF-Tanita or FF-Téfal were much greater than those obtained with HF-BIA (−7·7 and +4·3, −12·0 and +10·6 v. −2·1 and +6·7kg respectively). The differences between FM assessed using the FF-Tanita or the FF-Téfal analyser and DXA increased with the waist:hip ratio, and were higher in boys than in girls. The major limiting factor of FF-BIA was the inter-individual variability in FM estimates. In conclusion, FF-BIA and DXA are not interchangeable methods. FF-BIA could be acceptable to assess body composition in large groups of overweight or obese adolescents, but cannot be recommended for body composition assessment in obese subjects because of the large errors in individual estimates.