Introduction: studying the percentage of body fat (%BF) in children and adolescents is very relevant, since a high level of body fat in childhood and adolescence represents overweight and obesity. Objective: to identify the anthropometric indicators related to %BF and to validate regression equations to predict %BF in children and adolescents using dual-energy X-ray absorptiometry (DXA) as a reference method. Methods: a descriptive study (cross-sectional) was designed in 1126 schoolchildren (588 males and 538 females) from the Maule region (Chile). The age range ranged from 6.0 to 17.9 years. Weight, height, two skinfolds (tricipital and subscapular and waist circumference (WC) were evaluated. Body mass index (BMI), triponderal mass index (TMI), waist height index (WHtR) were calculated. Body fat percentage (%BF) was assessed by DXA scanning. Results: the relationships between ⊠ (Tricipital + Subscapular), TMI and WHtR with %BF (DXA) ranged from R2 = 52 % to 54 % in men, and from R2 = 41 % to 49 % in women. The equations generated for men were: %BF = 9.775 + [(0.415 * (Tr + SE)] + (35.084 * WHtR)-(0.828 * age), R2 = 70 %, and %BF = 20.720 + [(0.492 * (Tr + SE)] + (0.354 * TMI)-(0.923 * age), R2 = 68 %], and for women: %BF = 8.608 + [(0.291 * (Tr + SE)] + (38.893 * WHtR)-(0.176 * age), R2 = 60 %, and %BF = 16.087 + [(0.306 * (Tr + SE)] + (0.818 * TMI)-(0.300 * age), R2 = 59 %. Conclusion: this study showed that the sum of tricipital and subscapular skinfolds, IP and WHtR are adequate predictors of %BF. These indicators allowed the development of two regression equations acceptable in terms of precision and accuracy to predict %BF in children and adolescents of both sexes.