Bioelectrical impedance analysis (BIA), a device that analyzes the current conduction differences between the fat and water components is widely used for reasons that include convenience of use, non-invasiveness, safety, and low cost. Dual energy X-ray absorptiometry (DXA) allows for the assessment of total body and regional lean and fat tissues and bone mineral content (BMC). The objective of this study was to compare body composition assessments by BIA and DXA before and after a 6-week herbal diet intervention program in 50 pre-menopausal women [mean +/- SD: age 30.58 +/- 6.15, body mass index (BMI) 31.72 +/- 3.78]. Waist-to-hip ratio (WHR) was measured by BIA and anthropometry. Lean body mass (LBM), body fat (BF), BMC and percent body fat (%BF) were measured by BIA and DXA. Highly significant correlations were observed between BIA and DXA measurements for LBM, BF, BMC and %BF (r = 0.73, 0.93, 0.53, 0.79, respectively) before the intervention. Differences between BIA and DXA measurements were observed in LBM, BF, %BF and BMC before intervention (p < 0.01) where WHR by BIA was significantly higher compared to anthropometry before (p < 0.01) and after the intervention (p < 0.01). BIA underestimated LBM by 1.85 kg and overestimated BF by 2.54 kg compared to DXA before the intervention. Although BIA and DXA showed highly significant correlations for LBM, BF, BMC and %BF before the intervention, they did not produce statistically comparable results in pre-menopausal Korean women and therefore should not be used interchangeably when measuring body composition.