This study examined whether differences in waist circumference (WC) and pedometer placement (anterior vs. midaxillary vs. posterior) affect the agreement between pedometer and observed steps during treadmill and self-paced walking. Participants included 19 pairs of youth (9-15 years old) who were matched for sex, race, and height and stratified by WC (high WC: HWC; low WC: LWC). Participants performed 3-min treadmill-walking trials at speeds of 59, 72, and 86 m multiplied by min[superscript -1] and a 400-m self-paced walking trial on level ground. Bland-Altman plots were used to assess the agreement between pedometer and observed steps of spring-levered pedometers by WC, pedometer placement, and walking speed. In the HWC group, the posterior pedometer placement consistently agreed most closely with observed steps at all treadmill speeds and during self-paced walking. In the LWC group, no single pedometer placement consistently agreed most closely with observed steps at all treadmill speeds and during self-paced walking. We conclude that a posterior pedometer placement improves step-count accuracy in most youth with an HWC at a range of walking speeds on level ground. (Contains 1 table and 5 figures.)