This cross-sectional study examined electromyographic (EMG) and kinetic variables during reactive balance responses in children grouped according to developmental level as compared with chronological age. Purposes were to explore relationships between the two types of variables and the effectiveness of the two grouping methods. Forty-four children between 9 months and 10 years old were tested for reactive balance control on a moveable platform. Surface electrodes measured EMG activity in the gastrocnemius (GA), hamstrings (HA), paraspinals (PS), tibialis anterior (TA), quadriceps (QA), and abdominal (AB) muscles. Timing and distance of center-of-pressure (COP) movements and peak muscle torques at the ankle, knee, and hip were also examined. Significant relationships and group differences were found between postural muscle activity and both the torque generated in the lower limbs and the timing and distance of COP adjustments employed to restabilize balance. As postural muscle activity increased and became more coordinated in timing, peak torque at the ankle and hip also increased, while the distance of and time to complete COP readjustments decreased. Children in younger/developmentally lower groups had smaller-magnitude and less-synergic muscle activity, lower peak torques, longer times to restabilize the COP, and greater COP paths than older/higher developmental groupings. Grouping by developmental level produced more statistical differences than did grouping by age. The correspondence of GA, HA, and PS muscle activity with COP measures and joint peak torques confirms that these muscles are key contributors to the balance synergy correcting for induced forward sway. Additionally, developmental level appears to be a much better predictor of balance improvement than chronological age.