Different analysis methods for the objective measurements of body sway were examined to establish the most sensitive method for identifying changes in stability. Changes between stability with eyes open and eyes closed were used to assess the techniques. Body sway was measured in a group of normal subjects standing on a stable surface using "sway magnetometry," which measures hip movement in the sagittal and coronal planes by magnetic field detectors worn at the waist. The following analysis methods were studied; path length, mean distance from the centre, the area enclosed by the path of movement, and the areas of a convex hull, an ellipse and a circle enclosing the sampled points. The results showed that the path length technique had the smallest inter-subject variability, both with eyes open and eyes closed, and was the only technique whose mean Romberg Coefficient (Eyes Open/Eyes Closed) was more than two standard derivations from 1.0 (0.74 +/- 0.11, mean +/- S.D.), indicating that this technique showed greatest sensitivity for detecting changes in body sway.