ABSTRACT
Normalization is a process to remove the effect of a variable into another to decrease its variability. Normalization is also a process to scale a variable. For gait analysis, the most common normalization is to divide Ground Reaction Force by body weight. The objective of this paper is to analyze the effects of normalizalizing a set of ground reaction force parameters. Thirteen adults were the subjects of this study. A force plataform was used to measure the vertical component of ground reaction (vGRF) during stance phase. Twenty samples for each subject were obtained (10 for walking with and other 10 for walking barefoot). For each condition, 5 samples were due to the left support and other 5 were due to the rigth support. Some parameters were calculated using vGRF: intensity, impulse and inclination. The raw signal was filtered with low-pass, 2nd order, zero lag, Butterworth filter at 100 Hz. For statistical analysis, different produres were used: to compare groups, we used two-way analysis of variance; to study the effect of normalization into the signal's variability, we used the principal component analysis. The main results suggest that normalization affects those paraeters, and it is not affected by shoes condiction, chaging the metric among them. However, this is not a pure lenear effect and probably it is also affected by a covariable. We suggest that time is one covariate. We also suggest that other experiments should be run to analyze such effect, considering as covariables gait speed, time, shoes and frequency components.