Why Use Z-Scores for Motion Data Analysis?

[jophysics]

Hi,

I would like to know something about the possibility to convert data set into z-scores to ensure a common scale for analysis. More specifically, when this conversion is needed and why. I am working on motion data (from 3 persons) tracked from video and I can use a common reference for the whole data set or a reference for each single person. I found the z-score conversion in a paper, but I cannot understand when it is really needed.

thank you


jo

 

[Mark44]

The basic assumption in using z-scores is that the attribute you're measuring is normally distributed. The z-score allows you to transform values that have a mean of mu and a standard deviation of sigma to new values with mean 0 and standard deviation 1. The reason for doing this is that there are widely published tables with probabilities for z-scores, while such tables for the raw x-scores can be found rarely, if at all.

Hope that helps.