Normalization is a transformation that adjusts the values in one or more columns of a data table to a common scale, such as between 0 and 1, or with a mean of 0 and a standard deviation of 1. Normalization can be useful for comparing data from different sources, removing outliers, or preparing data for further analysis. Spotfire provides several methods for normalizing data, which are briefly described below1:
* Z-score calculation: This method transforms the values in each column by subtracting the mean and dividing by the standard deviation. The resulting values have a mean of 0 and a standard deviation of 1, and are also known as standard scores or z-scores. This method is useful for comparing data that have different units or scales, or for identifying outliers that deviate from the mean by more than a certain number of standard deviations.
* Normalize by standard deviation: This method transforms the values in each column by dividing by the standard deviation. The resulting values have a standard deviation of 1, but the mean is not changed.
This method is useful for comparing the variability or dispersion of data across columns, or for reducing the effect of outliers on the mean.
* Normalize by mean: This method transforms the values in each column by dividing by the mean. The resulting values have a mean of 1, but the standard deviation is not changed. This method is useful for comparing the relative magnitude or proportion of data across columns, or for scaling the data to a common unit.
* Normalize by mode: This method transforms the values in each column by dividing by the mode, which is the most frequent value. The resulting values have a mode of 1, but the mean and standard deviation are not changed. This method is useful for comparing the frequency or popularity of data across columns, or for scaling the data to a common unit.
* Normalize by median: This method transforms the values in each column by dividing by the median, which is the middle value when the data are sorted. The resulting values have a median of 1, but the mean and standard deviation are not changed. This method is useful for comparing the central tendency or location of data across columns, or for scaling the data to a common unit.
To add a normalization transformation to a data table, either when loading a new data table or after the data is already in Spotfire, follow the steps described in the references23. References:
* Details on Normalization
* Spotfire Tips & Tricks: Normalize/Standardize your data with Spotfire
* How to Use the Map Chart