data transformation (statistics)

In statistics, data transformation is carried in order to transform the data and assure that it has a normal distribution (a remedy for outliers, failures of normality, linearity, and homoscedasticity; this is usually done to prepare data for regression analysis, as this analysis assumes the data is linear, normal and homoscedastic.). This is also known as transformation to linearity. A good indicator of data having a normal distribution is skewness in the range of −0.8 to 0.8 and  kurtosis in range of −3.0 to 3.0.
See more at Wikipedia.org...