Angular Transformation


Menu location: Data_Transforming and Deriving_Common Transforms_Angular.


Angular transformation can be used to linearize sigmoid distributions and equalize variances of proportions (Bland, 2000). The angle is defined as arcsin (sqr(p)), where sqr(p) is the square root of a proportion. Logit and probit transformations do not equalize variances but logistic regression and probit analysis do give special treatment to variances in the way that they handle expected error distributions. Angular regression is seldom used but angle transformed data can be useful for exploratory data analysis by experts.


You can supply proportions or discrete data for angular transformation. If you specify discrete data then StatsDirect converts these to proportions by taking each value as a proportion of the maximum of the supplied data. The results are stored in a new column that is marked Angle:<name> where <name> is the original column label.


Indeterminable values are marked as missing data.