Category variables are non-numeric descriptive variables that describe a category that objects may belong to. They can be used in multivariate analysis in regression models after converting them to dummy variables. Category variables are also useful in data visualization and for validating models across categorical levels such as sample origin, treatment etc. They can be used to color code samples in plots and when used in a score plot, help to accentuate the visualization of groupings that may exist in the data based on a group.

Category variables can be added to data in The Unscrambler® by converting an existing data type to category. This is done by selecting a data column and then using a right mouse click to access the menu from which you can choose Change Data Type -> Category. This will open the Category Converter dialog where the categories can be defined based on the individual values in the selected column of data, or by creating levels based on the range of values in that column. The category names can be designated if the radio button Specify each range manually is selected.

Category Converter

Another way to add category variables to data is to add a new column of data as type category by selecting a column in the data matrix and using Edit->Insert Category Variable which opens the category variable dialog that allows categories to be added or created from existing row sets in the data.