All statistical programs make mistakes in typing some variables. R++ has a type viewer so that you’re not stuck with wrong parameters during analysis.
In the viewer all the types are matched with a colour:
- Turquoise for numerical variables
- Violet for nominal variables
- Blue for ordered variables
- Yellow for logical variables
- Green for integer variables
When the type is not correct you can go to the configuration menu and change it in a simple click.
I this example R++ thinks the French department number are integer variables. It’s meaningless, you can’t average departments out (e.g. 31 = Haute-Garonne, 91 = Essonne). To fix it, just pick “Nominal” in “Department” column.
Sometimes a type is not correct because some of its
levels are not correct.
For instance when there is “zero” instead of “0” the program recognises a nominal variable. You can change the type with the integer button. R++ highlights the incoherent levels in red at the top of the column. Hence you can spot and fix quickly them.
Statistical programs always recognise ordered variables as nominal variables. With R++ you can see the levels, drag & drop to order them, click the “ordered” button and confirm. The changes are made!
Once you’ve changed the format for a variable you can duplicate it to other variables with a copy and paste.
As always, all the actions are available in the code menu.