DataLab is a compact statistics package aiming at exploratory data analysis. Please visit the DataLab Web site for more information....

Home Features of DataLab Editing Data Data Editing Nominal and Ordinal Values
See also: Variable and Object Names, Matrix Survey Plot 

Editing Nominal and Ordinal Values

DataLab supports four different levels of measurement: nominal, ordinal, interval, and ratio-level variables. The type of a variable (column of the data matrix) can be set by the user in the matrix viewer. For nominal and ordinal variables the user may additionally specify the corresponding categorical names(1) (again, in the matrix viewer, by clicking the star-button right to the level of measurement).

DataLab supports up to 150 nominal/ordinal identifiers for each variable. When you click the star-button right to the level of measurement a window is opened which lets you enter the nominal or ordinal identifiers. You can edit or add an identifier by clicking into the corresponding field and entering the new string. In order to undefine an identifier, just clear it and leave it empty. Undefined identifiers are indicated by the string "(undefined)".

Hint: Categorical assignments can be re-sorted by "drag & drop". The re-sorting changes the assignment between identifiers and ordinal numbers (the data stay the same, only the corresponding identifiers change). Thus re-sorting should be performed only either before entering the data, or if the current assigments are wrong.


(1) If no identifiers are specified the missing identifiers are indicated by their ordinals in parantheses. Thus parantheses in the scale labels always point to missing identifiers.

Example: In the figure at the right you see correctly specified identifiers in the top diagram. The diagram below shows the results if the identifiers for the properties "dark brown", "brunette", and "brown" are undefined. The scale labels are showing the corresponding ordinals in parantheses.