Measurement Level (TREE command)

Optionally, a measurement level can be specified in square brackets following any model variable. The measurement level can be defined as scale ([S]), ordinal ([O]), or nominal ([N]). Categorical ([C]) is accepted as a synonym for nominal.

• If a measurement level is specified, it temporarily overrides the setting recorded in the data dictionary. If a measurement level is not specified for a variable, the dictionary setting is used.
• If a measurement level is not specified and no setting is recorded in the data dictionary, a numeric variable is treated as scale and a string variable is treated as nominal.
• If a string variable is defined as scale, a warning is issued and the variable is treated as nominal.
• A measurement level applies only to the variable that immediately precedes it in the variable list. For example, age income [S] and age TO income [S] assign the scale level of measurement to income only.
• The keyword ALL followed by a measurement level specification will apply that measurement level to all independent variables.
• If a variable appears more than once in the independent variable list and the measurement level varies across instances, the measurement level specified for the last instance of the variable is used. For example, if ALL [S] z [N] is specified, z is treated as a nominal predictor.