SCORES Subcommand (TREE command)

For CHAID or Exhaustive CHAID trees with ordinal dependent variables, the SCORES subcommand allows you to specify scores that are used to scale the dependent variable. By default, scores are assigned in sorted value order: the first category is assigned a score of 1, the second category is assigned a score of 2, and so forth.

  • If the growing method is CRT or QUEST, this subcommand is ignored, and a warning is issued.
  • If the dependent variable is nominal or scale, this subcommand is ignored, and a warning is issued.

The subcommand name is followed by an equals sign (=) and one of the two following alternatives:

EQUALINCREMENTS. Scores are ordinal ranks. The lowest category of the dependent variable is assigned a score of 1, the next highest category is assigned a score of 2, and so on. This is the default.

CUSTOM. User-specified scores. For each category of the dependent variable, specify the category value followed by a score in square brackets.

  • Scores must be numeric and unique. You cannot assign the same score to more than one category.
  • Scores must be specified for all values of the dependent variable included in the analysis (either all non-missing values found in the data or all values defined on the DEPCATEGORIES subcommand).
  • Category values must be consistent with the data type of the dependent variable. String and date values must be quoted. Date values must be consistent with the variable’s print format.
  • If you specify multiple score values for the same category value, the last one is used.
  • A warning is issued if you specify a score for a category that does not exist in the data or in the training sample if split-sample validation is in effect. See the topic VALIDATION Subcommand (TREE command) for more information.

Example

TREE income [o] BY age employment gender region
 /SCORES CUSTOM=1 [1] 2 [2.5] 3 [8].
  • By default, the CHAID method is used to grow the tree.
  • CUSTOM contains one score specification for each category of the ordinal dependent variable.
  • Category 1 is assigned a score of 1; category 2 is assigned a score of 2.5; and category 3 is assigned a score of 8. In other words, scores are used to rescale income so there is a smaller gap between categories 1 and 2 than between 2 and 3.