Introduction to Scoring, Standardization, and Ranking Procedures

 

Overview: Scoring, Standardization, and Ranking Procedures

Several SAS/STAT procedures are utilities that produce an output data set with new variables that are transformations of data in the input data set. SAS/STAT software includes four of these procedures. The RANK procedure produces rank scores across observations, the SCORE procedure constructs functions across the variables, and the STANDARD and STDIZE procedures transform each variable individually.

RANK

ranks the observations of each numeric variable and outputs ranks or rank scores. For a complete discussion of the RANK procedure, see the Base SAS Procedures Guide: Statistical Procedures.

SCORE

constructs new variables that are linear combinations of old variables according to a scoring data set. This procedure is used with the FACTOR procedure and other procedures that output scoring coefficients.

STANDARD

standardizes variables to a given mean and standard deviation. For a complete discussion of PROC STANDARD, see the Base SAS Procedures Guide: Statistical Procedures.

STDIZE

standardizes variables by subtracting a location measure and dividing by a scale measure. A variety of location and scale measures are provided. Such measures include the mean, median, Huber’s estimate, Tukey’s biweight estimate, and Andrew’s wave estimate.