Each NAEP data product contains sets of control statement files for converting each response data file into a SAS, SPSS, or STATA (2003 and later) system data file.
Each of the control statement files contains separate sections for variable definition, variable labeling, missing value declaration, value labeling, and creation of scored variables from the cognitive items. The variable definition section describes the locations of the fields, by name, in the file, and, if applicable, the number of decimal places or type of data. The variable label identifies each field with a 50-character description. (Note: SAS has a limit of 40 characters for variable labels; some editing of the labels in the syntax file may be necessary to retain important information in the last 10 characters.)The missing value section identifies values of those variables that are to be treated as missing and excluded from analyses. The value label section specifies the code values and their descriptors for each discrete numeric variable. The scoring section is provided to permit the user to generate item score variables instead of the item response variables.
A separate file is included among the SAS control statement files that uses the SAS PROC FORMAT procedure to generate value labels for the discrete numeric variables across all response data files. This file must be executed in SAS before any other SAS control statement file is processed.