This document describes the layout of the SAT-9 Score Summary file, which is produced during the scoring of the SAT, at the school's request.

The data file is sorted by last name. It consists of the identifying information and test level, followed by the scores for each student  14 subtest sets, each containing 7 items (scores, percentages, etc). Neither the 130 cluster raw scores, nor the item response grid (480 items) are included in this file.

All dates are in the format MM/DD/YYYY. Record number is just a sequence number that makes several database systems happy. (It can be used as a unique, primary key for database systems.) Batch, state code, reporting source code and ODS number are numbers that we sometimes use internally for filing purposes. They won't have any meaning for the average user. However, I've included them in case users need to call this office with questions. Building Code is a 3-digit number from the header sheet, that can be used to break schools into classrooms, teachers, or other classifications.

Grade Equivalents can be any of the following:

Definitions of some of the terms used above can be found in the GRI SAT-9 Glossary of terms as well as in the HEM Glossary of Measurement Terms at <http://www.hemweb.com/library/glossary.htm>.

Hearing Impaired Percentile Rank values normally range from 0 to 99. The only exception is a value of 999 which indicates that the subtest has not been normed for that age. Similarly, Percent Right values are between 0 and 100, except when a particular subtest does not have a valid score  either because it was not given or it does not exist at that level of the test  in which case the Raw Score will be between 990 and 999, and the Percent Right will be 999.

Three copies of the data file are provided, each in a different format. Also included are two copies of the layout description  one being the file you're reading now (README.HTML) designed for use with a web browser (e.g. Netscape), and the other being an ASCII version (README.TXT), suitable for printing on a dumb printer or perusing on the screen with a paging program such as more.

The filename ending with .POR is an SPSS portable file, and is the most complete, since it incorporates the field names, sizes, and data types into the data file, freeing the user from the need to specify these. Unfortunately, without some version of SPSS, the file is probably useless. Although the file is technically ASCII, it is not organized into any kind of human-readable format.

The filename ending in .CSV is a comma separated value file, which can be imported into a wide variety of programs. Each field value is separated from the next by a comma  hence the file type CSV. Fields containing text are surrounded by double quotes. The first line of the file contains the field names. The remaining lines contain the actual data.

The filename ending with .ASC is a fixed-width ASCII file. Each field has a fixed position and length. Short alpha (text) fields are padded with blanks and left-justified; numeric fields are padded with zeros and right-justified. (For example, the Last Name field is 11 characters long. So a value of "SMITH" will be stored as "SMITH      ", and a value of "1" for a 3-digit field becomes "001".) The record length for this file is 404 bytes/record. The description of each field, and its position and length appear in the table at the end of this document, but for those of you who understand the FORTRAN FORMAT statement, here's a quick layout:

(I6,2(I2,'/'),I4,I6,I2,2I3,2(I2,'/'),I4,A4,A10,A11,
A9,A1,2(I2,'/'),I4,2I2,I3,2A1,I2,14(5I3,A4,I3))

Here's the detailed layout:

If you have any questions regarding these files, please contact me at the address below: