Files created by PREPARE
The PREPARE module produces some files automatically and some when you request them. When you begin a new questionnaire, you specify a study name in the header statement that PREPARE will use to name the files it creates. The file names take the form of your study name plus a suffix or extension. Here are some of the files created and their extensions for study name DEMO:
| Description | Extension | |
| Default Files | Compiled Questionnaire | DEMO.QFF |
| Quota/CaseID | DEMO.QUO | |
| Files can be Suppressed |
Formatted Source/Backup Qnaire spec file | DEMO.QSP |
| Ordered Variable list with data location information | DEMO.CHK | |
| Variables data base | DEMO.DB | |
| Question summary list | DEMO.SUM | |
| Optional Files | Formatted Questionnaire with labels, data locations, etc | DEMO.HRD |
| Data validation and cleaning specifications | DEMO.CLN | |
| Crosstab definitions | DEMO.DEF |
NOTES:
- If a question has no label, information about it will be stored using the program-assigned label preceded by CFM (i.e., CFM0023).
- If the questionnaire has errors, only the QSP file will be created.
- Files not noted below as ASCII are not directly readable (i.e., by an editor); they are meant to be accessed by a Survox program.
Compiled Questionnaire File and Quota File (QFF)
PREPARE creates a compiled questionnaire and quota file to be used for the actual interview with the Survent program. The QFF file contains all the screen text, formatting commands and logic conditions needed to collect interview data. If necessary, you can generate any of the other informational files from the QFF file. (See the end of COMPOSING AND COMPILING QUESTIONS IN PREPARE for more details.)
Quota File (QUO)
PREPARE creates a file to keep track of quota values and counters even if no QUOTA statements or functions are present in your specs. This file can be accessed using the QUOTAMOD utility and the Supervisor’s QSS command, and can be suppressed using the -QUOTA_FILE header option. Supervised and shared-file Survent require a quota file. Standalone Survent can be run without a quota file.
Backup Questionnaire Specification File (QSP)
PREPARE creates a backup specification file when a questionnaire is compiled. This is an ASCII record of all the questionnaire specifications. You can edit it with a text editor and reprocess it by typing its name at the spec file prompt, or load it into SCRIPT COMPOSER to continue modifying your specifications. The specification file, or portions of it, can also be accessed for creating new questions. If you wish to edit this file and use it as input to PREPARE, you must rename it with an extension other than QSP.
This backup file will have all repeats expanded so that each question is displayed individually. A comment is generated with the question number and data location for each question. Comments (text after leading ‘s or ‘ ‘s) and {!COMMENT} blocks are retained.
The file will have data locations hard coded if you used the HARD_CODE compiler directive. You may then edit this file to make additional changes and guarantee that locations you let the program assign do not get moved when adding or removing questions (See MAKING CHANGES TO A LIVE STUDY).
Ordered Data List File (CHK)
PREPARE prints a map of column locations used, corresponding question numbers with types, and other uses of the location. This lets you check to make sure the same column was not inappropriately assigned to more than one question. This map is sorted in column order. This file can be suppressed using the -CHECK_FILE header option.
The CHK file lists:
- Case ID field
- Record ID and content fields (if CARD_FORMAT is used)
- Data column location and width
- Question label
- Question number
- Question type and subtype
- Other use of same data location
- Related labels referred to
- Gaps of columns
- Total case length
- Text area starting location
- Start of loop iterations
NOTE: All data modifying questions are listed. Any question type with a subtype A will not be listed because that question does not modify data. PHONE,GET_PHONE_INFORMATION and PHONE,PUT_PHONE_TEXT statements that show where data is referenced in the phone file will also be listed.
EXAMPLE:
Variables File (DB)
The DB file is created to store compiled variables for Mentor and the utilities to use. You will get a DB file unless you specify -DB_FILE on your header statement or compile with ~PREPARE COMPILE -SPEC_FILES. PREPARE will create the file using default specifications unless you have a CREATE_DB or DB_SIZES meta command before your header statement. PREPARE and Survent do not use the DB file.
Summary Question List File (SUM)
PREPARE creates an ASCII question list file that can be useful for checking the contents of a questionnaire file. This is a simplified listing which contains a one-line description for each question in question order. This file can be suppressed using the -SUMMARY_FILE header option.
Each line of the SUM file lists:
- Label
- Question number
- Data column location and width
- Question type and subtype
- The first n characters of the question text.
- N is determined by the SPEC_WIDTH setting on your study header (i.e., the text starts in column 39 so if your SPEC_WIDTH is 132, you’ll get up to 93 characters of text).
- This file will print up to position 80 by default and may be printed up to 5,000 characters wide, depending on the SPEC_WIDTH setting.
EXAMPLE:
It will also list columns used by ROTATE questions and the beginning and end of SUSPEND, RESUME, ROTATE and SPECIAL blocks.
Hardcopy Question Listing File (HRD)
PREPARE creates a hardcopy listing of the questionnaire when you use the compiler command HARD_COPY. You can use various options to control how the hardcopy is presented. If you have multiple languages in the questionnaire, you may specify which language to use in the hardcopy listing. This file could be used to do paper and pencil data collection or as an easy-to-read reference for a client. Syntax is converted to readable language wherever possible.
EXAMPLE:
July 13, 2015 - Exam1 - EXAMPLE Questionnaire - Page 1 Question: Q_COUNTRY (MAX 2 RESPONSES) (50.4) 01 ...........United Kingdom 02............France 03............Germany 04............Spain
In addition, the .HRD file is made automatically each time you compile. The “hardcopy” file is an easy-to-read description of the questionnaire with syntax removed and replaced by readable English. It gets created (<study>.hrd) each time you compile whether you have a HARDCOPY command or not. You can use the {!-hardcopy} command to turn off the hardcopy output for certain sections of the questionnaire or use {!hardcopy <options>} to change the looks of the hardcopy file. You can also use ”~prepare compile -hrd” to negate the creation of a .hrd file., if you choose.
Data Cleaning, Variable Definition, Table-Building Specs (CLN, DEF, TAB)
If you also have the Mentor software, you can create cleaning and table specs during the compile. The CLN file, in combination with the CLEAN.SPX model spec file, will let you clean the data by the questionnaire’s logic and rules. The DEF and TAB files, in combination with the TABS.SPX model spec file, will let you run tables automatically without you having to write the specs. (See your MENTOR manual for more details. See COMPOSING AND COMPILING QUESTIONS IN PREPARE for how to create these files.) The DEF file is also where other program table definitions are stored if specified on the COMPILE line. This includes SPSS, SAS, and SSS_XML specifications.
Renaming Files
PREPARE files are not automatically overwritten. If one exists when a new one with the same name is created, the old one is renamed by changing the first letter to the next ASCII character (DEMO.DB, for EXAMPLE, would be renamed EEMO.DB).
To replace rather than rename files, the PURGE_SAME meta command must be invoked. Be careful not to lose needed files using this command. The QUO file and the TR file are not renamed or purged by recompiling. The List File is not affected by a PURGE_SAME command –– use a dash before the name to get the same effect, eg. “-listfile”.