[R] read.spss question warning compression bias

[Hedderik van Rijn]

> So it would appear that if the above is correct, there is no user
> adjustment to the bias value. The only scenario that I can envision is
> if the user SAVE's the ".sav" file in an uncompressed format, where the
> bias value **might** be set to 0.
>
> Perhaps a r-help reader with access to current SPSS manuals can confirm
> the above.

See below for a copy and pasted text from the PDF Manuals for SPSS v11 
for Mac OS X, pages 1106/1107 in the file: [...] /SPSS\ 
11/SyntaxGuide/SPSSBase.pdf

Note that it doesn't tell you anymore whether the default is 
"compressed" or "uncompressed". The save dialog box also doesn't given 
any options, apart from selecting which type of .SAV file one wants 
(SPSS, SPSS v7, SPSS/PC+, SPSS Portable, and a load of other non-SPSS 
formats like tab delimited, Excel and even dBase II).

  - Hedderik.

SAVE
SAVE OUTFILE=file
[/VERSION={3**}]
{2 }
[/UNSELECTED=[{RETAIN}]
{DELETE}
[/KEEP={ALL** }] [/DROP=varlist]
{varlist}
[/RENAME=(old varlist=new varlist)...]
[/MAP] [/{COMPRESSED }]
{UNCOMPRESSED}
**Default if the subcommand is omitted.

Example
SAVE OUTFILE=EMPL /RENAME=(AGE=AGE88) (JOBCAT=JOBCAT88).

Overview
SAVE produces an SPSS-format data file. An SPSS-format data file 
contains data plus a dictionary.
The dictionary contains a name for each variable in the data file plus 
any assigned
variable and value labels, missing-value flags, and variable print and 
write formats. The dictionary
also contains document text created with the DOCUMENTS command.
XSAVE also creates SPSS-format data files. The difference is that SAVE 
causes data to be
read, while XSAVE is not executed until data are read for the next 
procedure.
See SAVE TRANSLATE and SAVE SCSS for information on saving data files 
that can be
used by other programs.

Options
Compatibility with Early Releases. You can save a data file that can be 
read by SPSS releases
prior to 7.5.
Variable Subsets and Order. You can save a subset of variables and 
reorder the variables that
are saved using the DROP and KEEP subcommands.
Variable Names. You can rename variables as they are copied into the 
SPSS-format data file
using the RENAME subcommand.
Variable Map. To confirm the names and order of the variables saved in 
the SPSS-format
data file, use the MAP subcommand. MAP displays the variables saved in 
the SPSS-format
data file next to their corresponding names in the working data file.

Data Compression. You can write the data file in compressed or 
uncompressed form using the
COMPRESSED or UNCOMPRESSED subcommand.

Basic Specification
The basic specification is the OUTFILE subcommand, which specifies a 
name for the SPSSformat
data file to be saved.

Subcommand Order
· Subcommands can be specified in any order.

Syntax Rules
· OUTFILE is required and can be specified only once. If OUTFILE is 
specified more than
once, only the last OUTFILE specified is in effect.
· KEEP, DROP, RENAME, and MAP can each be used as many times as needed.
· Only one of the subcommands COMPRESSED or UNCOMPRESSED can be 
specified per
SAVE command.

Operations
· SAVE is executed immediately and causes the data to be read.
· The new SPSS-format data file dictionary is arranged in the same 
order as the working file
dictionary, unless variables are reordered with the KEEP subcommand. 
Documentary text
from the working file dictionary is always saved unless it is dropped 
with the DROP DOCUMENTS
command before SAVE.
· New variables created by transformations and procedures previous to 
the SAVE command
are included in the new SPSS-format data file, and variables altered by 
transformations
are saved in their modified form. Results of any temporary 
transformations immediately
preceding the SAVE command are included in the file; scratch variables 
are not.
· SPSS-format data files are binary files designed to be read and 
written by SPSS only.
SPSS-format data files can be edited only with the UPDATE command. Use 
the MATCH
FILES and ADD FILES commands to merge SPSS-format data files.
· The working data file is still available for transformations and 
procedures after SAVE is
executed.
· SAVE processes the dictionary first and displays a message that 
indicates how many variables
will be saved. Once the data are written, SAVE indicates how many cases 
were saved.
If the second message does not appear, the file was probably not 
completely written.