ARCHIVE DOCUMENTATION

STUDY NO: 4379

STUDY TITLE: Scottish Health Survey, 1998


Details of Dataset received:
----------------------------
 
Format, software, media:
-------------------------
Data and documentation received from the National Centre for
Social Research (Natcen), in June 2001.
Data - SPSS version 10.0.5.
Documentation - Word 97 and paper format.


Validation and conversion:
--------------------------

Content validation:
------------------- 
All files checked for accuracy and completeness.
 
Conversion process:
-------------------
DATA
SPSS .sav files converted to SPSS portable format using SPSS 
10.0.5 for Windows - resulting files shs98h.por and shs98i.por.

DOCUMENTATION
Word files were converted to Adobe Acrobat PDF format using 
Acrobat Exchange 3.0. Isolated pages containing diagrams, etc. 
were scanned from paper copies of documentation instructions and
added to the PDF files. All files were combined into a 4-volume 
User Guide, and bookmarks added throughout for ease of navigation. 

N.B. At the request of the depositor, the PDF User Guide has been
compiled in a specific order.
 

Conversion validation:
---------------------- 
DATA
List, descriptive and frequency files created and checked.

Summary of value-added outputs:
------------------------------- 
SPSS portable files created.
PDF User Guide created.
Variable and value labels edited where necessary for clarity.


Constraints on the reliability and/or usability of the data:
------------------------------------------------------------
N.B. The interviewers' instructions in the documentation
(file a4379udb.pdf, p.59) contain an extra sheet detailing
'Supplementary Instructions for Alternative Measuring
Equipment' for height measurements. This is not contained
in the printed version of the file supplied at the time of
deposit.

See also below.


Any outstanding problems or queries: 
------------------------------------
The grouped carstairs deprivation variables are mislabelled. The current
label for carstg5 gives 1 as 'most deprived' and 5 as 'least deprived'.
carstg4 are similarly labelled as 1 'most deprived' and 4 ''least deprived' .
These labels should be reversed - i.e. 1 is least deprived.   

Technical report recommended/essential:
---------------------------------------
Tech report available via the web from the SHOW website
(http://www.show.scot.nhs.uk/scottishhealthsurvey/index.htm).
This URL is a better direct link than that given in the
User Guide.

Full bibliographic details for the published tech report are 
as follows:


Notes, useful tips and hints:
-----------------------------
The weight variable WEIGHTA should be set ON for all analysis.

The individual and household files may be linked using variable
ARCHSN (Individual Serial Number) as the 'match by' variable in
SPSS.


User comments and post-order corrections:
-----------------------------------------