[DDI-SRG] NCube Attribute question

[Joachim Wackerow]

Sorry, when I was not clear enough.

Some of the data cells have a count below a limit, where data items 
could not be published; in the example this belongs to data cells below 
6. These data cells will be set to the mean of the limit (here 3), so 
the totals get not biased to one side. I named these cells "closed", the 
details are not open to the public, just the mean (here 3) is shown. 
Each of these cells should get an indicator, that the number doesn't 
represent the true data.

My intention was to use Attribute in
/nci:RecordLayout/nci:NCubeInstance/nci:DataItem/nci:Attribute

This seems to be the right place, because the note is related to a 
specific data cell. There is no structural reason for this note, which 
can be described in the logical product.

Example:

nci:DataItem>
   <nci:Dimension value="2" rank="1"/>
   <nci:Dimension value="1" rank="2"/>
   <nci:Attribute>
     <nci:AttributeReference>
       <ru:ID>DisclosureControlNote</ru:ID>
     </nci:AttributeReference>
     <nci:Value>anonymized</nci:Value>
   </nci:Attribute>
   <nci:Measure>
   <nci:MeasureReference>
     <ru:ID>Measure_N</ru:ID>

The field-level documentation says
AttributeReference: Reference to attribute description
Attribute: Provides for a choice between an explicit value, or a 
location in a file where the value can be found.

But both fields are mandatory, there is no choice. This seems to be a 
bug. There should be a choice. The Value is probably the value of a 
related variable. But in my case the AttributeReference to an attribute 
description seems to be the right place.

Anyway my original question was where to store the attribute 
description. A regular note doesn't seem to be the right place. Then we 
would have some circular reference. The attribute points to the note and 
the note points to the attribute.

Did I overlook something?

Achim

Wendy Thomas wrote:
> IF this is a default value occuring in the same location(s) in each 
> potential instance of the NCube
> In
> <NCubeLogicalProduct>
>   <NCubeScheme>
>     <NCube>
>      <Attribute isIdentifiable>
>        <AttachmentValue>
> 
> If it is a potential value (e.g. suppression code) then it is an 
> alternate measure described by a variable with a coded value and label. 
> Where the measured value say count is blank or simply not there and the 
> measure suppression flag has a value of the appropriate code.
> 
> Actually without seeing your example I'm just speculating possible 
> scenarios.
> 
> Wendy
> 
> 
> 
> 
> On Sat, 31 Jan 2009, Joachim Wackerow wrote:
> 
>> I'm using the Attribute in inline ncube. Now I'm wondering where to
>> store the attribute description to which the AttributeReference is
>> pointing to. The intention is to use Attribute like a footnote with a
>> descriptive text for closed data items. This is not related to any
>> variable/dimension.
>>
>> Achim
>>
>> -- 
>> GESIS - Leibniz Institute for the Social Sciences
>> Postal address: P.O. Box 122155, 68072 Mannheim, Germany
>> Visiting address: B2 1, 68159 Mannheim, Germany
>> Phone: +49 (0)621 1246 262
>> Fax: +49 (0)621 1246 100
>> E-mail: joachim.wackerow at gesis.org
>> www.gesis.org/en/institute/
>> _______________________________________________
>> DDI-SRG mailing list
>> DDI-SRG at icpsr.umich.edu
>> http://www.icpsr.umich.edu/mailman/listinfo/ddi-srg
>>
> 
> Wendy L. Thomas                          Phone: +1 612.624.4389
> Data Access Core Director         Fax:   +1 612.626.8375
> Minnesota Population Center              Email: wlt at pop.umn.edu
> University of Minnesota
> 50 Willey Hall
> 225 19th Avenue South
> Minneapolis, MN 55455


-- 
GESIS - Leibniz Institute for the Social Sciences
Postal address: P.O. Box 122155, 68072 Mannheim, Germany
Visiting address: B2 1, 68159 Mannheim, Germany
Phone: +49 (0)621 1246 262
Fax: +49 (0)621 1246 100
E-mail: joachim.wackerow at gesis.org
www.gesis.org/en/institute/