[DDI-users] SAS format library-> ddi?

Hoyle, Larry larryhoyle at ku.edu
Mon Oct 18 01:19:40 EDT 2010


Bob-
As Jeremy points out the range to label association  is not directly represented in DDI3.1, and as you know there would be no conversion to an enumeration of the range for a continuous variable. It is somethinng of a conceptual issue, so I expect that the end user tools will have to be reconfigired.
In the meantime I will be posting code for embedding the SAS CNTLIN dataset and metadata into a DDI file containing the data (and optionally the metadata) of its associated dataset.
 
I would also like to point out Phil Wright's SAS Global forum 2010 paper. He has developed code to go from the SPSS style independent labeling of each variable to the SAS/Stata style of reference to a shared format definition, for example a set of commonly coded Yes,No questions.

________________________________

From: ddi-users-bounces at icpsr.umich.edu on behalf of Jeremy Iverson
Sent: Sun 10/17/2010 10:50 PM
To: ddi-users at icpsr.umich.edu
Subject: Re: [DDI-users] SAS format library-> ddi?



Bob,

Unfortunately DDI 3.1 does not support assigning labels to value ranges.
Of course this is an important capability, so my understanding is that
you will see this addressed in the next version, 3.2. Tools support
should follow from there.

Cheers,

Jeremy


On 10/16/2010 7:43 AM, Bob McConnaughey wrote:
> Larry et al.
>     being more or less ignorant of DDI - ie I want to use the end
> products but am not sure how much i need to know about xml details to do
> so effectively, but being pretty familiar w/ SAS,  your code allows the
> SAS format libraries to be completely described in XML as they are w/in
> SAS.  That is it looks like you're allowing SAS ranges in the xml -> ddi
> transformation/compliance process.  Which is terrific as i've had to go
> though the steps of converting SAS ranges to "one-one" value labels when
> converting SAS dataset to SPSS in order to use the DDI based
> documentation tools that I've been trying out over the last few months.
>   Can the end user tools themselves (eg Nesstar or something cheaper...)
> be reconfigured to accept the equivalent of the SAS/Stata format ranges?
> or is that a "conceptual" problem that the documentation packages avoid
> on principle?
>
> eg..it would be very convenient and more parsimonious not to have to do
> the equivalent of SPSS recodes or excessively lengthy sets of value
> labels and instead have variables, say, education or age "formatted"
> into the groups one wants for documentation w/out losing the original
> values. I had been under the misapprehension (I hope) that DDI had a
> philosophical problem w/ "data ranges" - and am rather hoping that this
> was my misreading of how documentation packages had been implemented -
> that is w/out incorporating format ranges for convenience rather than by
> fiat.
>
> thanks very much indeed.
>
> bob mcconnaughey
>
> westat/NIEHS epidemiology support
>
>
> "the more I think I know, the less I see."
> John Lennon
>
>
>
>
>
> _______________________________________________
> DDI-users mailing list
> DDI-users at icpsr.umich.edu
> http://www.icpsr.umich.edu/mailman/listinfo/ddi-users
_______________________________________________
DDI-users mailing list
DDI-users at icpsr.umich.edu
http://www.icpsr.umich.edu/mailman/listinfo/ddi-users


-------------- next part --------------
A non-text attachment was scrubbed...
Name: not available
Type: application/ms-tnef
Size: 6194 bytes
Desc: not available
Url : http://www.icpsr.umich.edu/pipermail/ddi-users/attachments/20101018/688ea595/attachment.bin 


More information about the DDI-users mailing list