[DDI-CDG] Fwd: Re: [DDI-SRG] Longitudinal data proposal

[Sanda Ionescu]

>
>Wendy,
>
>Sure, the respondent can change over time, but I've actually seen more 
>than one respondent on the same record (same dataset, same year, etc.)
>I have already replied to I-Lin, here's what I wrote to him (sorry I did 
>not realize he had copied all these groups, so I should have replied to all):
>
>
>>Date: Thu, 15 Sep 2005 08:27:15 -0400
>>To: ikuoikuo at gmail.com
>>From: Sanda Ionescu <sandai at icpsr.umich.edu>
>>Subject: Re: [DDI-SRG] Longitudinal data proposal
>>Bcc: maryv at icpsr.umich.edu
>>
>>Hi, I-Lin.
>>
>>No, respondent too is "variable-wide", to put it in your words.
>>We can have on the same record questions that were asked about, say, 
>>"person X"  both FROM
>>  person X herself and, for example, a relative or somebody else who 
>> knows this person or even the interviewer.
>>That is not even an issue as far as my proposal is concerned because we 
>>have a child-element of variable right now, in V 2.0, that covers for 
>>this type of info -it's
>>respUnit (def: Provides information regarding who provided the 
>>information contained within the variable, e.g., respondent, proxy, 
>>interviewer.)
>>
>>Also, respondent is definitely a comparison criterion - and a very 
>>important one !
>>
>>So what I'm assuming is that we will somehow be able to extract the type 
>>of info I'm listing in your quote below (respUnit, anlysUnit, universe, etc)
>>directly from the variable description bec. we already have tags that 
>>contain this type of info.
>>
>>Cheers.
>>Sanda.
>>
>>p.s. any news on the SSVD project?
>>
>>
>>
>>
>>At 02:02 PM 9/14/2005, you wrote:
>>>Hi Sanda,
>>>
>>>I have a question:
>>>
>>>"It will also be necessary to identify, for each variable, the
>>>respondent, analysis unit, universe, and certainly the physical data
>>>file containing the variable. "
>>>
>>>What's "the respondent"? I'm confused because the other things
>>>mentioned, "analysis unit", etc., are "variable-wide", so to speak,
>>>but my understanding of "the respondent" is that it pertains only to a
>>>single record, so wouldn't be a comparison criteria.
>>
>>
>>
>>Sanda Ionescu,
>>Research Associate
>>Inter-university Consortium for Political and Social Research (ICPSR)
>>The University of Michigan
>>P.O. Box 1248
>>Ann Arbor, MI 48106
>>
>>Phone: (734) 615-7890
>>Fax: (734) 615-7890
>>        (734) 647-8200
>
>
>
>Sanda Ionescu,
>Research Associate
>Inter-university Consortium for Political and Social Research (ICPSR)
>The University of Michigan
>P.O. Box 1248
>Ann Arbor, MI 48106
>
>Phone: (734) 615-7890
>Fax: (734) 615-7890
>        (734) 647-8200



Sanda Ionescu,
Research Associate
Inter-university Consortium for Political and Social Research (ICPSR)
The University of Michigan
P.O. Box 1248
Ann Arbor, MI 48106

Phone: (734) 615-7890
Fax: (734) 615-7890
        (734) 647-8200