[DDI-users] InParameter/OutParameter and Binding in Questionnaires

Wendy Thomas wlt at umn.edu
Mon Dec 14 13:21:31 EST 2015


This and the question grid usage of the array in In/Out Parameters and
Binding has been posted as an issue to DDILIFE 3511

Wendy

On Mon, Dec 14, 2015 at 11:41 AM, Wackerow, Joachim <
Joachim.Wackerow at gesis.org> wrote:

> Thanks. This is helpful.
>
>
>
> I’m also concerned regarding reusability. Some more comments:
>
>
>
> The binding definition is a child element of QuestionItem,
> QuestionConstruct, and Sequence. This means each of these elements are only
> reusable independently of the context if the binding is defined in the
> element where the InParameter is defined. This is just in other words what
> you said with “Binding for objects intended for reuse should be done at
> their point of use when possible.”
>
>
>
> The SourceParameterReference seems to combine the definition of an
> InParameter with a binding. I assume this can be limiting because no
> specific value of an array can be addressed. I.e. a variable can be the
> result of a question in a row of a QuestionGrid. The binding would only be
> possible by means of an intermediate GenerationInstruction. Or did I miss
> something?
>
>
>
> Anyway, the use of the SourceParameterReference makes a Variable not
> reusable.
>
>
>
> A thought on future DDI versions:
>
> A more consistent approach would be to use the same solution as an
> QuestionItem: the definition of an InParameter. The binding should be done
> independently of the Variable to support reusable variables.
>
>
>
> Achim
>
>
>
>
>
> *From:* ddi-users-bounces at icpsr.umich.edu [mailto:
> ddi-users-bounces at icpsr.umich.edu] *On Behalf Of *Wendy Thomas
> *Sent:* Montag, 14. Dezember 2015 15:29
> *To:* Data Documentation Initiative Users Group
> *Subject:* Re: [DDI-users] InParameter/OutParameter and Binding in
> Questionnaires
>
>
>
> A couple of points about binding in 3.2:
>
>
>
> First in answer to your earlier question regarding the path of the binding
> from the question item to the control construct to the generation
> instruction to the variable. You are correct that you could bind directly
> from the variable to the question. However, a question item may be used in
> many locations within a questionnaire (Age for example). It could be part
> of multiple sequences, in a loop, etc. By specifying the link between a
> question and its specific control construct the question can be reused
> multiple times, and each time one can create a binding to its specific
> usage.
>
>
>
> Binding itself may be done within the construct of the item or at the time
> of reference using the SourceParameterReference. This may not be as well
> distributed as it should be. I see it in VariableType and
> ConditionalTextType. This should be reviewed for a 3.3 version. Binding to
> a question outparameter in an instrument should be done at the level of the
> control construct to keep the question reusable. In short, you can make
> your binding links as short as you wish for your specific needs. Binding
> for objects intended for reuse should be done at their point of use when
> possible.
>
>
>
> Wendy
>
>
>
> On Sun, Dec 13, 2015 at 10:53 AM, Wackerow, Joachim <
> Joachim.Wackerow at gesis.org> wrote:
>
> Relationship class is in DDI 4.
>
> The related documentation says “Relationship specification between this
> item and the item to which it is related.”
>
> Using this approach for binding would conflict with the idea of
> reusability because the reference to a another object would be part of the
> object itself.
>
>
>
> The concept of binding is in 3.2 a little different. Two identifiable
> parameters are referenced by the binding definition. But the binding is a
> child element of the element having a parameter. This seems to be limiting
> regarding reusability.
>
>
>
> Back to the usage in DDI 3.2:
>
> Any thoughts on the question regarding the binding definition only in the
> Sequence (the OutParameter of Question 1 and the InParameter of Question
> 2).
>
> This is a DDI 3.2 question.
>
>
>
> Achim
>
>
>
>
>
>
>
> *From:* ddi-users-bounces at icpsr.umich.edu [mailto:
> ddi-users-bounces at icpsr.umich.edu] *On Behalf Of *Hoyle, Larry
> *Sent:* Freitag, 11. Dezember 2015 20:48
> *To:* Data Documentation Initiative Users Group
> *Subject:* Re: [DDI-users] InParameter/OutParameter and Binding in
> Questionnaires
>
>
>
> Would this be an application of the Relationship classes?
>
>
>
> --- Larry Hoyle
>
>
>
>
>
> *From:* ddi-users-bounces at icpsr.umich.edu [
> mailto:ddi-users-bounces at icpsr.umich.edu
> <ddi-users-bounces at icpsr.umich.edu>] *On Behalf Of *Wackerow, Joachim
> *Sent:* Friday, December 11, 2015 12:32 PM
> *To:* Data Documentation Initiative Users Group (ddi-users at icpsr.umich.edu)
> <ddi-users at icpsr.umich.edu>
> *Subject:* [DDI-users] InParameter/OutParameter and Binding in
> Questionnaires
>
>
>
> I’m wondering where the binding can be specified and which way is best to
> support reusability of QuestionItems.
>
>
>
> This topic is documented in the Technical Documentation of DDI 3.2. There
> is a related diagram on page 35 (and a related example XML file
> InOutBindingExample). This explains that binding is done in a chain from
> Question 1 up to the Sequence and then down again to Question 2. This shows
> all possibilities of defining bindings in this case.
>
>
>
> A simpler way would be to define the binding of directly in the Sequence
> (the OutParameter of Question 1 and the InParameter of Question 2). Then
> all other bindings are not necessary and the questions are totally
> reusable. This would also correspond to the actual use of these two
> questions in the sequence. Only then the binding is important.
>
>
>
> Are there any disadvantages to use this option?
>
>
>
>
>
> A more general thought: a definition of a binding is probably mostly of
> local use (here for questions in a sequence). Therefore it can conflict
> with the idea of reusability. In this sense, a better location for binding
> definitions would be an independent scheme. On the other hand, if the
> sequence should be reused like it is it makes sense to have the binding
> definition in the sequence.
>
>
>
> Achim
>
>
>
>
>
>
>
>
>
>
>
>
> _______________________________________________
> DDI-users mailing list
> DDI-users at icpsr.umich.edu
> http://lists.icpsr.umich.edu/mailman/listinfo/ddi-users
>
>
>
>
> --
>
> Wendy L. Thomas                              Phone: +1 612.624.4389
>
> Data Access Core Director                 Fax:   +1 612.626.8375
>
> Minnesota Population Center             Email: wlt at umn.edu
>
> University of Minnesota
>
> 50 Willey Hall
>
> 225 19th Avenue South
>
> Minneapolis, MN 55455
>
> _______________________________________________
> DDI-users mailing list
> DDI-users at icpsr.umich.edu
> http://lists.icpsr.umich.edu/mailman/listinfo/ddi-users
>
>


-- 
Wendy L. Thomas                              Phone: +1 612.624.4389
Data Access Core Director                 Fax:   +1 612.626.8375
Minnesota Population Center             Email: wlt at umn.edu
University of Minnesota
50 Willey Hall
225 19th Avenue South
Minneapolis, MN 55455
-------------- next part --------------
An HTML attachment was scrubbed...
URL: http://lists.icpsr.umich.edu/pipermail/ddi-users/attachments/20151214/ed116e1b/attachment-0001.html 


More information about the DDI-users mailing list