Hi Wendy,<br><br>With regards to the second email: this solution is meant to be a reusable attempt at solving this issue in DDI3.1 exclusively. If the standard changes this approach may also have to change as well, or be wholly unnecessary.<br>
<br>It also very much depends on what problem you are trying to solve. For me the issue is how can one document the structure and semantic logic of a form in a succinct way, keeping as much of that information in the DDI.<br>
<br>With this in mind, if you take the QuestionItem/StructureMixedResponseDomain approach to documenting these kinds of questions, the logic about when one can fill in the "Other please specify" text field has legitimate storage location within 'pure' DDI (although a not could be attached - however, I try to limit this approach when ever possible).<br>
<br>The MultiQuestion approach means that the Coded response and free text response can be linked together in a meaningful way, and have some presenational/semantic logic applied at the point within the metadata - along with the intent of this logic. This means that if there were multiple tools for presenting DDI forms, be they web, paper or other, if I grabs DDI from another agency who used a different tool to me, I could read the intent of the logic within their metadata surrounding the 'other please specify' question, and at most have to attach a new expression for my tool within the SubQuestionSequencing.<br>
<br>While the reality of the situation is that there are few publicly available DDI to web form tools or metadata-driven paper form creation/scanner systems people are probably working in this space. And a common way that both support automated design and is relatively sound DDI in principle will help ensure that as people begin announcing more tool that they will be as compatible as possible with each other.<br>
<br>However, as you said the changes to MultiQuestions in DDI3.2 will be vast and by the looks of it quite an improvement, and may render such approaches obsolete in the future making this a stop gap to support an immediate need, rather than an ongoing best practice.<br>
<br>Cheers,<br>Sam.<br><br clear="all">--- Specificity is the soul of all good communication ---<br>--- When the game is over, the king and the pawn go into the same box ---<br>Find out more about me: <a href="http://about.me/legostormtroopr" target="_blank">http://about.me/legostormtroopr</a><br>
<br><br><div class="gmail_quote">On 26 January 2012 03:33, Wendy Thomas <span dir="ltr"><<a href="mailto:wlt@umn.edu">wlt@umn.edu</a>></span> wrote:<br><blockquote class="gmail_quote" style="margin:0 0 0 .8ex;border-left:1px #ccc solid;padding-left:1ex">
Actually Sam the normal way is to use a single question and<br>
StructuredMixedResponse<br>
<br>
Field Level:<br>
ResponseDomains should be chosen that do NOT duplicate responses such<br>
as CodeSchemes with overlapping codes. Be aware that certain<br>
instruments may collect responses in such a way that confusion between<br>
a code response and text response may be possible. The process of<br>
resolving such conflicts should be addressed in the data processing<br>
instructions. There is an assumption that if a text or numeric<br>
response duplicates a coded response to a question, that the value is<br>
that of the coded category.<br>
<br>
See page 42 of DDI 3.1 Part II: User Guide<br>
<br>
At least that how it was designed to work.<br>
<br>
Wendy<br>
<div><div class="h5"><br>
On Sun, Jan 22, 2012 at 6:07 AM, Samuel Spencer<br>
<<a href="mailto:theodore.therone@gmail.com">theodore.therone@gmail.com</a>> wrote:<br>
> Hi all,<br>
><br>
> I had a bit of a brainwave late in the weekend looking at how to work with<br>
> the idea of complex questions in DDI.<br>
> The issue at hand is how to solve questions where a respondent either picks<br>
> from a list of codes or enters their own option. For an example see this<br>
> image from the ONS 2001 Census here: <a href="http://i.imgur.com/9xpvP.png" target="_blank">http://i.imgur.com/9xpvP.png</a><br>
><br>
> The normal approach is to use a MultipleQuestionItem, but there have been<br>
> issues about how to deal with the logic about what two answer when. I<br>
> believe a suitable approach that keeps the questions together, and supports<br>
> the logic is through the use of a SubQuestionSequence within the<br>
> MultipleQuestionItem.<br>
><br>
> Rationale and a larger example are available here: <a href="http://bit.ly/yEPcZ5" target="_blank">http://bit.ly/yEPcZ5</a><br>
> A more comprehensive example will be put into the DDI Examples Repository<br>
> later this week, pending critique of the suggested approach.<br>
><br>
> Cheers,<br>
> Sam.<br>
><br>
> --- Specificity is the soul of all good communication ---<br>
> --- When the game is over, the king and the pawn go into the same box ---<br>
> Find out more about me: <a href="http://about.me/legostormtroopr" target="_blank">http://about.me/legostormtroopr</a><br>
><br>
</div></div>> _______________________________________________<br>
> DDI-users mailing list<br>
> <a href="mailto:DDI-users@icpsr.umich.edu">DDI-users@icpsr.umich.edu</a><br>
> <a href="http://lists.icpsr.umich.edu/mailman/listinfo/ddi-users" target="_blank">http://lists.icpsr.umich.edu/mailman/listinfo/ddi-users</a><br>
><br>
<br>
<br>
<br>
--<br>
Wendy L. Thomas Phone: <a href="tel:%2B1%20612.624.4389" value="+16126244389">+1 612.624.4389</a><br>
Data Access Core Director Fax: <a href="tel:%2B1%20612.626.8375" value="+16126268375">+1 612.626.8375</a><br>
Minnesota Population Center Email: <a href="mailto:wlt@umn.edu">wlt@umn.edu</a><br>
University of Minnesota<br>
50 Willey Hall<br>
225 19th Avenue South<br>
Minneapolis, MN 55455<br>
<br>
_______________________________________________<br>
DDI-users mailing list<br>
<a href="mailto:DDI-users@icpsr.umich.edu">DDI-users@icpsr.umich.edu</a><br>
<a href="http://lists.icpsr.umich.edu/mailman/listinfo/ddi-users" target="_blank">http://lists.icpsr.umich.edu/mailman/listinfo/ddi-users</a><br>
</blockquote></div><br>