[DDI-ADG] UNC document

[Katherine McNeill-Harman]

Good points.  Although I imagine (but don't know where it fits in for sure) 
that this structure (the ability to refer to collection issues at the 
variable--or even response--level) is already built into the model for 
3.0.  Similar issues would apply to a survey, where "Field collection 
description/limitations" might be analogous to, in a survey, missing values 
(e.g. respondent refused to answer question) and Field collection dates to 
the fact that a survey is administered during a time period (e.g. a month), 
but not each one will be on the same day.

I think we've provided a structure for these, but some things will need to 
be decided by the study's author (e.g. what sort of missing values codes to 
allow, to what level of granularity to represent collection dates, etc.).

Kate

At 04:41 PM 11/8/2005 -0800, Ilona Einowski wrote:

>Dear all (I'm SURE I did "Reply All" this time...)
>
>After another look at this it seems to me that we have discussed most of
>these issues.
>
>I think there needs to be an agreement as to which elements go in which
>section of the DDI.  Section 2 (study description and methodology) could
>handle some of these elements but as I reread John's Project descriptions it
>occurred to me that there are some questions....whether we need to answer
>those questions right now is another matter...
>
>We could assume that some of these elements might apply to the entire study.
>
>
>For example, for "Collection personnel" could we assume you would start and
>end the study with the same company or group of surveyors? ...seems logical,
>and that would apply to the entire dataset and would belong up in Section
>2....
>
>But what about the element "Field collection description/limitations" where
>he talks about difficulty getting GPS signal lock due to tree cover or high
>buildings...that could vary for these agricultural plots or for the
>residential locations...some plots or houses would be easy to lock on,
>others would be difficult.  In that case, this element would need to be
>associated with each particular plot or residence location.
>
>Same thing with Field collection dates...on a particular study you might be
>able to just give the range of dates at the study level but for some
>agricultural plots where the survey took place over a period of several days
>or weeks, it might be important to know that there was a heavy rain storm on
>a particular date which "changed" the appearance of the crops...
>
>I'm sure we can come up with additional wrinkles but this is just to say
>that I think what we have already covers most cases...we already aware of
>the fact that we needed to build in flexibility.
>
>
>...just my 2 cents...
>
>Ilona
>
>-----Original Message-----
>From: ddi-adg-bounces at icpsr.umich.edu
>[mailto:ddi-adg-bounces at icpsr.umich.edu] On Behalf Of Wendy Thomas
>Sent: Tuesday, November 08, 2005 1:57 PM
>To: ddi-adg at icpsr.umich.edu
>Subject: [DDI-ADG] UNC document
>
>After reading this through it does seem like the vast majority of the
>information items noted are in the collection methodology section. After
>looking at the work that was done by the various working groups and the
>content of DDI version 2.0, the SRG concluded that significant work needed
>to be done on the structure of items currently in section 2 of the DDI
>(study description and methodology). The need for improvement in these
>sections touches on all types of data and should be conceptualized and
>approached in a unified way. For that reason, major restructuring of this
>section (aside from the instrumentation piece) will be scheduled for 3.x.
>
>In fact, it looks like a lot of what they want is already in the DDI but
>only as PCDATA buckets. It would be interesting to note what special needs
>they have in each area and whether those need to be machine-actionable or
>just captured.
>
>That said, the spacial data requirements here are not that complex and
>many are not unique to geographic based collection. We have addressed many
>of the temporal pieces and the availability of reusable classes on agents
>and roles should address some additional points.
>
>The main geographic items are:
>
>1) identifying the type of geography expressed by an observation (point,
>line, or polygon)
>2) uniquely identifying that within the observation
>3) linking it to an appropriate geographic boundary file (if required)
>
>To address each in turn:
>
>1) refer to FGDC/ISO to obtain a controlled vocabulary list of geographic
>types (there is also circle and arc that I know of) and provide an element
>that allows identification of the most descrete geographic type
>
>2) This can currently be done. A variable (or 2 variables) can be used to
>identify the long/lat of a point thereby identifying it uniquely. A line
>is simply a start and end point. A polygon is commonly linked to a
>geographic file by its identifier (code or name). If you have unique
>polygones such as self determined plots, it is best to describe the
>polygon using a geography based system (FGDC/ISO etc) providing a unique
>identifer to each polygon and then using that identify in the data record
>as a link.
>
>3) see 2
>
>or am I missing something in this?
>
>wendy
>
>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
>
>_______________________________________________
>DDI-ADG mailing list
>DDI-ADG at icpsr.umich.edu
>http://www.icpsr.umich.edu/mailman/listinfo/ddi-adg
>
>_______________________________________________
>DDI-ADG mailing list
>DDI-ADG at icpsr.umich.edu
>http://www.icpsr.umich.edu/mailman/listinfo/ddi-adg

___________________________________________
Katherine McNeill-Harman
Data Services Librarian
Dewey Library for Management and Social Sciences
Massachusetts Institute of Technology
77 Massachusetts Avenue, E53-100
Cambridge, MA 02139
mcneillh at mit.edu
617-253-0787