[DDI-ADG] initial use case ideas: user searching

[Katherine McNeill-Harman]

Following is my initial brainstorming on use cases for geography.  I'm 
starting from the perspective of ways in which users would want to search 
by geography, giving examples of types of data, rather than specific files 
(maybe we also want latter).  So this is coming from the perspective of how 
we want systems to be designed to be useful to users (as well as to be able 
to accommodate various file types).  Not sure if this is what the group had 
in mind, but hopefully it's a helpful start.

This is based on my experience in helping users find and work with data 
files, and many are actual questions I've received; this is not based on 
experience actually implementing 2.0, so I'm sure others who have will have 
other helpful ideas.  I've started to try to group them into categories, 
but even describing those categories proved to be a challenge.

Ways in which users want to use geography to search for data:

1. Searching for variable information within a specific geographic coverage 
area at a specific level of geographic unit (yet still aggregated), e.g.
- data on voter registration by precinct within the United States
- unemployment rate by county within the United States
- population and housing data for Eastern European cities (in this case the 
coverage and unit of analysis are more closely related than in the above 
examples)

2. Similar to "a" above, but for microdata, in which case the information 
is not aggregated to a certain level, but where the user wants a 
micro-level identifier at a certain geographic level, e.g.
- want public opinion poll on a certain topic and to be able to 
differentiate among user responses by state (i.e. where state of residence 
was a question asked of the respondent)

3. Situations in which a time series has a quasi-geographic aspect, but 
geography isn't exactly a variable, e.g.
-  historical prices for municipal bonds within the United States (i.e. the 
geographic level of a municipality is a quality of the bond (i.e., as 
opposed to federal bonds), so might users think to search by geographic level?)

4. User looking to compare information about similar-level (yet 
differently-defined) geographic areas (either aggregate or microdata), e.g.
- comparing literacy rates at the sub-national level among several 
countries (i.e. some countries call those areas "states," others 
"provinces;" how will they readily identify both?)
(Relating to some of this week's discussion about labels on variables)

5. Issues of geographic correspondence (this may seem to be more of a data 
management issue, so I'm not sure exactly how it applies to the DDI, but 
there are some data files designed to provide correspondence)
- users needing to match up two data files by a common geographic unit of 
analysis (e.g. a data file on education variables to be merged with one on 
economic variables, both with a common county code).
- user needing to expand their geographic understanding of a data file, 
either for its own sake or to match two data files with different 
geographic units, e.g. one user requesting a particular data product that 
matches zip codes to telephone area codes (something one might not think of 
as a geographic ID, but they can function as such in the U.S.)

Hope this is a helpful start.
Kate

___________________________________________
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