Folks,

  Some of you already have some ideas about what questions/data you want
  to consider for project 2 and that's great.  

  If you are stuck for data and can't think of anything else, please 
  take a look at the world wide web.  Here are some specific links that
  might prove useful.  Many of them are overlapping but it is still
  useful to look at several.

     * please do not forget our very own StatLib (http://www.stat.cmu.edu/);
       in particular the following sections are good; and StatLib has
       other good data sections as well:
	http://lib.stat.cmu.edu/datasets/
	http://lib.stat.cmu.edu/data-expo/ 
	http://lib.stat.cmu.edu/DASL/
	http://www.maths.uq.edu.au/~gks/data/index.html
       and there are also some interesting statistical archive links at
       the bottom of
	http://lib.stat.cmu.edu/otherplaces/

     * John Behrens at ASU keeps very good collections of data links.  I
       especially like
	http://research.ed.asu.edu/siip/webdata/
       although it is getting a little out of date; it is also worth 
       looking at 
	http://research.ed.asu.edu/siip/
       itself.

     * Clay Helberg has a famous set of links for statisticians, incl
       data sources:
	http://www.execpc.com/~helberg/statistics.html

     * The federal government has lots of interesting data:
	http://www.fedworld.gov/

       see especially the bureau of labor statistics, the census bureau,
       etc.  Some of these are also indexed on John Behrens' pages above.

     * An excellent resource for sociological data is
	http://www.icpsr.umich.edu

     * Some more specific directories include
        http://www.ics.uci.edu/~mlearn/MLRepository.html
       (which is oriented toward machine learning and medical data a bit)
	http://www.stat.ufl.edu/vlib/statistics.html
       (which has a few specific things I didn't find elsewhere)

  and do not forget search engines like Alta Vista (http://www.altavista.com/)
  or google (http://www.google.com).  I have found very good data on the web 
  that is not indexed anywhere else, by spending 90 minutes fooling around 
  with queries in Alta Vista.  For example searching for 

       +"Fortune 500" +data

  in Google produced approx. 192,000 hits(!), the seventh of which 
  included an Excel spreadsheet of summary statistics for the companies, 
  etc.

-BJ