Be sure to have a look at <a href="http://www-personal.umich.edu/~mejn/election/">http://www-personal.umich.edu/~mejn/election/</a> and <a href="http://www.princeton.edu/~rvdb/JAVA/election2004/">http://www.princeton.edu/~rvdb/JAVA/election2004/</a> as an introduction. The comments tell the story and give the credit, the text box describes how to create a fathom version of the map, and the resized collection gives you a display of the data that helps to think about population density.
For the serious statisticians in the crowd, I have two questions:
1. If you plot the pctBlack (Black/pop2000) against the pctBush (Bush/(Bush+Kerry+Nader)), you get a wedge-shaped graph. It is like the maximum possible level of support for various proportions has a linear correlation but there is lots of variation below that. Is this a typical behaviour or does it result from some simple structure in the data that I am missing? 2. Tim Erickson's suggestion to plot log(populationdensity) vs. pctBush to get a football shaped cluster with negative correlation is interesting. Is taking the log just a device to deal with the huge range or does it point to some other important property of the data?
Ross Isenegger IT Coordinator (JK-OAC) Near North District School Board P.O. Box 3110, North Bay, ON P1B 8H1 (mail) 963 Airport Road, North Bay, ON P1C 1A5 (courier) (705) 472-7015 x5022 Fax (705) 472-9927 firstname.lastname@example.org www.nearnorth.edu.on.ca
"If your plan is for one year, plant rice. If your plan is for 10 years, plant trees. If your plan is for 100 years, educate children" - Confucius
"The past is the only dead thing that smells sweet." - Edward Thomas
"Eschew obfuscation" - from a bumper sticker on a beat up Toyota pickup
To unsubscribe, send mail to email@example.com In the body of the message, use the following command: unsubscribe fathom
To reach the (actual, human) owner of the list, send your message to: firstname.lastname@example.org