Slightly OT: Graphic of Unemployment in the United States
in [SAS]
|
From: Ben Powell on 2 Dec 2009 09:41 How to Display Data Badly Howard Wainer The American Statistician, Vol. 38, No. 2. (May, 1984), pp. 137-147. Stable URL: http://links.jstor.org/sici?sici=0003-1305%28198405%2938%3A2%3C137%3AHTDDB%3E2.0.CO%3B2-N The American Statistician is currently published by American Statistical Association.
From: Ben Powell on 2 Dec 2009 10:35 Looking at the final map graphic, is it telling you unemployment is over 10%, or 7-9.9%? The difference is quite significant. The scale is also nonlinear which adds to the confusion, and you have to take the 8.5% as given but correlating it to the graphic is impossible. So its doing two things, giving a basic national average of unemployment over time (8.5%) and showing geographical non-population weighted unemployment spread. The former is unimpressive, the latter is irrelevant and misleading: unemployment data is only of use where it's population weighted. Rgds
From: Ben Powell on 2 Dec 2009 10:37 I agree, brings to mind the nerve density maps of the human body, where the finger tips are oversize, Rgds
From: Savian on 2 Dec 2009 11:25 On Dec 2, 8:35 am, ben.pow...(a)INFXSOLUTIONS.COM (Ben Powell) wrote: > Looking at the final map graphic, is it telling you unemployment is over > 10%, or 7-9.9%? > > The difference is quite significant. > > The scale is also nonlinear which adds to the confusion, and you have to > take the 8.5% as given but correlating it to the graphic is impossible. > > So its doing two things, giving a basic national average of unemployment > over time (8.5%) and showing geographical non-population weighted > unemployment spread. The former is unimpressive, the latter is irrelevant > and misleading: unemployment data is only of use where it's population weighted. > > Rgds Ben, I disagree. If the goal is to show how widespread, geographically, unemployment is, this graph works. It depends on what you pull out of it. Certainly population density is important but I never looked at it that way. Alan
From: "Choate, Paul on 2 Dec 2009 11:36
Haven't read the whole thread, so pardon if this was mentioned .... This is similar to the red vs. blue election maps - population cartograms are much better at proportionally representing geographic based information. Mark Newman of the Department of Physics and Center for the Study of Complex Systems at University of Michigan has a great web page on this. http://www-personal.umich.edu/~mejn/election/2008/ Paul Choate DDS Data Extraction (916) 654-2160 -----Original Message----- From: SAS(r) Discussion [mailto:SAS-L(a)LISTSERV.UGA.EDU] On Behalf Of Ben Powell Sent: Wednesday, December 02, 2009 7:38 AM To: SAS-L(a)LISTSERV.UGA.EDU Subject: Re: Slightly OT: Graphic of Unemployment in the United States I agree, brings to mind the nerve density maps of the human body, where the finger tips are oversize, Rgds |