Not content with merely reporting the news, the New York Times appears to have developed a research unit that it calls "The Upshot" which appears to conduct various types of research projects. Here is a sample of some recent projects: http://www.nytimes.com/upshot/ Of greatest relevance to Tips is an economic study they conducted that describes each county in the U.S. on six measures: (a) Educational attainment (e.g., either highest grade or highest degree obtained) (b) Household income (c) Jobless rate (d) Disability rate (i.e., people who are pre-retirement but unable to work; these people are considered in the work/unemployment statistics but would be if they were not disabled, implying they all have complete disability status) (e) Life expectancy (f) Obesity rate
I will leave it to the reader to find out which counties (as well as regions) of the U.S. are the "hardest" to live in, as described by lowest mean rank on the six measures above (at least that's what I think they are using as their summary metric; don't you hate it when newspaper articles like these don't follow APA style ;-) and which appear to have it best (the best may surprise you). The NY Times has two stories on this: (a) A briefer one with a map of counties of the U.S. color coded for their average ranking; see: http://www.nytimes.com/2014/06/26/upshot/where-are-the-hardest-places-to-live-in-the-us.html?emc=edit_th_20140629&nl=todaysheadlines&nlid=389166&_r=0 (b) A longer text piece that may count as qualitative research; see: http://www.nytimes.com/2014/06/29/magazine/whats-the-matter-with-eastern-kentucky.html The Times does not appear to make their data public (if someone does find it, please let us know) so one can't replicate the analysis to double-check their conclusions. It should be pointed out that the overall orientation of the articles is one based on descriptive statistical analysis and though comparisons of the "highest ranked" and "lowest ranked" counties are made there are no statistical tests. One might want to use these as examples of how once can do a statistical description of a situation without the use of inferential statistics. Also, as tempting as might have been, there does not appear to be any causal statements made about why the counties differ in the way that they do (but the use of additional variables, such as amount of federal government spending, might produce reasonable causal models). I'll leave the political implications (some of which are mentioned in the articles) to others. -Mike Palij New York University m...@nyu.edu . --- You are currently subscribed to tips as: arch...@jab.org. To unsubscribe click here: http://fsulist.frostburg.edu/u?id=13090.68da6e6e5325aa33287ff385b70df5d5&n=T&l=tips&o=37330 or send a blank email to leave-37330-13090.68da6e6e5325aa33287ff385b70df...@fsulist.frostburg.edu
