1,129
edits
Changes
Jump to navigation
Jump to search
Line 150:
Line 150: +
→Innumeracy
: Not that small. Several hundred at least. Probably a higher percentage than most polls of the American public.--[[User:Aschlafly|Aschlafly]] 01:22, 28 April 2007 (EDT)
: Not that small. Several hundred at least. Probably a higher percentage than most polls of the American public.--[[User:Aschlafly|Aschlafly]] 01:22, 28 April 2007 (EDT)
:: Sampling error isn't significantly affected by the percentage of the population included in the sample -- a representative sample of 1000 is as reliable in the USA as it is in Monaco. The problem in this case is that it's difficult to justify the assumption that those who self identify as liberal or conservative are representative of the community of Wikipedia editors. It may well be true that there are six times as many liberal Wikipedians than conservative, but it's not statistically reasonable to conclude that from a self-selected sample such as this.--[[User:Jalapeno|Jalapeno]] 04:52, 28 April 2007 (EDT)
:: Sampling error isn't significantly affected by the percentage of the population included in the sample -- a representative sample of 1000 is as reliable in the USA as it is in Monaco. The problem in this case is that it's difficult to justify the assumption that those who self identify as liberal or conservative are representative of the community of Wikipedia editors. It may well be true that there are six times as many liberal Wikipedians than conservative, but it's not statistically reasonable to conclude that from a self-selected sample such as this.--[[User:Jalapeno|Jalapeno]] 04:52, 28 April 2007 (EDT)
:: The last time I looked at Wikipedia, the same size was 274 (for conservatives + liberals). This is not a large sample. The total percent sample size is about half that for a Harris poll (that it admits that it has errors that gives that it cannot give a good confidence[http://www.harrisinteractive.com/harris_poll/index.asp?PID=548]), however the sample for wikipedia is self selected meaning that you cannot use it to make statistically responsible statements.[http://www.aapor.org/pdfs/2006/samp_err_stmt.pdf] --[[User:Mtur|Mtur]] 22:30, 9 May 2007 (EDT)
2. In the absence of its inverse, the conservative quotient, defining the Lq is silly knee-jerk reactionism.
2. In the absence of its inverse, the conservative quotient, defining the Lq is silly knee-jerk reactionism.