Block, SkipCaptcha, Automoderated users, edit, rollback
3,135
edits
Changes
Jump to navigation
Jump to search
Line 587:
Line 587: + + + + + + + + + + + + + + +
Talk:Essay:Best New Conservative Words (view source)
Revision as of 01:22, February 2, 2013
, 01:22, February 2, 2013→My problems with making any statistical conclusions: new section
::::::Using your method, we could show that there are as many Germans as there are Americans, just by alternately listing Germans and Americans! That could go on for quite a while, without proving anything about the real ratio. (Browsing through the archives, I see I'm not the first to point this out: see [[Talk:Essay:Best_New_Conservative_Words/archive3#Research_method|here]] and [[Talk:Essay:Best_New_Conservative_Words/archive3#Selection_Bias_and_Proposal_for_an_Unbiased_Test|here]])
::::::Using your method, we could show that there are as many Germans as there are Americans, just by alternately listing Germans and Americans! That could go on for quite a while, without proving anything about the real ratio. (Browsing through the archives, I see I'm not the first to point this out: see [[Talk:Essay:Best_New_Conservative_Words/archive3#Research_method|here]] and [[Talk:Essay:Best_New_Conservative_Words/archive3#Selection_Bias_and_Proposal_for_an_Unbiased_Test|here]])
::::::--[[User:AugustO|AugustO]] 16:57, 31 January 2013 (EST)
::::::--[[User:AugustO|AugustO]] 16:57, 31 January 2013 (EST)
== My problems with making any statistical conclusions ==
I will provide my honest opinion.
If this essay is meant to be a personal project in learning new vocabulary, etymology, and use of language, then it's certainly worthwhile. However, if the intent of this essay was to draw statistical conclusions, the methodology is so egregiously flawed that there is really nothing one can take away from the data in the essay.
Several methodology problems are the following:
*There is no objective definition as to what constitutes a word, or a conservative word, for that matter. For some reason, abbreviations, proper names, and multiple-word terms appear on the list. Very similar terms are alternative counted as conservative or not (contrast "welfare queen" and "welfare to work", for example). The fact that there are several arguments on this very page as to whether or not words are conservative indicate that the notion of a "conservative word" is not well-defined.
*The arbiter of whether or not a term is conservative appears to a strong interest in ensuring that the data collected fit into a predetermined "doubling" pattern. (In fact, it seems to me that this approach to the subject demonstrates a lack of open-mindedness that the empirical data may suggest a different conclusion than the one anticipated.) It's not even single blind, much less double blind.
*There is no consistency as to the date assigned to a term. It appears to be the date a term became conservative. There is no objective calculation of the date assigned to a term (like, for example, consulting the OED for the oldest extant print reference to a word).
*No comparison is made to the growth in the number of English terms since 1600 nor the growth of liberal terms over time.
*The words are not randomly sampled. They appear to be inserted in order to fit a predetermined pattern. A better method to solve the problem would be to randomly select words from the dictionary.
In all, if you want to keep this as a side project just for the educational value, be my guest. However, if you want to draw mathematical or statistical conclusions from this essay, I would advise you not to, as it only serves to make us conservatives at Conservapedia appear sophomoric in mathematics, statistics, and logic. Thanks, [[User:GregG|GregG]] 20:22, 1 February 2013 (EST)