Changes

Jump to navigation Jump to search
→‎A research proposal: Comparitive groups, problems, and biases. And who will pay?
Line 83: Line 83:     
: I'm fine with including all religions.  However, the control does not have to be [[atheistic]] children of [[atheists]].  Other control groups are possible, or perhaps simply use the general population (of comparable privilege) as a rough control.--[[User:Aschlafly|Aschlafly]] 17:08, 26 February 2008 (EST)
 
: I'm fine with including all religions.  However, the control does not have to be [[atheistic]] children of [[atheists]].  Other control groups are possible, or perhaps simply use the general population (of comparable privilege) as a rough control.--[[User:Aschlafly|Aschlafly]] 17:08, 26 February 2008 (EST)
 +
 +
:: He suggested the control group as atheist children of theists, not of atheists.  Atheist children of theists are the appropriate control if you are comparing ''second-generation'' atheists with ''first-generation'' atheists.  But it would also be interesting to compare both groups to first-generation Christians and second-generation Christians.
 +
:: And although you could add as separate groups other religions, lumping all theistic religions into one group vs. atheists is making a huge, and invalid, assumption about how similar various theistic religions are.  So after atheists, the first group you would compare against would be Christians, and preferably Christians from a reformed background.
 +
:: All proper studies have to grapple with problems of definitions, avoiding hidden biases, etc., and the same applies here, but I don't see any exceptionally difficult problems with a study like this.  Religion, at least, is already determined by many other studies.  It can be done.  Achievement would not be a particular problem, although one thing to be wary of is to avoid a reliance on academic achievement, as people who have achieved academically have, for the most part, gone through a system (i.e. university and the like) that is predominantly atheistic, and which would therefore cause a bias in the results.
 +
:: But such a study is the sort of thing that you get a grant of tens of thousands of dollars for, and takes many months of work.  I very much doubt that it would be within the capabilities of anyone here.
 +
: [[User:Philip J. Rayment|Philip J. Rayment]] 21:39, 26 February 2008 (EST)
    
== A quick check ==
 
== A quick check ==
nsTeam2RO, nsTeam2RW, nsTeam2_talkRO, nsTeam2_talkRW
13,254

edits

Navigation menu