RE: Athiest nations are more peaceful?

Collarchat.com

Home Login Search

RE: Athiest nations are more peaceful?

View related threads: (in this forum | in all forums)

Logged in as: Guest

All Forums >> [Community Discussions] >> Dungeon of Political and Religious Discussion >> RE: Athiest nations are more peaceful?

Page: << < prev 1 2 3 4 [5]

Message

<< Older Topic Newer Topic >>

RE: Athiest nations are more peaceful? - 6/6/2009 9:13:56 PM

CruelNUnsual

Posts: 624
Joined: 9/28/2008
Status: offline

quote:

ORIGINAL: DomKen

I am looking at that data. Based on which set you draw the data from you can have as few as 57, so not the 60 you claim, or well over 100.
http://www.worldvaluessurvey.org/

As to guessing a rho for an incomplete data set that would be foolish. Produce the complete data sets including all the rankings and I'll see if your rho is accurate or not.

its not an "incomplete data set", I used every country that has entries in both data bases. If you had the first clue what you were talking about you would also know that a sample rho from 52 out of even 200 data points is highly reliable. And I certainly dont need you to tell me my calculations are accurate. Its my business.

(in reply to DomKen)

Profile

Post #: 81

RE: Athiest nations are more peaceful? - 6/6/2009 9:45:11 PM

DomKen

Posts: 19457
Joined: 7/4/2004
From: Chicago, IL
Status: offline

quote:

ORIGINAL: Arpig

GHA!!! you two. This arguement has descended to the incomprehensible. WTF is a rho?

rho is a statistics term for a formula that determines how closely two data points correlate.
http://en.wikipedia.org/wiki/Spearman%27s_rank_correlation_coefficient

Basically it measures if the two data points have the same rate of change or not. A rho of 1 is a direct correlation, as one value increases so does the other. A rho of -1 is a direct inverse relationship, that is one decreases at the same rate the other increases. 0 indicates no direct correlation, IOW when one value changes it is not predictive of what teh other will do.

There is of course a major however to using rho as it only measures a direct correlation in a strict mathematical sense. It does not measure the situation described in the original blog post which just claimed that nations with larger athiest or non religious populations were more peaceful. Rho would only be directly relevant if the author claimed that the data showed a corresponding increase in peacefulness with an increase in atheists or decrease in relgious attendance. However I had been holding that back till I finished leading CnU down the garden path yet again but what the heck I've had my fun and it is getting tedious.

(in reply to Arpig)

Profile

Post #: 82

RE: Athiest nations are more peaceful? - 6/6/2009 9:49:44 PM

CruelNUnsual

Posts: 624
Joined: 9/28/2008
Status: offline

quote:

ORIGINAL: Arpig

GHA!!! you two. This arguement has descended to the incomprehensible. WTF is a rho?

Its a measure of the correlation between two variables, ranging from -1 to 1, with 0 being no correlation. A correlation of 1 or -1 means that changes in 1 variable can be fully "explained" by changes in the other variable. It examines each country's "atheism" score and its actual "peaceful score".

The P score in the article tells you nothing about the strength of correlation. It only tells you that when the author decided to aggregate all of the peace scores into two groups and then compare it to the average level of atheism, that the difference between the two groups is unlikely to be random.

For example, within the "atheist" countries there are varying "levels of atheism" and varying "violence scores", and likewise within the "non-atheist" countries. The P score tells you that when the author of the article chose the breakpoint for what is a "violent" country and counted the number of atheists within each, that the sampling was likely to be representative of the total population

You can also get a P score of the correlation coefficient (rho) itself, which tells you that given the size of the sample (in this case 50 "degrees of freedom") how likely it is that the sample rho is reprentative of the entire population.

Since KD is apparently unwilling to guess, the rho for "atheists" vs "peacelike" was a reasonably strong -.43. So the difference between two different countries "peace score" can be considered 43% "explained" by the level of atheism (Its negative because more atheists means "less violent".) However, the validity of the rho is fairly low because there are so few atheists reported in any country. (I am careful to use "explained" vs "attributed to" to avoid the "correation != causation" discussion.

A more robust and equally logical comparison is not just atheists, but atheists plus non-religious, since the hidden agenda of the article is to support the notion that religion results in war, and "non-religious" is a valid converse to "religious". There the correlation is only 31%, but the 31% is much more likely to be valid.

And for giggles, the correlation for "prostitution is never justifiable" vs "violent" is 54%....the more people who said it is never justifiable, the more violent the country.

(in reply to Arpig)

Profile

Post #: 83

RE: Athiest nations are more peaceful? - 6/6/2009 9:51:30 PM

CruelNUnsual

Posts: 624
Joined: 9/28/2008
Status: offline

quote:

ORIGINAL: DomKen

quote:

ORIGINAL: Arpig

GHA!!! you two. This arguement has descended to the incomprehensible. WTF is a rho?

rho is a statistics term for a formula that determines how closely two data points correlate.
http://en.wikipedia.org/wiki/Spearman%27s_rank_correlation_coefficient

Basically it measures if the two data points have the same rate of change or not. A rho of 1 is a direct correlation, as one value increases so does the other. A rho of -1 is a direct inverse relationship, that is one decreases at the same rate the other increases. 0 indicates no direct correlation, IOW when one value changes it is not predictive of what teh other will do.

There is of course a major however to using rho as it only measures a direct correlation in a strict mathematical sense. It does not measure the situation described in the original blog post which just claimed that nations with larger athiest or non religious populations were more peaceful. Rho would only be directly relevant if the author claimed that the data showed a corresponding increase in peacefulness with an increase in atheists or decrease in relgious attendance. However I had been holding that back till I finished leading CnU down the garden path yet again but what the heck I've had my fun and it is getting tedious.

Apparently you got stuck in quicksand laying your path, because this is your first correct post on the topic, and you carefully avoided your previous implication that P had anything to do with correlation.

(in reply to DomKen)

Profile

Post #: 84

RE: Athiest nations are more peaceful? - 6/7/2009 6:36:51 AM

kdsub

Posts: 12180
Joined: 8/16/2007
Status: offline

quote:

ORIGINAL: Arpig

GHA!!! you two. This arguement has descended to the incomprehensible. WTF is a rho?

It means someone is a grandiloquent sesquipedalian that is full of hyperbole who loves to read their own posts over and over while masturbating...I think

Butch

< Message edited by kdsub -- 6/7/2009 6:39:17 AM >

_____________________________

Mark Twain:

I don't see any use in having a uniform and arbitrary way of spelling words. We might as well make all clothes alike and cook all dishes alike. Sameness is tiresome; variety is pleasing

(in reply to Arpig)

Profile

Post #: 85

RE: Athiest nations are more peaceful? - 6/7/2009 10:34:44 AM

Arpig

Posts: 9930
Joined: 1/3/2006
From: Increasingly further from reality
Status: offline

Thank you all for the clarification. Especially Butch....LOL

_____________________________

Big man! Pig Man!
Ha Ha...Charade you are!

Why do they leave out the letter b on "Garage Sale" signs?

CM's #1 All-Time Also-Ran

(in reply to kdsub)

Profile

Post #: 86

RE: Athiest nations are more peaceful? - 6/7/2009 10:36:42 AM

CruelNUnsual

Posts: 624
Joined: 9/28/2008
Status: offline

BTW for this type of data there is a better alternative to the "rho" correlation coefficient, called the correlation ratio ("eta")

measures how much of the apparent statistical differences between the "violent" and "peaceful" countries" is due to variation of the NRs (number of non-religious plus atheists) between the two categories (ie NR does indeed contribute significantly to the difference in violent vs peaceful) and how much is due to variation of NR within the categories. Looking at the graph in the OP link, which Im not allowed to reproduce here, a high

indicates how much the apparent statistical differences is due to variation in NRs between the left bar and right bar, and a low

indicates how much of the statistical differences is merely due to varation in NRs up and down each bar separately. The authors implication is the

would be high...a meaningful difference between the two bars.

for this data is a very low .09...very little of the difference is due to differences between the two bars.

The author's premise is solidly refuted by

(in reply to kdsub)

Profile

Post #: 87

RE: Athiest nations are more peaceful? - 6/7/2009 1:35:38 PM

DomKen

Posts: 19457
Joined: 7/4/2004
From: Chicago, IL
Status: offline

quote:

ORIGINAL: CruelNUnsual

quote:

ORIGINAL: DomKen

quote:

ORIGINAL: Arpig

GHA!!! you two. This arguement has descended to the incomprehensible. WTF is a rho?

rho is a statistics term for a formula that determines how closely two data points correlate.
http://en.wikipedia.org/wiki/Spearman%27s_rank_correlation_coefficient

Basically it measures if the two data points have the same rate of change or not. A rho of 1 is a direct correlation, as one value increases so does the other. A rho of -1 is a direct inverse relationship, that is one decreases at the same rate the other increases. 0 indicates no direct correlation, IOW when one value changes it is not predictive of what teh other will do.

There is of course a major however to using rho as it only measures a direct correlation in a strict mathematical sense. It does not measure the situation described in the original blog post which just claimed that nations with larger athiest or non religious populations were more peaceful. Rho would only be directly relevant if the author claimed that the data showed a corresponding increase in peacefulness with an increase in atheists or decrease in relgious attendance. However I had been holding that back till I finished leading CnU down the garden path yet again but what the heck I've had my fun and it is getting tedious.

Apparently you got stuck in quicksand laying your path, because this is your first correct post on the topic, and you carefully avoided your previous implication that P had anything to do with correlation.

Actually P does have something to do with correlation. P is the probability of getting the actual outcome under the null hypothesis, in this case that atheism or low regular religious attendance has no relationship to peacefulness, i.e. a rho of 0. And with a P of 0.001 or less that is indicative that there probably is something to the data which is precisely what you rho calculations got as well.

(in reply to CruelNUnsual)

Profile

Post #: 88

RE: Athiest nations are more peaceful? - 6/7/2009 2:48:22 PM

CruelNUnsual

Posts: 624
Joined: 9/28/2008
Status: offline

quote:

ORIGINAL: DomKen

Actually P does have something to do with correlation. P is the probability of getting the actual outcome under the null hypothesis, in this case that atheism or low regular religious attendance has no relationship to peacefulness, i.e. a rho of 0. And with a P of 0.001 or less that is indicative that there probably is something to the data which is precisely what you rho calculations got as well.

For the last time, it doesn't. Your data can have a very high P test and 0 rho. They have absolutely nothing to do with each other unless you are specifically getting the P for rho itself.

The article in the OP is a perfect example. The data used for the article shows a low to medium rho, a low P test on the rho, and an extremely low correlation ratio despite the P=.001.

< Message edited by CruelNUnsual -- 6/7/2009 2:49:29 PM >

(in reply to DomKen)

Profile

Post #: 89

RE: Athiest nations are more peaceful? - 6/7/2009 8:56:09 PM

DomKen

Posts: 19457
Joined: 7/4/2004
From: Chicago, IL
Status: offline

quote:

ORIGINAL: CruelNUnsual

quote:

ORIGINAL: DomKen

Actually P does have something to do with correlation. P is the probability of getting the actual outcome under the null hypothesis, in this case that atheism or low regular religious attendance has no relationship to peacefulness, i.e. a rho of 0. And with a P of 0.001 or less that is indicative that there probably is something to the data which is precisely what you rho calculations got as well.

For the last time, it doesn't. Your data can have a very high P test and 0 rho. They have absolutely nothing to do with each other unless you are specifically getting the P for rho itself.

The article in the OP is a perfect example. The data used for the article shows a low to medium rho, a low P test on the rho, and an extremely low correlation ratio despite the P=.001.

For the absolute last time if the null hypothesis is no relation and P =< 0.001 then the odds are that there is some correlation, as in 999 out of a thousand or better odds. Your rho showed that as well. As to your new attempt at bringing in coeffecient ratio (eta) that is a measure of
more interest since it compares the means across the categories. However if the means are very small, as is certainly the case here, then the dispersion between the means would be very small as well. Any eta is indicative of some dispersion of the means which indicates the samples are not actually simply random variance with an almost 10% variance in the means between the categories the indication is that a weak correlation exists. Which is what all the other figures have shown so far.

(in reply to CruelNUnsual)

Profile

Post #: 90

RE: Athiest nations are more peaceful? - 6/7/2009 9:01:13 PM

MissDominae

Posts: 94
Joined: 8/9/2008
Status: offline

Shall we begin a thread called "Everything I never wanted to know about Statistical theory?" I studied this at University as part of my Psych degree; hated it then and have just discovered I have no more love for it now~!

I would have typed this answer in Sanskrit except that no one would have understood one word of it......... hang on??? LOL - that means it would have been the same as the last few answers.

_____________________________

***~ Success can not be measured in the respect gained from others, only by the respect that, with complete self honesty and freed of ego or delusion, we are able to give ourselves.~***

(in reply to DomKen)

Profile

Post #: 91

RE: Athiest nations are more peaceful? - 6/8/2009 9:00:29 AM

LadyEllen

Posts: 10931
Joined: 6/30/2006
From: Stourport-England
Status: offline

I think if nothing else, its been proven that threads lacking statisticians are likely more peaceful.

E

_____________________________

In a test against the leading brand, 9 out of 10 participants couldnt tell the difference. Dumbasses.

(in reply to MissDominae)

Profile

Post #: 92

RE: Athiest nations are more peaceful? - 6/8/2009 9:28:27 AM

CallaFirestormBW

Posts: 3651
Joined: 6/29/2008
Status: offline

Rather than "atheist" countries (which, to me, speaks to something like the USSR and modern China's enforcement of an atheist state), I would ask whether any correlation relates to countries that do not enforce or reinforce -any- religious perspective, and retain a religion-neutral domestic policy.

I can say that the United States has -never- been religiously 'neutral'. This country was founded, in part, by religious fanatics looking for a place to practice their particular brand of fanaticism where it wouldn't get them killed. The fact that religions is mentioned at -all- in our Constitution belies any desire to remain a-religious in our political practice. We are a -religious- nation... the only bone of contention is which religion is going to obtain the majority pressure to push its will on the general public.

Freedom of religion and decision-making regarding relationships to other world powers comes not out of statements of "religious freedom", but out of true separation of religion and politics, where it is not even -mentioned- in a political venue, and where laws are determined based solely on the protection of the physical and fiscal resources of a country rather than on establishing -any- moral platform.

If we could just 'get' that concept, I'm pretty sure we -could- get past some of the stupid, senseless, life-destroying reasons that we feel compelled to beat the crap out of one another... alas, that's about as likely as Joseph Alois Ratzinger ( the Pope, Benedict XVI), Pat Robertson, Benjamin Netanyahu, and Robert Mugabe sitting down at a rickety picnic table in Georgia over a rack of pork ribs, mayonnaise cole slaw, and a nice shiraz, to discuss the potential benefits of encouraging radical feminism.

_____________________________

***
Said to me recently: "Look, I know you're the "voice of reason"... but dammit, I LIKE being unreasonable!!!!"

"Your mind is more interested in the challenge of becoming than the challenge of doing." Jon Benson, Bodybuilder/Trainer

(in reply to NihilusZero)

Profile

Post #: 93