Recommend
1 
 Thumb up
 Hide
31 Posts
1 , 2  Next »   | 

BoardGameGeek» Forums » Everything Else » Religion, Sex, and Politics

Subject: Looking at assaults in Catholic countries rss

Your Tags: Add tags
Popular Tags: [View All]
James Myers
United States
Redmond
WA
flag msg tools
Mandelbrot/Simurgh hybrid etc etc
badge
I made both of these fractals, hurray!
Avatar
mbmbmbmbmb
http://www.r-bloggers.com/deaths-from-assault-over-time-in-4...

Base URL is from a stats blog. Caveat: you'll have to skip through some code if you want to read the source directly.

Some interesting graphs:



Remember that the above is a logarithmic scale (so that Columbia didn't squish that map by being an outstanding hellhole of violence.) The author is looking at male vs. female violence here, looking at if Catholicism protects women from violence, in a sense.



Going with the bottom graph -- which is the author's conclusion, though this is where statistics becomes art on top of science -- Catholicism doesn't necessarily protect women (that is, it doesn't favor men as the targets of assault) -- when you put traditionally Catholic countries also grouped by region (as Latin America vs. not,) you don't find an effect when you look at male vs. female assault victim ratios.

Pretty interesting and topical to RSP, I thought.
1 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
James Myers
United States
Redmond
WA
flag msg tools
Mandelbrot/Simurgh hybrid etc etc
badge
I made both of these fractals, hurray!
Avatar
mbmbmbmbmb
Koldfoot wrote:
I looked at the graphs for a while, mildly interested.

Then it occurred to me that a far more interesting study could be done on the level of violence, violence against women, and the corrupting nature of billions of dollars of drug money vs. oil money in Catholic vs. Muslim countries.


That would be interesting.

Presumably, you'd expect no effect -- and it's a little hard to hash out, really. We don't have lots of oil-rich Catholic countries, so it would be a bit of an empty cell. But sure, in theory, you'd expect Catholicism = Islam (if you're looking at it from a particular perspective,) and you would expect drug money and oil money to be equally corruptive... maybe.

I think the general idea is that Islam treats women more like property than Catholicism, so you would see less violence against women. It's not like, say, Hinduism, where honor killings are (remotely) prevalent... although I don't really know the differences in those rates.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
casey r lowe
United States
butte
Montana
flag msg tools
mb
Koldfoot wrote:
I looked at the graphs for a while, mildly interested.

Then it occurred to me that a far more interesting study could be done on the level of violence, violence against women, and the corrupting nature of billions of dollars of drug money vs. oil money in Catholic vs. Muslim countries.
you need stop the deflections and realize the root cause: catholicism is a religion of violence
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Boaty McBoatface
England
County of Essex
flag msg tools
badge
Avatar
mbmbmbmbmb
Koldfoot wrote:
single sentences wrote:
Koldfoot wrote:
I looked at the graphs for a while, mildly interested.

Then it occurred to me that a far more interesting study could be done on the level of violence, violence against women, and the corrupting nature of billions of dollars of drug money vs. oil money in Catholic vs. Muslim countries.
you need stop the deflections and realize the root cause: catholicism is a religion of violence


Agreed.

Pedophiles. And protecting pedophiles are horrible offenses of violence and condoning violence.

Why can't we agree suicide bombings are horrible acts of violence that are worthy of condemnation?

One of us has our head in the sand.
Has any one here said they are not?

According to the GPI, the ten most peaceful countries, in order, from 2013 to 2014 were Iceland, Denmark, Austria, New Zealand, Switzerland, Finland, Canada, Japan, Belgium, and Norway. The most violent countries were Syria, Afghanistan, South Sudan, Iraq, Somalia, Sudan, Central African Republic, Democratic Republic of the Congo, Pakistan, and North Korea.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Boaty McBoatface
England
County of Essex
flag msg tools
badge
Avatar
mbmbmbmbmb
Koldfoot wrote:
slatersteven wrote:
Koldfoot wrote:
single sentences wrote:
Koldfoot wrote:
I looked at the graphs for a while, mildly interested.

Then it occurred to me that a far more interesting study could be done on the level of violence, violence against women, and the corrupting nature of billions of dollars of drug money vs. oil money in Catholic vs. Muslim countries.
you need stop the deflections and realize the root cause: catholicism is a religion of violence


Agreed.

Pedophiles. And protecting pedophiles are horrible offenses of violence and condoning violence.

Why can't we agree suicide bombings are horrible acts of violence that are worthy of condemnation?

One of us has our head in the sand.
Has any one here said they are not?


Uhhhhhhhhhhhhhh.

Not sure if serious.........

Ok. I'll take the bait.

I condemn suicide bombings and the religion that condones it with the promise of a glorious life in heaven with 72 virgins. I also condemn pedophiles and the religion that conspired not only to protect them, but move them around to find more victims.

Your turn.

Edit: I see you pasted in a whole paragraph from Wikipedia while I was writing. I quoted your initial comment, which I suspect even you found too embarrassing to let stand. But go ahead: your turn.
NO it was not from Wikipedia, so try again.

I will condemn any act of violence (sexual or otherwise) and any religious authority that condones (or encourages) it (or condones or encourages any criminal act) or covers it up (and have here on RSP many times).
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
casey r lowe
United States
butte
Montana
flag msg tools
mb
Koldfoot wrote:
I condemn suicide bombings and the religion that condones it with the promise of a glorious life in heaven with 72 virgins. I also condemn pedophiles and the religion that conspired not only to protect them, but move them around to find more victims.

the majority of muslims condemn suicide bombings
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Boaty McBoatface
England
County of Essex
flag msg tools
badge
Avatar
mbmbmbmbmb
Koldfoot wrote:
single sentences wrote:
Koldfoot wrote:
I condemn suicide bombings and the religion that condones it with the promise of a glorious life in heaven with 72 virgins. I also condemn pedophiles and the religion that conspired not only to protect them, but move them around to find more victims.

the majority of muslims condemn suicide bombings
And? The majority of Catholics condemn child rape.

We both agree (apparently) on that point.

Now, you condemn suicide bombings and the religion that promises suicide bombers an exalted place in heaven.

This is not about them. They are not involved in this conversation. You and I and the Wikipedia guy are the ones involved in this conversation. They are not. And furthermore, you are wrong. But that doesn't matter in this context.

Shame I did not quote Wikipedia.

OK lets condemn the religion, and not the particular sect, so are you going to condemn Christianity?
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Boaty McBoatface
England
County of Essex
flag msg tools
badge
Avatar
mbmbmbmbmb
Koldfoot wrote:
You first.

Take the tiniest little baby step to condemn suicide bombers.
You can't do it, can you?

Can you condemn the religion that stones gay people to death?
I have already said I condemn all acts of violence in the name of religion.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Boaty McBoatface
England
County of Essex
flag msg tools
badge
Avatar
mbmbmbmbmb
Koldfoot wrote:
You are pathetic.

I condemn all acts of violence by black people against police, and police against black people. And I condemn violence committed in the name of English football. I am against violence perpetrated against the unborn. I am against violence against the perpetrators of such vile crimes against humanity.

You know where I stand. You don't even know where you stand. No fucking clue.

Why am I pathetic, I did what you asked, I condemned all religiously motivated violence. You (however, I note) refuse to.

So are you going to do what I did, and condemn all violence in the name of religion?

Are you going to do what I did and condemn all religious organizations that preach breaking the law?

I shall say it again (for the hard of reading), I condemn all violence committed by (and in the name of) religion. How can I be any clearer? That is all violence and any religion.

 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
casey r lowe
United States
butte
Montana
flag msg tools
mb
Koldfoot wrote:
Now, you condemn suicide bombings and the religion that promises suicide bombers an exalted place in heaven.

only a minority of muslims think suicide bombings will get you a place in paradise

Koldfoot wrote:
This is not about them. They are not involved in this conversation.

they are but im not surprised that went over your head

Koldfoot wrote:
And furthermore, you are wrong.

about what
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Neon Joe, Werewolf He-yump
United States
Lincoln
NE
flag msg tools
badge
Avatar
mbmbmb
Koldfoot wrote:
You are pathetic.

I condemn all acts of violence by black people against police, and police against black people. And I condemn violence committed in the name of English football. I am against violence perpetrated against the unborn. I am against violence against the perpetrators of such vile crimes against humanity.

You know where I stand. You don't even know where you stand. No fucking clue.



Fan-fucking-tastic. Gold star for you.

Can we get back to the original topic, or do you need yet another thread to discuss how horrible you think Muslims are?
2 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
James Myers
United States
Redmond
WA
flag msg tools
Mandelbrot/Simurgh hybrid etc etc
badge
I made both of these fractals, hurray!
Avatar
mbmbmbmbmb
Koldfoot wrote:
Couple things. First: there are several oil rich catholic countries. Mexico and Venezuala spring to mind. Afghanistan and Turkey are likewise Muslim drug countries, although I'm not sure we could call them rich.

But that was not the point.

Violence in these Latin American countries (which happen to be poor and catholic) is mostly due to the violent nature of the drug trade.

Vast amounts of money are a corrupting influence on poor countries regardless of the legality of the source and the religion.

The religion may be (fill in the blank) but their God is money.


I'm not sure drug trade is necessarily violent -- e.g., Amsterdam, Colorado, etc. I believe there have been violent crime downticks or null changes, but I'm not sure about that.

I'm pretty ignorant about oil production per country, here's a link if any readers are too:

https://en.wikipedia.org/wiki/List_of_countries_by_oil_produ...

Quote:
The religion may be (fill in the blank) but their God is money.


Agreed.

That would be interesting to look at as well -- do different religions interact w/ low religiosity folks differently -- e.g., Hinduism has following money as a valid (early) life path, that sort of thing -- but that would be insane to try to study as a psychology... that's more in the land of an ethnographer, I figure.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Steve
Thailand
flag msg tools
My take on the original post is:

The graph is about deaths or killings and the title is about "assaults".

There is a big difference between a murder and an assault.

I assume the deaths/killings are only murders. Natural causes and executions, etc. are not counted.


 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
James Myers
United States
Redmond
WA
flag msg tools
Mandelbrot/Simurgh hybrid etc etc
badge
I made both of these fractals, hurray!
Avatar
mbmbmbmbmb
Steve1501 wrote:
My take on the original post is:

The graph is about deaths or killings and the title is about "assaults".

There is a big difference between a murder and an assault.

I assume the deaths/killings are only murders. Natural causes and executions, etc. are not counted.


The link has more information.

Yes, it's only looking at assaults that resulted in deaths. I think there are some instances where that isn't necessarily murder, but I don't know the particulars there.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Steve
Thailand
flag msg tools
OK, so I need to add "killings" in self-defense.

Still my point is valid. There is a big difference between assaults and killings [= murders plus killing in self-defense]. It just seems strange to use the word assault as the only word of what one is talking about when one is talking about killings.

In my dialect "assault" is a fist fight.

Legally the term would be "assault and battery". Assault therefore seems legally to be about threatening words, or something.

 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Steve
Thailand
flag msg tools
The post said I have to wade through "code" to get to the "meat" of the article.

At 69 years of age, I don't try to "wade thru code". So, I didn't see the article. And, for some reason until now the lower graph never showed up on my computer.

My 2nd take on the graphs you posted is:

I see no correlation between Catholicism and killing of women.

I see 2 "curves". The bottom one shows a flat line with scatter. In the top right I see a group of Latin American nations that are outliers of all the rest. There is no "line" just a scatter. Since I don't see the data points for % Catholic it is hard to say much.

Therefore, I conclude that there is something about Catholic nations that fused native American cultures and Spanish/Portuguese cultures [which were Catholic].

In the top left "corner" we also see South Africa. It is not as far above the flat line but it is above it. There should be a cause for this also.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
James Myers
United States
Redmond
WA
flag msg tools
Mandelbrot/Simurgh hybrid etc etc
badge
I made both of these fractals, hurray!
Avatar
mbmbmbmbmb
Steve1501 wrote:
The post said I have to wade through "code" to get to the "meat" of the article.

At 69 years of age, I don't try to "wade thru code". So, I didn't see the article. And, for some reason until now the lower graph never showed up on my computer.

My 2nd take on the graphs you posted is:

I see no correlation between Catholicism and killing of women.

I see 2 "curves". The bottom one shows a flat line with scatter. In the top right I see a group of Latin American nations that are outliers of all the rest. There is no "line" just a scatter. Since I don't see the data points for % Catholic it is hard to say much.


You have the same conclusion as the author (and myself, I found it persuasive.)

For reference, those grey areas around the curve are error bars -- it's showing what you said to be demonstrably true. Though we notice a slight positive correlation, it isn't shown to be statistically significant.

I'm not sure why you scare-quoted code -- it simply is what it is. I just posted something straight from a statistics blog I read, it occasionally shows some political or religious data.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Vapix
msg tools
mbmbmb
Graphing those countries vs e.g. their GINI coefficient would be more useful than using religion.

https://en.wikipedia.org/wiki/Gini_coefficient
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
James Myers
United States
Redmond
WA
flag msg tools
Mandelbrot/Simurgh hybrid etc etc
badge
I made both of these fractals, hurray!
Avatar
mbmbmbmbmb
Vapix wrote:
Graphing those countries vs e.g. their GINI coefficient would be more useful than using religion.

https://en.wikipedia.org/wiki/Gini_coefficient


The guy had a particular research question and he answered it.

Why, exactly, do you think income inequality is related to male vs. female death from assault ratio? It's tenable but I dunno.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Vapix
msg tools
mbmbmb
As far as I can see, the conclusions claimed can't be drawn from the data graphs you showed. It looks like deliberate BS actually. Unless the original includes controls for a bunch of other stuff (GINI is just an example) the conclusions are worthless.

Remove the high-GINI countries and the three graphs are almost flat. Which suggests GINI, or at least a different type of index, might be a significantly more important factor than religion.

At a glance the raw data suggests that the spread (male vs female) is bigger for Catholic countries, which is the opposite of what the text claims.
Of course it's hard to tell, because the facts have been carefully hidden by the use of a log scale, rather than showing % male vs female on a linear scale for the base data too.

 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
James Myers
United States
Redmond
WA
flag msg tools
Mandelbrot/Simurgh hybrid etc etc
badge
I made both of these fractals, hurray!
Avatar
mbmbmbmbmb
Vapix wrote:
As far as I can see, the conclusions claimed can't be drawn from the data graphs you showed. It looks like deliberate BS actually. Unless the original includes controls for a bunch of other stuff (GINI is just an example) the conclusions are worthless.

Remove the high-GINI countries and the three graphs are almost flat. Which suggests GINI, or at least a different type of index, might be a significantly more important factor than religion.

At a glance the raw data suggests that the spread (male vs female) is bigger for Catholic countries, which is the opposite of what the text claims.
Of course it's hard to tell, because the facts have been carefully hidden by the use of a log scale, rather than showing % male vs female on a linear scale for the base data too.


If you can't understand a log scale (or when using one is appropriate) that's a problem with you, not the graph.

Ridiculously, I'm pretty sure you're agreeing with the author's position and missing that entirely.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Vapix
msg tools
mbmbmb
Terwox wrote:
Vapix wrote:
As far as I can see, the conclusions claimed can't be drawn from the data graphs you showed. It looks like deliberate BS actually. Unless the original includes controls for a bunch of other stuff (GINI is just an example) the conclusions are worthless.

Remove the high-GINI countries and the three graphs are almost flat. Which suggests GINI, or at least a different type of index, might be a significantly more important factor than religion.

At a glance the raw data suggests that the spread (male vs female) is bigger for Catholic countries, which is the opposite of what the text claims.
Of course it's hard to tell, because the facts have been carefully hidden by the use of a log scale, rather than showing % male vs female on a linear scale for the base data too.


If you can't understand a log scale (or when using one is appropriate) that's a problem with you, not the graph.

Ridiculously, I'm pretty sure you're agreeing with the author's position and missing that entirely.

FWIW:
* I know what a log scale is, and I know what they're normally used for.
* My criticism was valid: a log scale is unsuitable for that data, and while it might have a useful purpose in the original article, it should not have been selected as the only source of raw data for the OP.

I can't see the bottom three pictures any more, but as I remember them, they didn't actually support your claim (which is a bit fuzzy at best - it may reflect the study, but isn't supported by the data you chose to include in your post):
Quote:
Going with the bottom graph -- which is the author's conclusion, though this is where statistics becomes art on top of science -- Catholicism doesn't necessarily protect women (that is, it doesn't favor men as the targets of assault) -- when you put traditionally Catholic countries also grouped by region (as Latin America vs. not,) you don't find an effect when you look at male vs. female assault victim ratios.



This section is taken from my memory of those three graphs - I thunk it's accurate enough, but I might update it if I can get those pix back.

The three graphs showed countries as points against Y-axis: % M vs F assaults, and X-axis: % catholics in a country.

Taken as a whole, the graphed data, including the Latin American countries, showed that a higher % of males than females are assaulted in "catholic countries" compared to others.

OTOH you could see at a glance that by excluding the Latin American countries you could get a "flat" line (y value more-or-less constant), and that reduced data set included a wide range on the x-axis (% catholics). If excluding those countries was valid (which I strongly doubt) this would lead to the contrary result (that the % is constant regardless of the % catholics)

This doesn't make the second result true through. It's actually a strong indicator that there is at least one factor that's more important than what's shown on the graph.

In this case it interesting to consider that Latin America is relatively violent - but probably not because it's largely catholic. Here you'd look to other kinds of explanations. One example is the kind of thing the GINI index measures.

There's also the issue with the limited set of countries - the lack of African and Asian countries (at least in the data you showed in the OP) is suspicious. For example it's the kind of thing someone might do to remove obvious evidence that there are important factors other than the independent variable (% catholic) that affect the independent variable (%M vs F assaults).

Why remove well over half the world's population (China, India, Indonesia, Nigeria (12% catholic), Bangladesh, ...)? Perhaps because in a serious study it's necessary to control for every factor that's known or likely to affect the dependent variable, which increases the scale and cost of data collection. But there's another possible explanation - leaving out inconvenient controls also makes it a lot easier to "rig" the results.


I can't tell if this is an issue with the original work or the data you selected for your original post. And I'm not going to check because the restricted set of countries is a good indicator that the original work is not worth reading.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
James Myers
United States
Redmond
WA
flag msg tools
Mandelbrot/Simurgh hybrid etc etc
badge
I made both of these fractals, hurray!
Avatar
mbmbmbmbmb
Vapix wrote:
Terwox wrote:
Vapix wrote:
As far as I can see, the conclusions claimed can't be drawn from the data graphs you showed. It looks like deliberate BS actually. Unless the original includes controls for a bunch of other stuff (GINI is just an example) the conclusions are worthless.

Remove the high-GINI countries and the three graphs are almost flat. Which suggests GINI, or at least a different type of index, might be a significantly more important factor than religion.

At a glance the raw data suggests that the spread (male vs female) is bigger for Catholic countries, which is the opposite of what the text claims.
Of course it's hard to tell, because the facts have been carefully hidden by the use of a log scale, rather than showing % male vs female on a linear scale for the base data too.


If you can't understand a log scale (or when using one is appropriate) that's a problem with you, not the graph.

Ridiculously, I'm pretty sure you're agreeing with the author's position and missing that entirely.

FWIW:
* I know what a log scale is, and I know what they're normally used for.
* My criticism was valid: a log scale is unsuitable for that data, and while it might have a useful purpose in the original article, it should not have been selected as the only source of raw data for the OP.

I can't see the bottom three pictures any more, but as I remember them, they didn't actually support your claim (which is a bit fuzzy at best - it may reflect the study, but isn't supported by the data you chose to include in your post):
Quote:
Going with the bottom graph -- which is the author's conclusion, though this is where statistics becomes art on top of science -- Catholicism doesn't necessarily protect women (that is, it doesn't favor men as the targets of assault) -- when you put traditionally Catholic countries also grouped by region (as Latin America vs. not,) you don't find an effect when you look at male vs. female assault victim ratios.



This section is taken from my memory of those three graphs - I thunk it's accurate enough, but I might update it if I can get those pix back.

The three graphs showed countries as points against Y-axis: % M vs F assaults, and X-axis: % catholics in a country.

Taken as a whole, the graphed data, including the Latin American countries, showed that a higher % of males than females are assaulted in "catholic countries" compared to others.

OTOH you could see at a glance that by excluding the Latin American countries you could get a "flat" line (y value more-or-less constant), and that reduced data set included a wide range on the x-axis (% catholics). If excluding those countries was valid (which I strongly doubt) this would lead to the contrary result (that the % is constant regardless of the % catholics)

This doesn't make the second result true through. It's actually a strong indicator that there is at least one factor that's more important than what's shown on the graph.

In this case it interesting to consider that Latin America is relatively violent - but probably not because it's largely catholic. Here you'd look to other kinds of explanations. One example is the kind of thing the GINI index measures.

There's also the issue with the limited set of countries - the lack of African and Asian countries (at least in the data you showed in the OP) is suspicious. For example it's the kind of thing someone might do to remove obvious evidence that there are important factors other than the independent variable (% catholic) that affect the independent variable (%M vs F assaults).

Why remove well over half the world's population (China, India, Indonesia, Nigeria (12% catholic), Bangladesh, ...)? Perhaps because in a serious study it's necessary to control for every factor that's known or likely to affect the dependent variable, which increases the scale and cost of data collection. But there's another possible explanation - leaving out inconvenient controls also makes it a lot easier to "rig" the results.


I can't tell if this is an issue with the original work or the data you selected for your original post. And I'm not going to check because the restricted set of countries is a good indicator that the original work is not worth reading.


...

Region is the alternate explanation over religion.

Perhaps that'd Gini dependant.

To be more frank:

Research question: Is Catholicism protective of women?

Research answer: Only if you treat Latin America as a specific group. It might be that Latin America exhibits violence differently, rather than Catholicism.

Your response: No, it's GINI, and you graphed it wrong.

My response: Yeah, Gini might be the reason between regions. Why do you think Gini affects gendered violence?

Your response: No, it's GINI, and you graphed it wrong.

My response: RTFS already.
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
Vapix
msg tools
mbmbmb
Terwox wrote:
...

Region is the alternate explanation over religion.

Perhaps that'd Gini dependant.

To be more frank:

Research question: Is Catholicism protective of women?

Research answer: Only if you treat Latin America as a specific group. It might be that Latin America exhibits violence differently, rather than Catholicism.

Your response: No, it's GINI, and you graphed it wrong.

My response: Yeah, Gini might be the reason between regions. Why do you think Gini affects gendered violence?

Your response: No, it's GINI, and you graphed it wrong.

My response: RTFS already.

Quote:
Research answer: Only if you treat Latin America as a specific group. It might be that Latin America exhibits violence differently, rather than Catholicism.

This isn't an answer at all. It's "lets get the answer we want by ignoring the data points we don't like".

In a sense it's what our discussion is about. Removing "inconvenient" data points without a rational reason is a favorite technique of pseudo-researchers who are misusing science.

If something like GINI partially explains the data then you would need 3 sets of 3 graphs as follows:
* % M/F assault vs GINI
* % M/F violence vs % Catholiic
* % M/F violence vs BOTH (there are techniques for this, but the graphs aren't directly comparable to the other two sets)

You can persuade me the source material is interesting easily: show that it has a good statistical reason for removing the data points for Latin America, backed up with numbers, and integrated into the overall study - for example by pasting in the "missing graphs" as I explained above. Note that using "region" can only be valid if the analysis includes all the countries in the world. This is why I mentioned the large proportion of the world that the study casually ignores. If you want to use GINI (or similar approaches), you can't exclude countries that have few Catholics.

If the study just uses the same claim as you just did:
Quote:
It might be that Latin America exhibits violence differently, rather than Catholicism.
then the original is junk.

A claim to remove otherwise significant data has to have substantial supporting evidence, or the stated result isn't a result at all. It's just "depending on which data points we include we can support either of the two possible conclusions".
 
 Thumb up
 tip
 Hide
  • [+] Dice rolls
1 , 2  Next »   | 
Front Page | Welcome | Contact | Privacy Policy | Terms of Service | Advertise | Support BGG | Feeds RSS
Geekdo, BoardGameGeek, the Geekdo logo, and the BoardGameGeek logo are trademarks of BoardGameGeek, LLC.