Liberals Lying about Guns Again

[houndawg]

Maybe. If want to ignore other factors like economics and law enforcement.

I think it's fair to say that. In fact I've been thinking about taking the time to look up something like per capita income by country then do something to take that into account when looking at the gun ownership rate/homicide rate association. Don't know how you'd quantify "law enforcement" to take that into account.

But that's the kind of thing you'd do if you were honestly evaluating. You'd propose factors that might have impacts on homicide rates then put them in a model that included data from all countries you had data for. You wouldn't just kind of arbitrarily select countries to compare the United States to.

And that relates to what I was getting at with the post at the start of this thread. The guy just selected a small handful of countries to compare the United States to with absolutely no basis for doing that. To me it REALLY smacks of "cherry picking."

Why don't you get on that and report back when the stats support your preconceptions?

[JohnStOnge]

Why don't you get on that and report back when the stats support your preconceptions?

I did do it before I came back here. And there's still no association with per capita gun ownership. I used the data at http://data.worldbank.org/indicator/NY.GDP.PCAP.CD" onclick="window.open(this.href);return false;. I Googled "per capita income by country" but everything I got was about per capita GDP. I guess it's basically the same thing.

Anyway when I combined that with the gun ownership rate and homicide rate info at the site I linked at the top of the thread I ended up with 103 countries for which gun ownership rate, per capita GDP, and homicide rate were all available.

There is significant negative correlation between homicide rate and per capita GDP. It's significant at the 99% confidence level. In other words, as per capita GDP (or I guess income) goes down, homicide rate goes up. There's a definite association such that poorer countries tend to have higher homicide rates. I don't think that's a shock.

But there's no indication of an association whereby homicide rate goes up as gun ownership rate goes up. I already mentioned it but you're still left with a negative correlation that's "not significant" there.

And when I "controlled" for income the association to the extent that there is one was still negative (i.e., homicide down when gun ownership rate goes up) when I "controlled" for per capita GDP. Not "significant' but negative' No suggestion at all that homicides tend to go up as gun ownership rates do.

Look, I think common sense tells us that if you're in a given country and you could somehow get rid of all access to guns the homicide rate for that country would go down at least SOME just because it's an efficient tool. But the idea that "more guns means more homicides" on a country by country basis just is NOT supported by the data. It's just not.

And you've got me completely backwards. What you need to understand about me is that data as well as the rules for how data are supposed to be analyzed drive my conceptions. I am not at all a person that starts off with a preconception than tries to distort the data or the analysis to fit it. I'm a person that loves to play with data like these, see what they say, the form my perceptions based on the results.

In fact, if you're worried about people who start off with a preconception what you really do need to look at people in this country who continue to focus on guns as the reason for the violence problem this country has to the extent that it has one. They're taking the easy way out and blaming inanimate objects for a problem that is caused by human behavior.

[houndawg]

Why don't you get on that and report back when the stats support your preconceptions?

I did do it before I came back here. And there's still no association with per capita gun ownership. I used the data at http://data.worldbank.org/indicator/NY.GDP.PCAP.CD" onclick="window.open(this.href);return false;. I Googled "per capita income by country" but everything I got was about per capita GDP. I guess it's basically the same thing.

Anyway when I combined that with the gun ownership rate and homicide rate info at the site I linked at the top of the thread I ended up with 103 countries for which gun ownership rate, per capita GDP, and homicide rate were all available.

There is significant negative correlation between homicide rate and per capita GDP. It's significant at the 99% confidence level. In other words, as per capita GDP (or I guess income) goes down, homicide rate goes up. There's a definite association such that poorer countries tend to have higher homicide rates. I don't think that's a shock.

But there's no indication of an association whereby homicide rate goes up as gun ownership rate goes up. I already mentioned it but you're still left with a negative correlation that's "not significant" there.

And when I "controlled" for income the association to the extent that there is one was still negative (i.e., homicide down when gun ownership rate goes up) when I "controlled" for per capita GDP. Not "significant' but negative' No suggestion at all that homicides tend to go up as gun ownership rates do.

Look, I think common sense tells us that if you're in a given country and you could somehow get rid of all access to guns the homicide rate for that country would go down at least SOME just because it's an efficient tool. But the idea that "more guns means more homicides" on a country by country basis just is NOT supported by the data. It's just not.

And you've got me completely backwards. What you need to understand about me is that data as well as the rules for how data are supposed to be analyzed drive my conceptions. I am not at all a person that starts off with a preconception than tries to distort the data or the analysis to fit it. I'm a person that loves to play with data like these, see what they say, the form my perceptions based on the results.

In fact, if you're worried about people who start off with a preconception what you really do need to look at people in this country who continue to focus on guns as the reason for the violence problem this country has to the extent that it has one. They're taking the easy way out and blaming inanimate objects for a problem that is caused by human behavior.

Of course you did.

[JohnStOnge]

Of course you did.

Yes I did. You guys need to stop disbelieving me just because I'm me and realize that I'm right about this. And by that I mean that the presentation I referenced at the start of this post was extremely misleading. It's pretty clear that the rate at which people own guns is not a major factor in country by country homicide rates. If you're focusing on the tool, you're focusing on the wrong thing.

I'm not saying I don't think it's a factor. Again, common sense tells you that it's an effective tool. But when you can't see an association in the data that tells you something. And when you look at another factor that makes sense, like per capita income, and you immediately and very easily see an association that tells you something.

I don't know why Liberals focus so much on the tool. But when Obama and others in his camp stand up there and act like the United States has the problems it has with violence because of the availability of guns they are full of crap. And really when they stand up there and act like the United States has a big violence problem in relative terms when considering the rest of the world they are also full of crap. It is not, in the context of the population of countries, an unusually violent country. It's a middle of the pack country in terms of violence. The average homicide rate for the 107 countries for which data are available at that Washington Post link I provided is 8.2 per 100,000 population. The United States homicide rate is 4.7 per 100,000 population.

People have GOT to stop buying the crap these people are serving.

[Skjellyfetti]

The average homicide rate for the 107 countries for which data are available at that Washington Post link I provided is 8.2 per 100,000 population. The United States homicide rate is 4.7 per 100,000 population.

People have GOT to stop buying the crap these people are serving.

Can you extract the data to include only OED countries?



Poverty is a huge factor in the prevalence of violent crime. For a reasonable comparison, you have to compare wealthy countries vs. wealthy countries or impoverished countries vs. impoverished countries.

[houndawg]

Of course you did.

Yes I did. You guys need to stop disbelieving me just because I'm me and realize that I'm right about this. And by that I mean that the presentation I referenced at the start of this post was extremely misleading. It's pretty clear that the rate at which people own guns is not a major factor in country by country homicide rates. If you're focusing on the tool, you're focusing on the wrong thing.

I'm not saying I don't think it's a factor. Again, common sense tells you that it's an effective tool. But when you can't see an association in the data that tells you something. And when you look at another factor that makes sense, like per capita income, and you immediately and very easily see an association that tells you something.

I don't know why Liberals focus so much on the tool. But when Obama and others in his camp stand up there and act like the United States has the problems it has with violence because of the availability of guns they are full of crap. And really when they stand up there and act like the United States has a big violence problem in relative terms when considering the rest of the world they are also full of crap. It is not, in the context of the population of countries, an unusually violent country. It's a middle of the pack country in terms of violence. The average homicide rate for the 107 countries for which data are available at that Washington Post link I provided is 8.2 per 100,000 population. The United States homicide rate is 4.7 per 100,000 population.

People have GOT to stop buying the crap these people are serving.

Its not just you, there are lots of other people that find what they want to find with stats too.

[kalm]

The average homicide rate for the 107 countries for which data are available at that Washington Post link I provided is 8.2 per 100,000 population. The United States homicide rate is 4.7 per 100,000 population.

People have GOT to stop buying the crap these people are serving.

Can you extract the data to include only OED countries?



Poverty is a huge factor in the prevalence of violent crime. For a reasonable comparison, you have to compare wealthy countries vs. wealthy countries or impoverished countries vs. impoverished countries.

Good luck!