Bayesian Analysis of US Mass Shooters.

[jeremyboycool]

I finished my analysis of US mass shooters. I used data from 1982 to 2018. There was another data set that went back to the 1960's, but this one from 1982 - 2018 seemed like a more reliable data set. It detailed where all the information came from, which I liked.

https://drive.google.com/file/d/1eO4...ew?usp=sharing

Feel free to share your opinions and if there are other statistics from that data set you'd like to see, it should not be very hard for me to punch something up.

[Philip]

Thank you for doing all this work!

So, it seems that the majority of weapons were obtained legally, about half the shooters had some history of mental health issues, and that the majority are between 18-46yo with a mean ~35yo.

After reading your analysis and looking at the statistics, it is still my opinion checking mental health history and vetting licensed people properly on a national level would do much more than changing the legal age of ownership.

[jeremyboycool]

Thank you for doing all this work!

So, it seems that the majority of weapons were obtained legally, about half the shooters had some history of mental health issues, and that the majority are between 18-46yo with a mean ~35yo.

After reading your analysis and looking at the statistics, it is still my opinion checking mental health history and vetting licensed people properly on a national level would do much more than changing the legal age of ownership.

I can understand that perspective. In debates I debate, but with my analyses my only goal is to provide information.


----

I also thought the number of weapons used that were obtained legally and the numbers of shooters with prior signs of mental health issues was concerning; it seems like an open door to do something. Something to look at, that I didn't put in the analysis, is the fact there were only 3 female shooters. So it is mostly males doing these shootings. I don't know why that is, but probably something that should be investigated.

[TonyT]

Very nice job!!!!!

I'd be interested in another factor and variable: the medications that the shooter was on prior to the shootings and/or during the shootings. In many cases shooters were on some form of psych meds and in some cases that data has not been made public.

[jeremyboycool]

I definitely agree that the details of mental health should be the next area of investigation. I am sure a statistical model could be made to throw red flags when a gun is purchased based on prior mental health issues. However, I don't have access to such data.

[Ken]

Any info on type of weapon used; hand gun, rifle, modified, automatic, etc.?

[jeremyboycool]

Any info on type of weapon used; hand gun, rifle, modified, automatic, etc.?

The details for the weapons are in the data set here: https://www.motherjones.com/politics...nes-full-data/

I'll have to think about how to display those details graphically.

[Easto]

I didn't read the attachment and I don't have a dog in this fight. But, when authorities gather together all these statistics don't they include suicides? Personally, I do not think it is fair to include suicide in gun violence statistics.

It's sort of like a life expectancy chart. For example I'm supposed to live until age 84. But if when I drive, I stop at every stop light and check the intersection and if I wear my seat belt, if I don't smoke and exercise regularly it adds something like 10 - 12 years to that figure. I find most statistics can be skewed or made meaningless if everyone just used an "ounce of prevention".

Sorry if this is a bit off topic, I just wanted to throw that out. Whenever I start seeing people throwing around statistics I take them with a grain of salt.

[TonyT]

I didn't read the attachment and I don't have a dog in this fight. But, when authorities gather together all these statistics don't they include suicides? Personally, I do not think it is fair to include suicide in gun violence statistics.

It's sort of like a life expectancy chart. For example I'm supposed to live until age 84. But if when I drive, I stop at every stop light and check the intersection and if I wear my seat belt, if I don't smoke and exercise regularly it adds something like 10 - 12 years to that figure. I find most statistics can be skewed or made meaningless if everyone just used an "ounce of prevention".

Sorry if this is a bit off topic, I just wanted to throw that out. Whenever I start seeing people throwing around statistics I take them with a grain of salt.

Correct. In another thread, jeremyboycool stated that he was a stats major and acknowledged the fact that stats don't actually prove out anything but they are useful, when accurate, to predict possible outcomes, correlate data and measure probablilites.

Even the stats you refer to re driving don't really determine length of life because there are so many variables involved. One could follow all of the suggested driving tips and that other guy who runs a red light changes everything.

[jeremyboycool]

I didn't read the attachment and I don't have a dog in this fight. But, when authorities gather together all these statistics don't they include suicides? Personally, I do not think it is fair to include suicide in gun violence statistics.

It's sort of like a life expectancy chart. For example I'm supposed to live until age 84. But if when I drive, I stop at every stop light and check the intersection and if I wear my seat belt, if I don't smoke and exercise regularly it adds something like 10 - 12 years to that figure. I find most statistics can be skewed or made meaningless if everyone just used an "ounce of prevention".

Sorry if this is a bit off topic, I just wanted to throw that out. Whenever I start seeing people throwing around statistics I take them with a grain of salt.

These data are about mass shooters and every observation in the data set is a mass shooting. Feel free to review the data set; that is why I provided a link to it, so you don't have to wonder about such things. Also, while I am not sure what life expectancy charts you are talking about, but in statistics generalized inferences from a sample size to a larger population is estimating the likely outcome when samples are taken randomly. Which forms a distribution around a central trend with what are known as standard deviations, as the variances around that center.

For example, let's say the estimated average height of US males is 69 inches. Now that does not mean if you are a US male then you are 69 inches tall, it means if I start taking random samples (random is the keyword) then on average I should get males who are close to 69 inches tall. In fact, with enough samples we'll find there is a normal variation around the average, which we can do all types of cool things with.


What you must keep in mind is that before the media, politicians and just the Internet, statistics is primarily used for science. Statistics was never meant to taken as factual, and it is unfortunate that it is so often presented that way; however, in science, the aim of statistics is to provide evidence. Evidence that is supposed to be considered alongside other evidence. The goal of the statistician is not to prove things, but to provide information.

I'll agree that politicians and the media abuses statistics to just flat out lie, and there is also a problem with the type of studies that are published, as studies are chosen based on how well they will sell and not how well they are conducted, but you can't blame the science for how people misuse it. The truth is when you push all that crap aside, statistics is a valid scientific method which does contribute to the exploration of the reality around us. I don't think the solution here is to just brush off all of statistics as worthless, as it is not, there are sound philosophies behind statistical methods, the solution is to learn how to read the stats. If the public was more informed about the nature of these methods then it would be harder to lie to them with statistics. My point is: Don't blame the science, blame the people abusing it.

[TonyT]

I'll agree that politicians and the media abuses statistics to just flat out lie, and there is also a problem with the type of studies that are published, as studies are chosen based on how well they will sell and not how well they are conducted, but you can't blame the science for how people misuse it. The truth is when you push all that crap aside, statistics is a valid scientific method which does contribute to the exploration of the reality around us. I don't think the solution here is to just brush off all of statistics as worthless, as it is not, there are sound philosophies behind statistical methods, the solution is to learn how to read the stats. If the public was more informed about the nature of these methods then it would be harder to lie to them with statistics. My point is: Don't blame the science, blame the people abusing it.

Well said.

Another example is Wall Street. If stats actually proved true & false then we'd all be rich. Same goes with gambling. Roll those bones baby! The casinos know the odds by evaluating stats and thus remain in business.

There is science in obtaining, organizing and coordinating stats. There is art in evaluating them.

[jeremyboycool]

I made a plot of the type of weapons mass shooters like to bring to the shooting. This graph does not tell you how many of each type of weapons a shooter had with them, as they typically had more than one weapon, it is just a graph showing what types are most common. Sorry for the double entries on the graph, but there was some issues that I could not figure out which made two similar text entries be seen as different entries to R.