Thursday, December 3, 2015

Counting Violence

[Johnathan Clayborn]

Okay, I'll preface this post with a disclaimer that the entire point of this is in no way related to the morality of violence or gun violence in any way shape or form. I felt like the vast majority of people who would read these comments would not be able to examine it from the realm of academic inquiry that I'm trying to present it from, and, frankly, that was a debate that I didn't want to have with many of the people on my page at this time.

As you may be aware, I'm actively working on my dissertation. Part of my dissertation involves the analysis of crime data, particularly as it relates to high school neighborhoods. But, it does particularly focus on violent crime. As a result, I've learned quite a lot of how violent crime is classified, categorized and reported over the last few years.

I read this article yesterday: https://www.washingtonpost.com/news/wonk/wp/2015/12/02/the-san-bernardino-mass-shooting-is-the-second-today-and-the-355th-this-year/
The title struck me immediately as sensationalized garbage. 355 Mass Shootings this year? How can that be?

The problem is that there is a bit of Baader-Meinhof Phenomenon going on. For those of you unfamiliar with this concept, it's what happens when you buy a new type of car, and then suddenly, you see THAT car everywhere. It's not that a bunch of other people rushed out and bought that same car. It's been there the whole time. The difference is that now you're used to what that car looks like - its part of your schema (the way your brain interprets information subconsciously) so you start to see it everywhere. (This is also the case when a new disease is suddenly increasing...it's not necessarily more prevalent, but our familiarity with it allows us to spot it more, but I digress).

All of the crime in the United States is reported back to the FBI. No, scratch that. All of the crime from the major metropolitan cities is reported back to the FBI. Small towns like Gila Bend or Podunkville, don't usually make it into the database. The data is collected, analyzed and reported on each year in a source called the Uniform Crime Report. One of the things that comes up is the question "if people from different cities are reporting the crime to the FBI, how do you get them to report it consistently?" It's a valid question. There's a huge Uniform Crime Reporting Handbook (https://www2.fbi.gov/ucr/handbook/ucrhandbook04.pdf) that breaks down every possible crime into different categories and classifications, and even "intensities". Suppose that a criminal stole something at knife-point (burglary), got into a knife fight (aggravated assault), then stole a car (grand theft auto), and then ran over a mailbox (property crimes) all in the same incident. You know, because criminals only limit themselves to one type of crime, right? The handbook includes a complex algorithm to report these crimes accordingly and consistently. In this particular case, Armed robbery would be the most heinous of these crimes, so the entire incident would be coded as that. Unless of course the perpetrator killed his attacker, then it would be coded as a homicide.

Now, that brings me back to this Washington Post article. 355 Mass Shootings? Really? Looking at the data, the media is getting this information from Reddit. That's right. Not the FBI. Not Homeland Security, DHS, or any other federal agency charged with our protection and safekeeping, but rather internet users on a chat board. Why is this important? Because it means that Reddit users are defining a "mass shooting" differently than the FBI, and as a result of this different definition, they have different numbers....very different numbers.

The Reddit Group defines a "Mass Shooting" as a single incident where a bullet strikes more than 4 people, including the shooter. So, if a shooter shoots three people in the leg and then shoots himself in the arm, then that gets counted by the Reddit Group and viola, another mass shooting.

The FBI, on the other hand, is much more stringent about how they define "Mass Shooting". For one thing, the FBI does not count domestic violence as a "Mass Shooting". That's a different type of crime that has a different type of personality and different reasoning behind it. Behavioral Analysis of these perpetrators reveal a vastly different profile than other mass shooters, so it's scored differently. Also, Gang Violence is not included in the FBI's count. So if there are two street gangs that have a beef over some turf, maybe MS13 is throwing down against some Bloods in south LA and they shoot each other up, that's not counted as a "mass shooting" by the FBI, because, again, behaviorally, it's a different profile, with different motivating factors and different outcomes and prevention. The FBI only defines a "mass shooting" as an incident were 4 or more people are killed, particularly by a shooter who has no connection to them and is targeting unarmed people with the intent to kill them.

The Washington Post article cites "Another mass shooting" in GA earlier yesterday. In it, 1 person was killed and 3 more were injured. That meets the Reddit Group's criteria for "Mass Shooting", but not the FBI's definition. See where I'm going with this? The FBI isn't counting that, but the Reddit Group is. This creates a huge difference in numbers.

To take an example, let's examine the numbers from 2013. According to the FBI's data there were 17 separate incidents that left 44 people dead and 42 people wounded. According to the Reddit Group for that same year, there were 363 events, 502 deaths, and 1,266 wounded. This is a huge discrepancy. One that, seemingly, over-inflates the severity of gun violence.

There are also reports that the US has more "mass shootings" than any other country in the world, but that's also a problem of definition. Many countries do not even have a formal definition of "mass shooting" and downplay the incidents. Also, even in countries where firearms are banned, such as China, there are still mass assaults and mass murders, but because no firearm is used, they can't be included, which automatically makes that statement true. It's the old "Audio Express" logic. "No one installs more car stereos for $1 than Audio Express". Well, considering that every other company either does it for free or charges more than a $1, then the statement is automatically true regardless of how many stereos they actually install.

According to the FBI's data, Violent Crimes in general (all types, including shootings), are going down: https://www.fbi.gov/news/stories/2014/november/crime-statistics-for-2013-released/image/violent-crime-graphic-from-crime-in-the-u.s.-2013-report.
In fact, all crime, across the board, is declining: https://www.fbi.gov/news/stories/2009/january/image/declining_crime_rates.jpg
This table, from the UCR, shows that since 1994, all crime, in every category, including gun violence, has been steadily and significantly trending down.
https://www.fbi.gov/about-us/cjis/ucr/crime-in-the-u.s/2013/crime-in-the-u.s.-2013/tables/1tabledatadecoverviewpdf/table_1_crime_in_the_united_states_by_volume_and_rate_per_100000_inhabitants_1994-2013.xls

And yet the media latches on to data compiled by the Reddit as their source for their statistics and people are up in arms over it. There is a definite discrepancy in the data. The FBI's data says the problem is getting better. The media's data says the problem is getting worse. Both sets of data cannot be right. In this particular case, since the FBI has had a longer track record of tracking this data more consistently than any other group, and they are the one's responsible for national security of this type, I'm going to look to their data and take it as being more credible first.