Episode 1 - Gun Control Statistics

Hey all, going to start a series analyzing media dishonesty to try and encourage people to critically analyze what they hear on the news and other popular publications. If there's an interest in a series like this please follow and comment so I know to continue creating more content!

The article

On February 22nd, 2018 an article was published by Christopher Ingraham for the Washington Post called "The real reason Congress banned assault weapons in 1994 - and why it worked." After reading the article I was yet again disappointed by an intellectually dishonest representation of statistics for an agenda. This sort of thing is not new, but when these popular publications post articles with cherry picked stats that the public have learned to trust as reality, and it relates to a fundamental right granted by the United States constitution. Then we get into dangerous territory. As such, I feel it is my civic duty to bring to light some of the dishonesty displayed in the aforementioned article.

The setup

The article's title and opening sentence both contain the term "assault weapon" which in itself is a rather ambiguous term that's used for media sensationalism more than anything. I won't complain too much on that one, pretty widely used this way and there are more important things to discuss here. They go on to talk about some survey results for support on a ban of "assault weapons." The next paragraph is the one where they concede some points to attempt to appeal to the 2nd amendment advocates that may be reading. Here's a couple quotes from the article.

• "The NRA correctly points out that assault weapons are used only in a tiny fraction of gun crimes." Source
• "The gun rights group also notes that a federally funded study of the previous assault weapons ban, which was in place from 1994 to 2004, concluded that “the ban’s impact on gun violence is likely to be small at best, and perhaps too small for reliable measurement.”" Source

They then write off the assault weapons ban as never being intended to be the ultimate fix for gun crimes saying it was only meant to cut back on mass shooting frequency and lethality. There is some extra fluff for emotional appeal talking about a couple of the shootings that took place, leading into a description on how/when the assault weapon ban got implemented. Before they go into the data where my disagreement lies, they also mention what the ban entails.

• "The 1994 law included a ban on 18 specific models of assault weapons, as well as a ban on any firearms containing certain military-style features, like a bayonet mount, a flash suppressor or a folding stock. It also banned high-capacity magazines capable of holding more than 10 bullets. The bill allowed individuals already in possession of such weapons to keep them." Source

What's good, what's suspicious, and what's dishonest

I'm uncertain about copyright restrictions on copying the exact graph they have on their website, feel free to check the source from any of the quotes above to see it. Here is the same metrics plugged into a generic Excel sheet and charted.

Data Source*

* no source for underlying data, just the publication presenting it

The good

Here's the good thing I took away from the article. It does go to show that when comparing 1994-2004 to 2004-2014 that there has been an alarming increase in mass shootings assuming we can trust the underlying data (more on that later). We already know in the years between 2014 and now that mass shootings are becoming more common. So it does start a conversation about an important topic.

The suspicious

Here's the suspicious part, the author chose an unconventional definition of mass shooting as 6+ casualties and has left us with reason to question the underlying data. Let's explore this for a minute. When tasked with reporting on "active shooter incidents" the FBI used the definition as "an individual actively engaged in killing or attempting to kill people in a populated area" (source). The federal government says 4+ casualties is the definition of a mass shooting later in the same fbi article when discussing the "Investigative Assistance for Violent Crimes Act of 2012." www.shootingtracker.com also uses the 4+ bar for their reporting. So why 6+ for the study? No reason is presented for this.

Also, it seems there isn't large swaths of good data regarding mass shootings for the years he analyzed from the FBI or other typical sources (see this article from NPR referencing the lack of good data out there) so I'm wondering where this data is from? Even for the years that the FBI collected data, such as this data from 2000-2013 there are some questions as to the validity of the information due to how reports were collected and how it left room for listing incidents with no casualties under certain situations. See this article from the New York Post disputing that data.

I decided to look for the data from the article to see what I could find. I went to the article to look for a link to the data. Here's what they said, "Klarevas has compiled data on gun massacres involving six or more fatalities for the 50 years before 2016." I noticed a link to another article with the same data in it here. That article says more or less the same thing about the source but this time mentions that the research was done for a book named "Rampage Nation" written by Louis Klarevas. Okay, so I guess I need to take a look at the book for sources. Did some Googling around but I wasn't having any luck finding a list of sources. I did find this review someone posted on goodreads that implies that Klarevas collected the data himself...

Okay, so that's conveniently hard to dispute then isn't it. I'll be digging into this further to strengthen my talking points on this issue, but for the time being let's just place a nice little red flag here. My next step will be to look at the physical copy of the book in hopes that some information on where to find the data will be listed there, It would be great if he was able to collect the data accurately and post it publicly, the data would be quite valuable. Until then, take the underlying data with extreme caution.

The Dishonest

Now for the dishonest part, the data shows a difference in mass shooting deaths and paints the assumption that a limited weapon ban caused the difference. However, if you look at other data you'll see that all violent crime was falling during the span of the weapon ban. Comparing all mass shooting deaths from 1984-1994 (Dec 31st 1993) to 1994-2004 (Dec 31st 2003) to violent crimes in those same spans produces a similar result.

Note: Violent crimes here were collected from disastercenter.com looking at the compiled FBI UCS annual crime reports section under the "Violent" column

In addition to all violent crime FBI Statistics, all interpersonal violent deaths, as well as only gun related interpersonal violent deaths also followed a similar pattern as collected from the vital registrations records (as presented by vizhub.healthdata.org).

Note: When analyzing the healthdata statistics above, the url is not dynamic and therefore requires you filter properly. It also splits results between male and female so to get records to match mine you'll need to analyze the data by taking the sum of both numbers.

It seems far more logical to me that the overall trend was a decline in violent crime, not that it was caused by the weapon ban. From the verbiage of the ban, it wouldn't have seemed that it would have much of an impact either. The fact is, that ban only stopped new sales of very specific kinds of weapons. Some of the restrictions were cosmetic driven based on accessories on the gun, other restrictions were around functionality such as the magazine size. You could still buy a semi-automatic rifle that would be absolutely sufficient for a mass shooting if that was your prerogative.

Another reason for concern with the theory that the ban was the reason is that mass shootings did not elevate back up to the previous rate after the ban was lifted. Instead, they went much higher. Your first reaction might be to say this solidifies the idea that the ban worked. However, if that were the case you would actually expect the rate to head towards a similar number as before the ban, not one much larger. This tells us that other variables are at play. If the experiment is as simple as a control group and a variable group with one variable changing (gun laws) then one would expect the years after the gun laws expired to be within the variance of the previous deaths adjusted with a linear growth/decline rate from before the ban. Here's another very important thing to consider when analyzing this data, especially if you think you're going to derive some relevant point from it, the number of mass shooting deaths is so vastly small compared to total deaths in the US and even that of violent crime death or gun deaths that almost any outlier in the data would skew everything. Look at the numbers side by side for scale.

To be clear so I'm not taken out of context, I'm not saying that the ban had no impact on deaths. I'm saying that we can't infer that it did with the current data/analysis available. It's easy to cherry pick a chart that paints the picture you want but by adding only a couple other variables we can see other correlations which means this article and the premise it's built on is inconclusive and shouldn't be reported as fact.

Summary

So, why do I feel the need to post this? Firstly, because it irks me when people draw conclusions from inconclusive data that will be believed by large swaths of the public... 14 children and 3 adults were murdered and people are looking for answers. When you post an article like this that is not factually sound, you play on peoples emotions to lobby for gun control that we don't even know would help. We can't talk about solutions if media and/or politicians are simply lying to try and push their agenda. People need a clear understanding of what outcomes can be expected from different ideas for this issue. Most people don't have time to dig into every thing that's said to find out if it's true, they have jobs, families, hobbies, passions, and any number of other things they need to do with their time. So quit being bad journalists...

Until we can clear up some of these facts and have a more detailed analysis, I propose a new headline for this article. "Some information regarding the assault weapons ban in 1994 - and why we don't have a clue if it helped."

If you enjoy this content, please consider upvoting/following.

This post is unofficially brought to you by Zipline Copper Alt from Zipline Brewing Co. here in my home state of Nebraska. If you want to encourage more content like this, consider a donation. Creating content is easiest when paired with the smooth finish of a well crafted beer, and beer is most delicious when it is generously funded by those who enjoy good content!

Bitcoin (BTC): 1Q1fwSzXn9ECJAusqCFmEqSawsErahTyVJ
Ethereum (ETH): 0x05a90d283f60752a71a0aace07daa2f5a14c1106