One of my favorite stats-esque topics is graphs. Specifically how we misrepresent with graphs, or how we can present data better. This weeks gun control debate provided a lot of good examples of how we present these things….starting with this article at Slate States With Tighter Gun Control Laws Have Fewer Gun Deaths. It came with this graph:
Now my first thought when looking at this graph was two-fold:
- FANTASTIC use of color
- That’s one heck of a correlation
Now because of point #2, I looked closer. I was sort of surprised to see that the correlation was almost a perfect -1….the line went almost straight from (0,50) to (50,0). But that didn’t make much sense….why are both axes using the same set of numbers? That’s when I looked at the labels and realized they were both ranks, not absolute numbers. Now for gun laws, this makes sense. You can’t count number of laws due to variability in the scope of laws, so you have to use some sort of ranking system. The gun control grade (the color) also gives a nice overview of which states are equivalent to each other. Not bad.
For gun deaths on the other hand, this is a little annoying. We actually do have a good metric for that: deaths per 100,000. This would help us maintain the sense of proportion as well. I decided to grab the original data here to see if the curve change when using the absolute numbers. I found those here. This is what I came up with:
Now we see a more gradual slope, and a correlation of probably around -.8 or so (Edited to add: I should be clear that because we are dealing with ordinal data for the ranking, a correlation is not really valid…I was just describing what would visually jump out at you.). We also get a better sense of the range and proportion. I didn’t include the state labels, in large part because I’m not sure if I’m using the same year of data the original group was.1
The really big issue here though, is that this graph with it’s wonderful correlation reflects gun deaths, not gun homicides….and of course the whole reason we are currently having this debate is because of gun homicides. I’m not the only one who noticed this, Eugene Volokh wrote about it at the Washington Post as well. I almost canned this post, but then I realized I didn’t particularly like his graph either. No disrespect to Prof Volokh, it’s really mostly that I don’t understand what the Brady Campaign means when it gives states a negative rating. So I decided to plot both sets of data on the same graph and see what happened. I got the data on just gun homicides here.
That’s a pretty big difference. Now I think there’s some good discussion to have around what accounts for this difference – suicides and accidents – and if that’s something to take in to account when reviewing gun legislation, but Volokh most certainly handles that discussion better than I. I’m just a numbers guy.
1. I also noticed that Slate flipped the order the Law Center to prevent Gun Violence had originally used, so if you look at the source data you will see a difference. The originally rankings had 1 as the strongest gun laws and 50 as the weakest. However, Slate flipped every state rank to reflect this change, so no meaning was lost. I think it made the graph easier to read.↩
graph paper diaries 
I will have more to say on this, and link back. But two things occur to me; gun homicides, gun suicides, and gun accidental deaths are different categories. We might argue that the homicides or suicides would have remained about the same and occurred by other means. We can’t say that about gun accidents, however. It’s hard to argue that a person whose child accidentally shot herself would have accidentally poisoned herself were the gun not present. Or can we, in a very distant way? Gun owners are not identical, and there may be continua or subgroups of people who are careless about their guns who are careless about everything.
Second, what gun laws are in place may be a _result_ of lower homicide rates rather than a _cause_ of them. Europe already had very low levels of internal homicide before WWI and WWII. After the latter, in their drive to embrace peacefulness and have no more killing, they embraced very restrictive gun laws. But the gun laws did not cause the low rates – those were already low. So too with American states. Massachusetts adopted a mandatory one year in prison for committing a crime while in possession of a gun about 20 years ago. But Massachusetts already had a low homicide rate. The law does not appear to have had much additional effect.
LikeLike
Last time I looked into this, I found the “Fatal Injury Statistics” compiled by the CDC to be helpful. Their reports on homicide tend to be close to, if not exactly the same as, the FBI numbers. However, it is very easy to compare homicide/suicide/accidental-death in the CDC statistics.
Of interest: the numbers for accidental-death-by-firearm in the United States are very low. Low enough that the nationwide numbers reported by the CDC rarely rise above 500 total deaths per year.
Secondly, the Brady Campaign used to give “Gun Control Grades” to States…then switched to a “points” system on a scale of 0 to 100.
However, very few States got the exact same number of “points”…and some were credited negative “points”.
LikeLike
With no correlation between gun murders and gun laws….It is very striking to me that 96% of the States (48/50) have 5 or less murders per 100,000, with one that looks like it is 6, and the “outlier” at 8/100,000–I’d love to know which state that is. Another interesting graph would be gun murders vs. murder by other means (maybe a two-color bar graph with a bar for each State and murders per 100,000). Thanks for making the graph.
LikeLike
There is a chart on Wikipedia, which claims to derive from FBI data for 2010.
https://en.wikipedia.org/wiki/Gun_violence_in_the_United_States_by_state
On that chart, the numbers for Louisiana are 7.7-per-100k.
However, that chart also includes the city of DC, which runs something like 16.5-per-100k. (This is not high, even among large cities in the U.S. The cities of Detroit, New Orleans, and Baltimore all have homicide-by-firearm rates above 30-per-100k. Which might give some indication of why Louisiana, Michigan, and Maryland are all in the top-5 States for homicide rates.)
There is a separate Wikipedia chart with all homicides. The rates are moderately higher, but the ordered list looks the same.
https://en.wikipedia.org/wiki/List_of_U.S._states_by_homicide_rate
But that chart also has data from 1996, 2000, 2005, 2010, and 2013. Which gives some indication of how far the rates have fallen since the 1990s.
LikeLike
Louisiana was the outlier. I like the other graph idea….I will have to play around with that for a post this week.
LikeLike
Pingback: Guns and Graphs Part 2 | graph paper diaries
Pingback: 5 Definitions You Need to Remember When Discussing Mass Shootings This Week | graph paper diaries