In a recent message to his Facebook followers, Arizona Congressman Paul Gosar called DACA a “Trojan Horse” and posted the following meme:
The statistic, he said, was from a study by the Crime Prevention Research Center. Fueled by this study, scads of right-wing websites jumped on the DACA-bashing bandwagon.
Breitbart wrote, “DACA-aged illegals commit crimes at twice the rate of young Americans, says a comprehensive summary of crimes and convictions in Arizona during the past 32 years.”
What is this “comprehensive summary”? It’s a bit of a Trojan Horse itself. At first glance, it’s a monument of a study, covering more than 300,000 crimes committed by citizens and non-citizens in Arizona between 1985 and 2017. But when you open it up, out swarms an army of manipulations and misleading figures designed to spread fear and hate throughout the land.
The study was conducted by John R. Lott, Jr., president and founder of the Crime Prevention Research Center. Lott runs the CPRC out of his home in Pennsylvania. Rocker and NRA board member Ted Nugent is on the CPRC board of directors. The organization mainly conducts studies attempting to prove that more gun ownership reduces crime. Lott’s studies have never been published in peer-reviewed journals, and they have been criticized numerous times by researchers for being “faulty,” “erroneous” and “garbage.”
That doesn’t necessarily mean this study, “Undocumented Immigrants, U.S. Citizens, and Convicted Criminals in Arizona,” is faulty, erroneous garbage. But it should make one suspicious.
So should the fact that it’s no accident a study attempting to disparage DACA is being released now. Congress is currently debating a solution for DREAMers after Trump pulled the plug on the Obama-era program that temporarily protects young immigrants from deportation and gives them a renewable work permit. Anti-immigrant hardliners are motivated to attack the public’s widespread support of DACA.
Note, first of all, that this study is not about DACA recipients. Young immigrants cannot qualify for DACA if they have committed crimes, so crime rates among DACA recipients are extremely low. Lott’s study talks about DACA-age immigrants, deliberately linking DREAMers with convicted criminals based merely on age. For most of the years covered by Lott’s study, DACA didn’t even exist yet.
Furthermore, the study focuses on numerous categories of drug-related crimes. And Lott admitted to the Phoenix New Times that an unknown number of those drug cases attributed to “undocumented immigrants” may actually have been committed by criminals who crossed the border merely to make a drop, then got caught and were referred to state court.
So let’s say, for example, a 25-year-old Mexican drug cartel employee drives across the border with a bunch of marijuana and a small group of immigrants. A chase ensues, resulting in a crash in which two immigrants are killed. Border Patrol arrests the man, and he is charged with a variety of crimes including drug trafficking, two counts of manslaughter, smuggling, and reckless driving. Each of these charges is logged as separate incidences in Lott’s study. Thus, a drug cartel member who intended to stay in the U.S. for only a few hours is now categorized as a “DACA-age illegal.” And his crimes are used to disparage, say, an ASU engineering student and part-time worker who has been living in the U.S. since her parents brought her here at age 3.
In addition, it’s important to remember that just because a group is arrested at higher rates doesn’t mean they’re committing more crimes. For example, white people smoke marijuana at a slightly higher estimated rate than black people. But black people are much more likely to get arrested for it.
Also, anytime you see a percentage increase in a study or an article, it should raise red flags. (That goes for statistics you want to believe as much as for statistics you don’t want to believe.)
If you see a headline that says “Small town’s murder rate increases 200 percent” you might think this unfortunate town has an epidemic on its hands. But if one person was murdered in 2016 and three people were murdered in 2017, that’s a 200 percent increase.
So readers should be skeptical when Lott says in his study: “Undocumented immigrants are 163 percent more likely to be convicted of 1st degree murder than are U.S. citizens, 168 percent more likely to be convicted of 2nd degree murder, and 189.6 percent more likely to be convicted of manslaughter.”
It might seem like the difference between 4.8 percent of criminal convictions and 12.6 percent of criminal convictions is about 8 percent. But it isn’t. It’s 163 percent.
If you go through the study, you’ll see that U.S. citizens commit a higher percentage of crimes in every single category except smuggling. In most cases, U.S. citizens commit a far higher percentage of each crime. For example, U.S. citizens committed 86.5 percent of 1st degree murders, while undocumented individuals committed 12.6 percent. (Again, some of those undocumented individuals may not have been actual immigrants.)
Now, it’s not surprising that U.S. citizens commit a far higher percentage of crimes, because there are so many more U.S. citizens than there are undocumented individuals.
So what Lott did was compare the percentage of crimes committed by undocumented individuals with their percentage in the total population. That’s a common practice in statistics, but it can be misleading and manipulated.
According to the Pew Research Center figures Lott used, undocumented individuals make up 4.8 percent of Arizona’s population. So you might expect that they make up 4.8 percent of criminal convictions. But according to Lott’s crime statistics from the Arizona Department of Corrections, undocumented individuals made up 12.6 percent of criminal convictions.
If you’re not a statistician, it might seem like the difference between 4.8 percent and 12.6 percent is about 8 percent. But it isn’t. It’s 163 percent.*
Let’s look at 1st degree murder. According to Lott’s study, between 1985 and 2017 there were 1,790 murders. 12.6 percent of those were committed by undocumented individuals. That’s 225 murders. If undocumented individuals had committed 4.8 percent of murders, it would have been 85 murders. That’s a difference of 140 murders. Over 33 years, it’s an average of four murders a year.
So when Lott says “Undocumented immigrants are 163 percent more likely to be convicted of 1st degree murder than are U.S. citizens,” what that translates to is: “Undocumented individuals, some of whom are likely drug cartel members shooting at each other and aren’t even part of the resident population, commit four more murders per year than you might expect based on undocumented people’s share of the total resident population.”
But that doesn’t make a very clickbait-y Breitbart headline, does it?
(* That’s because a 100 percent increase of 4.8 percent would be another 4.8 percent, bringing the total to 9.6, and a further 63 percent of the original 4.8 is 3, bringing the total to 12.6.)