[content warning: discussion of violent crime, including sexual assault]
I plan to write an article on misuse of statistics by online news organizations, but looking back through my archives most of the examples I’ve got are from a couple of liberal sites that aren’t the worst offenders so much as the only ones I can even bear to read. I’m worried that some of my readers have gotten the impression that liberal sites are the only ones that routinely misuse statistics, which would be grossly false. So before I write the article, I thought I’d give one example of how a lot of conservative sites have statistics that are so bad they’re not even fun to dissect.
I chose Breitbart’s “Rape Deniers: 9 Facts About Illegal Alien Crime The Media Covers Up” because it sounded promising – talking about rape and calling people who disagree with you “deniers” are two pretty reliable red flags that an article will be terrible. It’s a series of 9 facts meant to show that illegal immigrants to the US are involved in a lot of crime, especially sexual crime.
I’m skipping Fact 1, which is just a methodological point I don’t dispute, Fact 2, which just says some native-born Americans are unemployed, and Fact 3, which says that a lot of our heroin comes from Latin America; I don’t really disagree with any of these. I’m also skipping Fact 7 because it’s a repeat of Fact 6 and the same points apply. That leaves Facts 4, 5, 6, 8, and 9.
4. Because local and state prisons don’t track legal status, we don’t know how many illegals are in those prisons. As my colleague Ben Shapiro points out, the lack of this number is being used dishonestly by the media against Trump…Trump’s repeated statements about immigrants and crime underscore a common public perception that crime is correlated with immigration, especially illegal immigration. But that is a misperception; no solid data support it, and the data that do exist negate it. Trump can defend himself all he wants, but the facts just are not there. Except the facts are there. The Feds do track legal status, and the numbers are startling. Of 78,022 primary offense cases in fiscal year 2013, 38.6 percent were illegal immigrant offenders. The majority of their cases (76 percent) were immigration related. Of total primary offenses, 17.6 percent of drug trafficking offenses and 3.8 percent of sex abuse were illegal immigrants. Of 22,878 drug crime cases, 17.2 percent were illegal immigrants.
The first part of this is a really weird complaint. They’re saying that 38.6% of federal prisoners are illegal immigrants, which is true and indeed very high. Then they’re admitting that 76% of their cases are immigration related. That is, the Feds are imprisoning them because they immigrated illegally. I think that most people would be willing to concede that illegal immigrants are more likely to have immigrated illegally than other populations.
The second part, the part about federal drug trafficking, is complicated; it’s different from “drug having”, “drug dealing”, and even “drug trafficking” as a broader non-federal category. To get a federal drug trafficking arrest, you have to move really large quantities of drugs “across state or national borders”, preferably in a “High Intensity Drug Trafficking Area”. That is, crimes that would give you state charges in a normal place become federal charges in one of these areas. Where are they? The entire US border with Mexico is a gigantic High Intensity Drug Trafficking Area (see this map). So a high rate of illegal immigrants among federal drug trafficking prisoners just means that they’re more likely to be involving in transporting drugs across the US-Mexico border than, say, a lumberjack in Wisconsin is. I am prepared to believe this.
Finally, illegal immigrants do commit 3.8% of federal sexual abuse cases. I give Breitbart credit for finally getting a number that is entirely correct and not biased at all. Unfortunately for them, illegal immigrants are 3.8% of the US population.
5. According to the Justice Department…There are 94 federal court districts in this country and the five located near the southern border see a large portion of criminal cases, according to the Justice Department’s annual report on criminal prosecutions. The five federal districts also have the biggest number of defendants actually convicted of federal crimes. Of the 61,529 criminal cases initiated by federal prosecutors last fiscal year, more than 40%—or 24,746—were filed in court districts neighboring the Mexican border….Nearly 22% (13,383) were drug related, 19.7% (12,123) were violent crimes and 10.2% (6,300) involved white-collar offenses that include a full range of frauds committed by business and government professionals. Read those stats closely because the media will lie and claim the crimes involve border enforcement. As you can see, over 40% involve drugs and violence.
So here’s some interesting math. “More than 40% – or 24,746 – were filed in court districts neighboring the Mexican border.” Ellipsis. “13,383 were drug-related, 12,123 were violent crimes, and 6,300 were white-collar fraud.” Wait a second. Those three numbers add up to 31,806, more than the total number of cases. Apparently we can’t trust illegal immigrants to obey any laws, including the laws of mathematics.
Breitbart’s reference goes here and their reference is here. When I look it up, these turn out to be the numbers across the entire US, not the numbers for southern border regions. This makes sense – do we really think illegal immigrants commit 6,300 cases of white-collar fraud per year in their dynamic illegal immigrant megacorporations?
But this is really embarrassing for Breitbart’s case. Their whole point is that the disproportionate number of crimes committed in immigrant-heavy areas are not just immigration offenses, but in fact representative of the country’s crime load as a whole. But that’s only because they’re accidentally looking at the country’s crime load as a whole instead of at crimes committed in immigrant-heavy areas! I can’t actually find the immigrant-heavy-area data, but I’d be willing to bet that the disproportionate number of federal crimes along the border do in fact involve border enforcement, exactly the argument they claim to debunk.
6. There are more than 2,000 sex offenders deported by ICE every year in Texas alone…of the 862 alien sex offenders deported by the Texas-based offices, about 27 percent were convicted of sex offenses against children?
Quick, how many illegal aliens in Texas? If it’s 10 million, then their sexual offense rate is far lower than that of any other population. If it’s 10,000, their sexual assault rate is far higher than that of any other population. Not only does Breitbart not give these numbers, but they don’t even seem to understand that they should give them, or expect their readers to care.
I can’t actually figure this number out because it depends on knowing what percent of immigrant sex offenders are deported each year. Consider: there are 86000 sex offenders in Texas. About 6% of Texans are illegal immigrants, so by chance we should expect about 5000 sex offenders total. Given that more than 2000 are deported per year, and that this has been happening for more than three years, that sounds like there are (or were) more than 5000, unless some deportees came back, which we know some do. But I don’t know that the sex offender registry is measuring the same kind of sex-offenderness as the illegal immigration numbers, so I can’t be sure of this. Also, the Texas statistics for immigrants include Oklahoma and possible other areas, so it’s hard to directly compare.
According to this page, there are 5017 arrests for illegal aliens for sexual assault over 4.5 years, so about 1100/year. That should be the sort of number we can work with. But it’s well above the total number of Hispanics arrested for sexual assault given here, immigrant and native-born alike, which doesn’t make sense. So I don’t know what to do.
The only good source I can find for percent violent crime by illegal immigrants in Texas is this one, which says that they commit 7.5% of murders. But they’re 6% of the population, so that’s pretty much what we’d expect. This accords with the numbers mentioned above, where in federal prisons the percent illegal immigrants serving time for sexual assault was proportionate to their percent of the population.
I can’t find good numbers here. A very rough inference from one source of Texas sex offender numbers would suggest that the number of sexual offense deportations is unusually high, but inferring based on the murder and sexual abuse numbers suggests that sexual offenses are about average. I’m not sure which method is more correct – but in any case, whatever the truth is there’s no way Breitbart’s numbers could be expected to get anyone any closer to it.
8. In 2013 the Obama administration released 36,007 criminal immigrants who had nearly 88,000 convictions between them. Those convictions included 193 homicide convictions, 426 sexual assault convictions, 303 kidnapping convictions, and 1,075 aggravated assault convictions In January, the DHS admitted to Sen. Grassley that 1,000 of the 36,007 released had gone on to commit more crime including: terroristic threats, lewd acts with a minor, various types of assault, DUI, robbery, hit-and-run, gang activity, rape, and child cruelty.
Politifact rates this half true. All immigrants involved served full prison sentences appropriate for their crime; the concern isn’t that they didn’t serve their time but that after their time was up they didn’t get deported. The DHS says that their policy is that if they haven’t finished a deportation case by the time a criminal gets out of prison, the criminal may be released until the deportation case is finished, mostly because the prison system won’t keep them and the DHS doesn’t have enough immigration-related prisons of its own. These people are subjected to the usual monitoring and may be rearrested and deported after their deportation case comes through. There is some reason for concern in that about 3/4s of these people manage to lose themselves before the deportation case is complete, but the DHS reasonably says that if people want this to stop happening people should give them more funding.
And once again, without more information we can’t tell whether any of this involves a higher crime rate than any other population.
9. ICE is finding and removing more criminal aliens each year. The number ordered removed has gone up from 7,000 in 2007 to 79,000 in 2010. These criminals are not being stopped at the border. These criminals are being deported after making it across the border and committing tens of thousands of crimes.
Sheesh. You get angry when we don’t deport immigrants. You get angry when we do deport immigrants? Make up your minds! This is literally just complaining that we’re getting better at solving the problem you complained about above!
There are between ten and twenty million illegal immigrants in the United States – about equal to the number of New Yorkers. If somebody wanted to expel New York from the country, they could point out that New Yorkers commit 616 homicides, 2,534 rapes, and 45,206 cases of aggravated assaults per year. Or that we need seventy-one prisons just to contain all the New Yorker criminals in our justice system, and we don’t have nearly enough funding to run all of them effectively, such that literally thousands of New Yorker criminals, including murderers and rapists, are released into the general population every year. You could report that [number] of New Yorkers commit violent sexual assaults on children each year – I don’t know what the number is, but I guarantee you that it is a number, that it has a certain number of digits, and that it is worse than zero.
Breitbart consistently fails to give numbers that would mean anything or inform anybody, and when it either uses numbers that are loaded in its favor or ones that don’t mean what it thinks they mean. The actual numbers on the question of interest don’t seem particularly outrageous.
This isn’t to say that there can’t be legitimate concerns about illegal immigration. In fact, that’s my whole point and something that I wish conservatives better understood: we need to practice Gettier politics. There’s a theory on the Right that since the media has created a giant edifice of lies to justify liberalism, liberalism must be false. But other parts of the media have created a giant edifice of lies to justify conservativism. Instead of assuming our opponents are necessarily gullible morons who believe the giant edifice of lies on their side, we should kind of awkwardly go “Oh, there’s a giant edifice of lies on your side too? Yeah, I know that feeling,” and listen to what they have to say.
A major obstacle to changing one's mind is the fallacy that if the other side is right, it must be decent and intellectually honest.
— Steven Kaas (@stevenkaas) August 21, 2010