Texas Style Statistics

This evening, Mark had a Twitter post that made the point that Trump’s statement that there had been 63,000 murders by “illegal aliens” since 9/11 was, in Mark’s words “a LIE; a completely made-up number.”

In response, I decided to look at the White House Fact Sheet issued today entitled “President Donald J. Trump Stands with the Victims of Illegal Alien Crime.”  Since I have a day job, I couldn’t possibly fact-check all of the numerous bullet points.  So I picked out one:  “In Texas alone, more than 250,000 criminal aliens have been arrested and charged with over 600,000 criminal offenses within the last seven years.”  That’s a big number.  So I Googled the phrase “texas percentage charges convicted” and came up with a release from the Texas Department of Public Safety, “Texas Criminal Alien Arrest Data.”

The release states:

Between June 1, 2011 and May 31, 2018, these 171,000 illegal aliens were charged with more than 265,000 criminal offenses which included arrests for 495 homicide charges; 29,526 assault charges; 5,264 burglary charges; 33,637 drug charges; 350 kidnapping charges; 14,794 theft charges; 21,674 obstructing police charges; 1,537 robbery charges; 3,107 sexual assault charges; and 2,673 weapon charges. DPS criminal history records reflect those criminal charges have thus far resulted in over 110,000 convictions including 219 homicide convictions; 12,244 assault convictions; 2,900 burglary convictions; 16,358 drug convictions; 144 kidnapping convictions; 6,642 theft convictions; 10,508 obstructing police convictions; 929 robbery convictions; 1,528 sexual assault convictions; and 1,167 weapon convictions.

Emphasis added.  Look quick: Not 250,000 aliens, but 171,000 aliens.

Immediately following the statement, there’s a nifty chart which the reader can enlarge.  Because of my technical incompetence, I cannot put the file, which is a pdf, inside of this post.  So let me summarize: There are ten specific offenses listed and a catch-all “All Other Offenses.”  The period covered is June 1, 2011 through May 31, 2018.  For each category, the chart sets forth the number of arrests and the number of convictions obtained.

Examining the chart, I noticed something weird:  Of the eleven categories, only two of the smallest categories had a conviction rate greater than 50% of the number of arrests.  And, of course, the chart only set forth the number of arrests and number of convictions, not the number of individuals charged.  Since it is likely that most individuals were charged with numerous crimes, the chart does not support the implication of the Trump “fact sheet” that serious criminality (the ten crime categories listed in the Trump “Fact Sheet”) is a hallmark of illegal aliens in Texas.  The conviction rate for these ten specific serious crimes is only 46.55%.  Stated differently, Texas arrested 171,000 aliens and managed to get only 52,639 convictions for serious crimes.  (The “All Other Offenses” resulted in an additional 57,633 convictions.)

Now, I have limited professional and personal experience with criminal law.  However, I do know that in criminal courts, the deck is stacked against the defendants, particularly minorities.  That being the case, I found the conviction percentages to be suspiciously low.  However, a quick search turned up this nifty piece by the Cato Institute, “Criminal Conviction Rates in Texas 2016.”   That study found that “The native-born criminal conviction rate was . . . 2.4 times as high as the criminal conviction rate for illegal immigrants in [2016] and 7.2 times as high as that of legal immigrants.”  The Cato study also found that:

The 2016 criminal conviction rates in Texas are similar to that of 2015 with one major exception: The illegal immigrant homicide conviction rate is far lower. There were 31 convictions against illegal immigrants for homicide in Texas in 2016 but 51 in 2015. Illegal and legal immigrants had lower homicide, sexual assault, larceny, and overall criminal conviction rates relative to native-born Americans in 2016.

(Note to self: Favorably citing a Cato Institute study is yet another sin that I’ll have to pray forgiveness for on Yom Kippur.)

I managed to do the basic research to uncover the misleading use of statistics by the Texas Department of Public Safety in about five minutes.  The Trump staff, if it were so inclined, could have done the same and made an honest presentation.  Of course, no one in the Trump White House is so inclined.

6 thoughts on “Texas Style Statistics”

  1. Your image. I took a copy from the pdf and pasted into a text document in LibreOffice. Then I right-clicked and saved it again. This time the software gave me the PNG option for file type, which is an image not a document format. I saved it to my usual file for RBC images, and then uploaded that. There seems to a lot of luck in getting hold of images. I lost some resolution in the process, readers should click the link for the best view.

    1. PS: It's a terrible chart. The creator is trying to emphasize the conviction ratio at the expense of proportion (the 350 arrests for kidnapping are given the same visual weight as the 29,526 for assault). But the chart doesn't even manage that, as the convictions and arrests are summed and then normalised to 100. The ratio is nowhere visible as a number. What I would do is (1) rank the offences by frequency (2) construct a double bar chart of arrests and convictions for each offence, with the absolute numbers as labels to the bars (3) under the bar pairs, give the conviction ratio as a percentage. Only the last step is tricky with standard spreadsheet chart software, you would have to edit the chart in a graphics app.

  2. Do we know what the conviction rate, as a percentage of arrests, is for other residents of TX? I tried to find it and failed.

    I'd also like to know what "other offenses" are, since they are about 57% of all the arrests here, as well as the rate of arrest for these offenses among other residents.

    It does not seem implausible that convictions/arrests is low – if it is – because of aggressive arrests of some (cough, cough) segments of the population.

    1. I have the same thought as you: the low conviction to arrest ratio suggests there are a lot of unjustified arrests of illegal aliens.

      Some statistics can be found on the following page: https://www.dps.texas.gov/administration/crime_re
      Click on “Criminal History Arrest and Conviction Statistics” and a set of four spreadsheets should be shown.
      Adding everything up, if my calculations are correct, there were 3,178,681 arrests and 2,116,916 convictions during 2013 through 2016, for a conviction/arrest ratio of 66.60%. This doesn't cover the entire period that the data on illegal aliens covers, but the difference in conviction/arrest ratio seems too large to be explained by the differing time periods.

Comments are closed.