Thursday, January 10, 2019

Immigrant crime: the fallacy of composition and the Usual Suspects

Neighbor (after reading a letter to the editor): "Aren't you afraid Russell Pearce's henchmen will come to get you."
Gilligan:  "No ... they should be afraid of me [and if not me ...  my uncle, Louie, The Greek].  The only person I'm afraid of is Keyser Soze."

"Do undocumented migrants commit more crime proportionally than native-born Americans? No."  
-- Ed Montini, AZ Republic, 1/10/19

Comrade Professor Montini, King of the Nibelung:

What's your source for this observation?  The Cato Institute, i.e., Landgrave and Nowrasteh?  Their study focused on incarceration rates, a dubious proxy for crime.

Landgrave and Nowrasteh found in their study of those they called "illegal immigrants" that "illegal immigrants are less likely to be incarceratedthan native-born Americans."  Lott, on the other hand, using a 100 percent sample of Arizona incarcerations found that illegal immigrants were over-represented,  criticizing those (e.g., Landgrave and Nowrasteh) who relied on the American Community Survey sample (ACS).   Lott study raises the pertinent issue of African-American incarceration rates and African-American recidivism rates skewing overall incarceration rates for the native-born.   Lott, to the extent he did, might be guilty of a fallacy of composition where he suggests Arizona data extrapolates to the entire US.

Disaggregating  Landgrave and Nowrasteh's findings shows that they are not inconsistent with Lott's, however,  since Arizona has a relatively small African-American population and the likelihood that Arizona's African-American population may not suffer as much from institutions and institutional status (in an Ayresian sense) as African Americans do in other parts of the US.   According to Landgrave and Nowrateh,  Hispanic Americans of any race, illegal immigrants and not, have higher incarceration rates than White and Asian Americans.  Lott, for his part, attributes differences in native-born Hispanics and Whites to the well-known association of youth with crime and allows that but for that demographic native-born White and Hispanic crime rates are about the same.

Why are African American incarceration rates so high?  1) The institutional inheritance, old and new.   2)  The Usual Suspects:  in criminal investigations the police first look for known culprits: mugshots, fingerprints, DNA.  

But, facts don't really matter here, Ed.   This wall business is just a political pissing contest.   Even if the wall accomplished nothing,  the lousy $5 Billion isn't worth shutting down the TSA and Coast Guard ... a rational opposition would trade the wall for something they believe is worthwhile.

Suggesting that facts matter is dissembling on your part, Ed.   You either don't understand the study you cite or you don't care ... probably both.   You just want to be part of the pissing contest.   You get up in the morning and without even changing out of your pajamas, you start throwing BS at the wall.   You're an embarrassment to all the people in your professor who work hard at putting together stories that are interesting and bear some semblance to reality.  You, Ed, are guilty not only of the fallacy of composition, but also, the fallacy of division.

Slainte,
Gilligan
Data Scientist
Linebacker Strike Group, Freedom Train Strike Group, Pocket Money Strike Group, North SAR
Cook/Deckhand, MV Mugwump


Wednesday, January 9, 2019

The Wall and the Spitfire

"But data from his own administration repeatedly show that a rising number of smuggled drugs, especially more potent ones like heroin and cocaine, are increasingly seized at legal ports of entry, which are not impacted by a wall."  -- AZ Republic, 1/8/19

We don't need a wall because we catch the most smuggled drugs at ports of entry?   This brings to mind how the British decided where to armor their Spitfire fighter planes during WWII.  Fighters need to be light to be fast and maneuverable.   When you want as little extra weight as possible, where do you add the armor.   The British put their crack operations researchers to work and gathered data on bullet holes.   When they were done and presented their data to the Royal Air Force, a young pilot shouted:  "Eureka!   Put more armor where the most bullet holes are."   The old professor leading the operations research team quietly shook his head and then spoke:  "No, young man, that is not where to put the armor.  The planes with the bullet holes all came home safely.   We need to put the armor where our data shows there aren't any bullet holes.  Where our data shows no bullet holes is where the fighters were shot that didn't come back."   Because we are catching smugglers at one place, doesn't tell us that there isn't smuggling where we're not trying hard to intercept it and, therefore, have no data, ie, bullet holes, about the incidence of smuggling in those places. 

Thank God, they didn't let "journalists" design the Spitfires.

Wednesday, January 2, 2019

Times and AZ Republic cover up their Irish problem and other mistakes

In his spirited column about the New York Times and Alice Walker's anti-Semitism, Richard Cohen complains that the Times refuses to admit a mistake, interviewing Walker and not being shocked by her bigotry.  He just discovered that?  The Times fired their mistake department back in 2017 and even when they had a Public Editor some "mistakes" were off limits, eg, "OP-ED mistakes are not within the purview of the Public Editor."

Some years ago OP-ED columnist Bob Herbert wrote an egregious, error-filled column that blamed Irish immigrants who [allegedly] refused to fight to free poor blacks for the New York City draft riots.   Herbert neglected the part about the commanding generals at the pivotal Civil War Battle of Gettysburg being  Irish Americans,  George Gordon Meade and John Reynolds.   He also left out the part about New York City's Irish Brigade committing mass suicide on the eve of Emancipation while storming the stonewall at Fredericksburg,  12/13/1862.   In fact, the people who incited the riots were Americans with perfectly American pedigrees.  The one immigrant who was part of the upper-class cabal opposing Lincoln wasn't Irish.   The Times has never apologized for Mr. Herbert's nativist column or even published a letter of complaint.

Likewise,  the Times has never addressed it's problematic interview with James McPherson where it failed to challenge McPherson on the nativism and anti-Catholicism in his Battle Cry of Freedom and For Cause and Comrades,  where McPherson claimed immigrants were skulkers who gained "inglorious visibility" through late war enlistments.  How did the Times and Professor McPherson miss the part where immigrants won a quarter of the Civil War Medals of Honor and that these awards are highly correlated with enlistments, indicating a high immigrant presence in the Union army and navy throughout the war -- immigrants from literally all over the world.

The Times isn't alone in refusing to acknowledge mistakes.   The Arizona Republic has the same flawed blind spot.   It too endorsed Professor McPherson and his Battle Cry of Freedom and casually published Clarence Page's nativist column that blamed the Irish and their gangs for slavery not going away by itself.  Apparently Mr. Page believes that without immigration the South would have never seceded and there wouldn't have been a Civil War where the Meades and Reynoldses won the biggest battle.   No apologies from the Arizona Republic or Mr. Page.