US Justice Dept. refuses to correct bad report linking terrorism and immigration

[elix]

https://www.washingtonpost.com/world/national-security/justice-dept-admits-error-but-wont-correct-report-linking-terrorism-to-immigration/2019/01/03/cd29997a-0f69-11e9-831f-3aa2c2be4cbd_story.html

Title rewritten to fit. One of the major justifications for Trump's 2017 travel ban turns out to be full of misleading cherrypicked statistics and the DoJ admits it but won't retract or correct them.

One flaw the Justice Department acknowledged was the report’s assertion that between 2003 and 2009, immigrants were convicted of 69,929 sex offenses, which “in most instances constitutes gender-based violence against women.”
But, Allen said in his letter, “the alleged misrepresented data constitute mere editorial errors which the [law] does not obligate the agencies to withdraw or correct.”
...
Berwick said the errors were “not merely editorial.” The nearly 70,000 offenses spanned a period from 1955 to 2010 — 55 years, not six; the data covered arrests, not convictions; and one arrest could be for multiple offenses, he said, citing the General Accountability Office, which provided the underlying data.
...
Critics also decried the report’s inclusion of eight “illustrative examples” of foreign-born individuals out of a pool of 402 convicted of international terrorism. Six of the eight cases involved people admitted to the country as family members of legal residents or U.S. citizens, which Protect Democracy contended was an effort to portray “chain migration” as a threat.
“On reconsideration, the department acknowledges that a focus on eight seemingly similar ‘illustrative examples’ from a list of more than 400 convictions could cause some readers of the report to question its objectivity,” Allen wrote.

[StrawberryClock]

Predictably...

[Borealis]

If they used that report to argue before the Supreme Court, it's just like when FDR withheld information to get the court to uphold the internment camps.

[ASparkle]

Just to see what the simplest possible number correction would be: since the 69,929 figure comes from 10 times as long a period we can take a tenth of that at 6,993. Then since those are arrests and not convictions we can take the average conviction rate for sex crimes which is 11%. So, assuming those averages even can be applied like that which frankly they shouldn't and it would be much better to just use the actual data from the time frame given with the proper criteria, we get a total of 770, which is 1% of what this "minor editorial error" reports.

I think the last time I saw a report this horribly made it was written by Emil Kirkegaard.