[Pharmakon]

Can you cite those numbers? I checked one and even being generous you inflated it by quite a bit.

https://www.bop.gov/about/statistics/statistics_inmate_citiz...

Thst indicates that if every single non U.S. citizen in federal custody were there illegally (unlikely) they’d represent 20% and not 25%. Further clarification from the Cato Institute: https://www.cato.org/blog/another-confusing-federal-report-i...

The evidence that legal and illegal immigrants are less likely to be incarcerated, convicted, or even arrested for crimes is so overwhelming that even immigration restrictionists like Mark Krikorian at the Center for Immigration Studies admit that, “A lot of data does suggest immigrants are less likely to be involved in crime.”

The Cato Institute being famously not left wing I’d add, founded by the Koch’s.

[escape_goat]

He is probably using reports from the Centre For Immigration Studies. I believe they can be trusted to use real data, but can't really be trusted to represent it in proper context. I'm not sure if they cherry-pick cynically or just out of confirmation bias.

1. 26% of Federal Prisoners are Aliens [https://cis.org/Huennekens/DOJ-26-Federal-Prisoners-Are-Alie...]

2. Welfare Use by Immigrant and Native Households [https://cis.org/Report/Cost-Welfare-Use-Immigrant-and-Native...]

[dfilppi]

Confirmation bias works both ways.

[escape_goat]

Shrugging and saying "confirmation bias works both ways" is neither factual nor a valid rebuttal. Confirmation bias is a value-neutral phenomenon. It is neither innate, unavoidable, nor inherent to sociopolitical analysis. It does not "work both ways." It works only one way.

There are numerous sources of policy opinion all across the political spectrum that exhibit confirmation bias. There are also those who attempt to limit their own bias and do scholarly policy work in good faith.

When a person or publishing entity exhibits a tendency to systematical devalue evidence that contradicts their assumptions, that is confirmation bias.