If you don’t put your work out there for review, it’s not science.
If you claim something that isn’t science is, in fact, science, that’s pseudoscience.
That’s exactly what this is.
I’ve heard it a few times before that the premise of all dystopian fiction is “what if the government got so powerful it could do what it’s already doing, but to white people.”
19 Likes
Betcha there won’t be a whole hell of a lot of CEO’s, Wall Street traders, violent gay bashers, lobbyists or other white collar “pre-criminals” on this list.
3 Likes
This is a horrible idea,
It is morally wrong.
It will be used to discredit the police who already have low credibility to its target groups.
It will be used to accuse police of being racist (and many people are even if subconsciously).
It ignores the root causes in favor of the symptoms. (Blaming the victims of poverty and poor education)
With all that said.
That doesn’t mean its not accurate and not real science. For example personality tests are great predictors of behavior, that is why they discriminate against oppressed groups. That is also why they are basically illegal for hiring purposes (the outcome is illegal).
Hate it for all the right reasons, not because the methodology is not being revealed.
Or became president of the USA via the Supreme Court…
1 Like
Differential Association is not pseudoscience; it’s a legitimate and validate theory of criminal behavior. This an automated version of predicting those associations. Deal with it.
You can perform a randomised control trial, perhaps. Randomise whether or not you respond to the system’s predictions or not. Then look at the outcome. Look at whether you actually prevented crime in the intervention cases. Look at what proportion of false positives you got in the non-interventional cases. Etc. You really don’t need to know specifically how the system works to determine whether it’s working or not, if you go about it honestly.
Keeping algorithms secret is always going to controversial, but there’s a reasonably good precedent for it. The point would be that if the algorithm was publicised, an adversary would be able to respond to it. For example, Google will not make public the details of pagerank for fear it would make SEO’s jobs easier, and I don’t think BB would jump on google’s work as being unscientific and ineffective. There’s a reasonable case that if the details of this crimerank is made clear, criminals will learn how to tailor their profiles to look less like criminals.
Why not both?
After all, if they won’t reveal the methodology there stands a good chance that it is among the weak links of their chain. They don’t even have to reveal any pre-convicts identity to demonstrate how they determine who to foster fear & self-loating within.
2 Likes
Science does not hide its methods. This is engineering. Social engineering. Not science.
3 Likes
I don’t know if comparing it to a trade secret is applicable, but it certainly would be the wrong reason to keep it under wraps in this instance.
2 Likes
“Cops” are a pretty large dataset. It would be statistically unlikely they wouldn’t include those guilty of every possible infraction.
Especially given their disinclination to prosecute their own.
2 Likes
Would you say that the highly secretive inventors of the atomic bomb were doing real science? What about the decoders of the Enigma codes, should they have published what they were doing to be real scientists?
That is not always true. For example the Rorschach test images are well guarded to improve reliability. Even ones shown on TV and in Movies are purposely fakes.
Furthermore secrets of the design of atomic weapons were hidden for years… that doesn’t mean they were not science.
I think the engineering was kept secret, but the science was mostly pre-war, and it was shared with the germans and italians, a few of whom helped on ‘our’ side.
But that was war. Is the Chicago PD at war?
tachin1 writes:
“Why are you more sure that it will be used effectively than you are that it will be misused?
Do you know how it works? if not, then how can you be so sure?
Please tell me.”
…Will it be used properly? Possibly.
There, now no conclusions are drawn; as aluchko concluded the OP:
“Now the fact it’s a scientifically valid approach still doesn’t mean it’s a good idea. The article suggests these people will be targeted for minor police harassment which would be a big problem.”
Spin away…
1 Like
No, the science was kept extremely secret. The most secret thing was how much uranium you needed to produce a critical mass, and the methods to calculate that. After the war, it turned out that the Nazis thought you needed a lot more uranium than you actually did, which meant they never put much of an effort into it.
Whether you believe in the rhetoric of a war on crime or not, the similarity between Chicago PD’s situation and the WWII Manhattan project’s situation is that both are interested in the maximum effectiveness deployment of a technology against an adversary that could adapt to it, not in spreading knowledge and fostering co-operation. In such cases, secrecy will at least temporarily make things more effective. Now, there is an issue of morality, but being amoral does not make it unscientific.
And the tests were purposefully developed among peers, and reviewed by peers.
That one element is kept from the view of anyone except those being medically treated is not analagous to this topic. That’s medicine. Not policing.
I wonder how many Police’s own children are on the list?
1 Like
“My fear is that these programs are creating an environment where police can show up at anyone’s door at any time for any reason.”
Isn’t that true now?
2 Likes
Higher level politically, true, but I don’t think the same level in the business. Mind you, to really succeed in that field, you have to be a kid who joins the CIA…
