Hacker News Reader

Comment by legalcorrection | Hacker News Reader

If that were the case, we would then expect liberal DAs in liberal states to have an even higher murder rate. And I assure you, none of the policies you mentioned are being implemented in the Deep South states of SC, AL or MS.

> none of the policies you mentioned are being implemented in the Deep South states of SC, AL or MS.

These states may have started with extremely high murder rates. It's the growth in rates that would show whether OP is right or not, and those states with rapidly growing murders are not in the deep south.

(That said, none of the red states with high growth strike me as particularly likely to have liberal local DAs either, so I also question their argument.)

> Red states tend to have a much higher population of black residents, a group that is well established to have much higher crime rates. Black people are 13% of the national population but commit 52% of the murders.

Expanding this to causation - that black people cause more crime - is not only a complete misuse of statistics - nothing there shows correlation, much less causation - it is an old racist saw. For example, a society which relegates people based on their skin color to poverty, lack of access to jobs, education, and healthcare, and continual discrimination, might be the causal agent. It could be that people in poverty, with poor educations, and difficult employment, are correlated with crime (but we'd need to look that up).

> Many of those red states have very liberal DAs who are refusing to prosecute certain crimes, asking for extremely low bail, and pursuing very short prison sentences, while also aggressively prosecuting alleged police misconduct.

Those DA's seem less prevelant in red states - is there some correlation there? And do you have evidence that these things increase crime rates? Most of the research says otherwise, the opposite practices had little impact for many years, and the people most affected by crime vote for these DAs.

The correlation with race holds even when you control for poverty and other socioeconomic factors (such as urban/rural). But of course, you can't control for the unique historical experience of black Americans. So this remains an open question.

So what are you saying the causal relationship is, beyond the racist assertion that skin color causes crime?

>Red states tend to have a much higher population of black residents

It sounds like you're taking the article to mean "voting Republican is causal to murder", and are rebutting with "being black is causal to murder" (but are using a corollary fact to imply it). However, the first statement isn't being made by the article, and your implied rebuttal is neither verifiable nor does it even rebut the first statement, were the article making it. It's also unnecessarily provocative (unnecessary due to its irrelevance as described above). Of course this is all assumption you've forced me to make. By itself, the sentence you wrote doesn't say anything directly - you haven't connected it to the article explicitly.

Your second paragraph has the same problem, but the implication is more obvious: You believe the properties you've listed cause high murder rates. That doesn't seem to be true, though - can you either elaborate subjectively, or provide some other writeup that points to that conclusion?

Mask off time. I wonder how @dang will handle egregious racism disguised as "data".

What's the difference between egregious racism disguised as "data" and data relating to race that contradicts liberal priors?

The difference is contextual, as with anything. The parent obviously disagrees that the data provided rebuts the article's content (I too disagree, and think it mostly only serves to spur hatred in this particular context, the same way the title of the article does).

13/52 is a known hate symbol: https://www.adl.org/education/references/hate-symbols/1352-1...

This is weak. Yes, these numbers can be (and usually are) used as emotion-targeting propaganda. However, assuming they are true in a reasonably non-cherry-picked context, the prevalence of dishonest usage then does not disqualify them from being used in a useful context, free from hatred. Your post implies otherwise, though (regardless of whether they are or are not being used that way in the particular case of the OP).

It’s literally a crime statistic that’s true.

Lots of propaganda and hate-bait is true. Context is what matters in those cases. I agree that the parent is making an incorrect implication though, but not for the reason you give.

First, that's irrelevant. Second, all we have is your claim. We do have a citation that it's hate speech.

Is there literally a single thing which you couldn’t cite as hate speech in 2022?

While that is a common response, it doesn't address the problem or make any specfic claim. Are you saying there is no hate speech? We shouldn't do anything about hate speech? Make a claim here.

The article is just lazy journalism. There are a lot of possible alternatives variables to run a regression on that could have stronger explanatory power and they just chose the first that fit the narrative they wanted to present. Southern states and the legacy of racism looks like a viable explanation for the higher murder rates in 'Red' states that should have been explored as part of the article.

Can data actually be racist? Or just the way we use it (information we form from it)?

I’d argue that pointing at a user and calling them racist is quite egregious.

I would love to see the supporting data you used to reach that conclusion.

They are likely quoting DOJ arrest statistics [0]. Lots of interesting stuff there, and the overall conclusion one can draw from them is quite different than what the GP asserts.

0 - https://www.ojjdp.gov/ojstatbb/crime/ucr.asp?table_in=2

We have decades of evidence that prison sentences being long or short has no effect (or nearly so) on deterring crime. Likely the murder rate differences in these red vs blue areas has more to do with the economic well being of these areas that police sentiment or how tough on crime the police and DA are. Someone here should be able to whip up a script to correlate that in like an hour.

Could you please share some examples of a very liberal DA in a very conservative state outside of major metropolitan areas? No offense but your response sounds very brainwashed, where the problems in a conservative area are obviously caused by the dirty local liberals mucking everything up.

> No offense but

then proceeds to tell OP that they're brainwashed. You were deliberately trying to offend, let's not mince words ok?

Because they are.

Sounds like you are trying to justify the results to fit your existing narrative. You immediately assume there is a fault in the data instead of analyzing your own baseline assumptions. The idea that Alabama, Mississippi, Louisiana, etc. have higher murder rates because they are "very liberal" doesn't pass the smell test.

Your comment also seems to assume that the states which vote blue somehow don't have locally elected DA's? That doesn't explain why Missouri (with much less urbanization) has a higher murder than Illinois or Michigan (with "out--of-control" cities like Chicago and Detroit.

Reply on news.ycombinator.com