Welcome to the Utopia Forums! Register a new account
The current time is Fri May 24 18:13:12 2019

Utopia Talk / Politics / For Sam: ethnicity and knife crime
Seb
Member
Sun Apr 28 03:47:30
Sam, you've repeatedly said UK knife crime is due to ethnicity.

http://wha...me%5FEvidence%5FBriefing%2Epdf

Seb
Member
Sun Apr 28 03:48:06
"Ethnicity – recent analysis of data collected in the UK indicates that there is no statistically significant relationship between ethnicity and weapon carrying"
Forwyn
Member
Sun Apr 28 04:25:46
"weapon carrying"

Imagine being such a cuck that you conflate carrying a pocket knife with violent crime
Seb
Member
Sun Apr 28 05:17:46
A pocket knife has a blade less than 4 inches and so isn't a weapon under the relevant act.

I used the term knife crime, which includes carrying a weapon.

Seb
Member
Sun Apr 28 05:18:48
Quite dishonest post there Forwyn: two lies in one sentence.
Rugian
Member
Sun Apr 28 07:39:14
Given that this is the UK that we're talking about, observer bias cannot be discounted as a possibility in the conclusion stated here.
Dukhat
Member
Sun Apr 28 07:42:47
Rugian's statements are always deeply drenched in irony when he attempts to say something intelligent instead of just trolling.
Rugian
Member
Sun Apr 28 07:48:23
Seb,

Forwyn did have a point here. Sam claims that ethnicity and violent crime tend to have a strong correlation; your rebuttal to that was a statistic on "weapon carrying."
Rugian
Member
Sun Apr 28 07:52:39
Dukhat,

Except that first post of mine was being semi-trollish. Thank you for this latest display of Dunning-Kruger in action.
Seb
Member
Sun Apr 28 08:56:02
Rugian:

If that is Sam's intent, then he should stop presenting statistics on knife crimes which are dominated by carrying offences; and should instead produce statistics on arrests and convictions for violent offences where the weapon was a knife.

I cannot be to blame if in Sam's desperation to substantiate his opinion that he has confused a category that is dominated by a non-violent crime.

Instead, you should focus your criticism at Sam for another example of his poor handling of evidence and statistics.
Rugian
Member
Sun Apr 28 09:30:21
Seb,

I haven't paid attention to Sam's arguments here, so I'll bow out for now. That being said, I'm rather surprised at the claim that minorities wouldn't be over-represented in crime rates due to lower aggregate socioeconomic status. Given that the analysis supporting that is behind a paywall, I don't know what the basis for that is.
Seb
Member
Sun Apr 28 09:35:55
Rugian:

That's the point of statistical significance.

If the rate of knife carrying in ethnic cohorts is statistically the same once common factors are controlled for (socio economic status etc.) then ethnicity isnt a factor.

I.e. as I've been saying all along, they aren't actually over-represented. They are represented exactly as much as you would expect them to be given factors other than their ethnicity.
Renzo Marquez
Member
Sun Apr 28 15:52:01
Seb
Member Sun Apr 28 03:48:06
"Ethnicity – recent analysis of data collected in the UK indicates that there is no statistically significant relationship between ethnicity and weapon carrying"

This citation for this is

Brennan, I.R. (2018). Weapon-Carrying and the Reduction of Violent Harm. British Journal of Criminology. Online access: https://doi.org/10.1093/bjc/azy032.

Brennan didn't actually study ethnicity though. Under the heading "Demographic factors", Brennan states:

"As the distribution of ethnicity was heavily weighted towards white respondents, using multiple categories of ethnicity would have yielded unstable statistical models. Consequently, a binary indicator, with ‘white’ and ‘non-white’ categories was generated."

How are "white" and "non-white" categorized? Brennan doesn't provide a detailed explanation. Does "white" include all Caucasians (including Middle Easterners, North Africans, Latinos, etc.)? We don't know. I'm sure "non-white" includes East Asians that will drive the numbers for that group down a bit.

Under the "Demographic risk factors" heading, Brennan discloses that:

"Studies that have directly compared ethnic groups have found that minority groups are at heightened risk of carrying a weapon compared to white respondents (Molnar et al. 2004; Hemenway et al. 2011; Swahn et al. 2013)."

I believe Brennan's conclusions are also based on self-reported data.
Renzo Marquez
Member
Sun Apr 28 15:59:43
Seb
Member Sun Apr 28 09:35:55
"If the rate of knife carrying in ethnic cohorts is statistically the same once common factors are controlled for (socio economic status etc.) then ethnicity isnt a factor."

What about if "lack of trust in the police" is one of the common, controlled-for factors? Brenann (2018)'s Abstract states:

"The study also shows that defensive factors, such as victimization and concerns about personal safety, are relevant to understanding weapon-carrying, but they are outweighed by criminogenic factors such as violence, neighbourhood disorder and, importantly, lack of trust in the police."

Seems like Seb might be relying on propaganda.
Seb
Member
Sun Apr 28 16:14:37
Renzo:

It would be a remarkable coincidence if the exact proportions and hypothetical "ethnic propensity to violence factor" just happened to be a combination such that they cancelled out to statistically equal the white population?

Is that what you are going with? Seems far less parsimonious hypothesis than race isn't an issue.

Re your second point, what if it is? The poor white population is also unlikely to trust the police.

See for example the police in Rotherham (a case I know you are familiar with) who refused to investigate crimes against poor white girls - describing them as slags who were asking for it - until a nice middle class (i.e. professional parents) girl was a victim.

Renzo Marquez
Member
Sun Apr 28 16:35:19
Seb
Member Sun Apr 28 16:14:37
"Renzo:

It would be a remarkable coincidence if the exact proportions and hypothetical "ethnic propensity to violence factor" just happened to be a combination such that they cancelled out to statistically equal the white population?"

Is this a statement or a question?

"Is that what you are going with? Seems far less parsimonious hypothesis than race isn't an issue."

I'm not "going with" anything. Just pointing out that Brennan (2018) did not actually provide any analysis on "ethnicity" and that Brennan cites the work of others who did find "that minority groups are at heightened risk of carrying a weapon compared to white respondents". You cited a report for the proposition that "there is no statistically significant relationship between ethnicity and weapon carrying". The journal article cited for the proposition in the report indicates that the author didn't actually analyze ethnicity. Instead, he only compared "white" and "non-white" (without explaining how individuals were categorized) and he may have controlled for factors (e.g., trust in police) that he shouldn't have if he wanted to accurately compare "white" vs "non-white".

"Re your second point, what if it is? The poor white population is also unlikely to trust the police."

Why not just control for IQ? We know that has a much stronger link to violent crime than parental socioeconomic status and would help dispel the rayciss notions wrongthinkers have about race/ethnicity and knife crime in the UK.
Seb
Member
Mon Apr 29 01:39:02
Renzo:

"and he may have controlled for factors"

Yes. Obviously. That's the point.

If two populations propensity to carry knives can be explained entirely due to difference in factors that are common to both, then there's no difference left to be explained by their differing ethnicity. Ethnicity is just a poor proxy for the causative factors.

Your argument that to compare ethnicities you should *not* control for other factors is scientifically illiterate.

I'll leave it there.
Sam Adams
Member
Mon Apr 29 11:02:43
"Weapon carrying"

Lol seb fail.

Meanwhile this gem just popped up.

Britain's 'pervasive horror of knife crime' reaches record for number of stabbing homicides

http://www...stabbing-homicides/3470942002/

Keep on importing third worlders seb.
Seb
Member
Mon Apr 29 11:50:23
Sam:

Carrying a knife classified as a weapon is a knife crime Sam.

Sam fail.
Seb
Member
Mon Apr 29 11:55:06
Sam, we already established that despite having a BAME demographic the same as many major US cities, London's "unprecedented" murder rate is below must equivalent US cities.

Not is the current rise driven by demographic changes.

Sam, this is practically a religion for you.
Sam Adams
Member
Mon Apr 29 12:24:53
Nah, you have fewer africans than almost all US cities, a major demographic advantage.

"Carrying a knife classified as a weapon is a knife crime Sam."

Lol. A nation of cucks.
Seb
Member
Mon Apr 29 12:53:09
London has 14% black population.

New York has 25%.

New York murder rate is close on twice London's. More than you would expect even by your bizarre theory.


"Lol. A nation of cucks."

Hilarious, you are castigating without evidence those with Afro-Caribbean heritage for having a higher propensity to commit a crime (even though they do not infact have such higher propensity) which you don't even think should be a crime.

Very logical. Very rational. Not at all a ball of emotional bumfluff devoid of any kind of logic
Sam Adams
Member
Mon Apr 29 13:09:00
"New York murder rate is close on twice London's. More than you would expect even by your bizarre theory. "

You just said that NYs african percentage was close to twice londons too. Rofl. You dont see how twice the crime with twice the africans might correlate well? Rofl again!!
Forwyn
Member
Mon Apr 29 13:11:49
Sam pretty clearly posts headlines and articles referencing knife violence; assaults and murders.

You can criminalize carrying all you like; you're the first person on this forum to attempt to use it as a data point. No one gives a shit about a knife in the pocket that hasn't been used to harm someone, except you.
Sam Adams
Member
Mon Apr 29 15:21:32
http://ktl...r-plot-in-southern-california/

One of sebs muslim converts in LAX caught in bomb terror plot.
Seb
Member
Mon Apr 29 17:57:16
Sam:

2 * 14 is not 25 Sam, and while NYC murder rate is close to twice Londons, it's close the wrong side.

So there's actually something like at least 50% of the difference that couldn't be explained even if your assumed all murders in both London and NY were committed by black people, which isn't the case anyway.

I love how your claim to be guided by stats degenerates into wishful thinking "nearly" "it's close" "if we just close our eyes and wish".

Forwyn:
No. He regularly posts stats on knife crime in aggregate. The fact you haven't taken note of that is irrelevant to me. Of which you do not know, do not speak.
Sam Adams
Member
Mon Apr 29 18:42:46
"2 * 14 is not 25 Sam"

Because 28 is so much different from 25? You really suck when it comes to real world stats. Close is actually good enough for simple comparisons in complex datasets.
Renzo Marquez
Member
Mon Apr 29 18:44:54
Seb
Member Mon Apr 29 01:39:02
"Your argument that to compare ethnicities you should *not* control for other factors is scientifically illiterate."

I never argued against controlling for other factors. For some factors (e.g., age), that's reasonable. For others, it's not.

Controlling for trust in police when trying to determine if "ethnicity" is a causal factor in knife crime is close on the reasonableness scale to controlling for serum testosterone and sex chromosomes when trying to determine if gender is a causal factor in violent crime.
Seb
Member
Tue Apr 30 02:07:28
"Because 28 is so much different from 25?"

Well to be precise its 10% bigger which is significant.

Quite shoddy. No wonder you get things so wrong if you repeatedly accumulate 10% errors but fail to account for them. Basically innumerate.
Seb
Member
Tue Apr 30 02:10:44
Renzo:

I see. In your mind then, not trusting the police to investigate a crime *against* you has the same relationship to propensity to commit crime as sex chromosomes does to biological sex?

This is certainly a courageous argument, but one that I fear is unsustainable.



smart dude
Member
Tue Apr 30 05:14:47
According to Wikipedia, the black population of London is 13.3%. The black population of NYC is 25.5%. (13.3)*2 is actually about 4.3% more than than 25.5.

So Seb's "10%" is actually off by around 132%. That's pretty extreme.
TJ
Member
Tue Apr 30 09:03:11
A PowerPoint Presentation:

Justice Matters:
Disproportionality
MOPAC Evidence & Insight?????
March 2018

http://www...ortionality_slides.pdf#page=11
Seb
Member
Tue Apr 30 11:23:33
Smart dude:

The number both Sam and I were using was 14% however and the point of dispute was his ability to understand relative magnitudes.

Sam Adams
Member
Tue Apr 30 11:50:49
Seb clearly does not have the math ability to understand estimates and reasonable error rates in different datasets.

If i am off by 10% on the velocity of my spacecraft, an environment with a very simple set of influences, i am completely wrong.

If i am off by 10% on flight time between heathrow and kennedy, thats a little more excusable but still not good.

However, if I am within 10% on the height of a tsunami or the crime rate of a major city, that is good enough... reasonably accurate in a much more complex set of datas.
Seb
Member
Tue Apr 30 15:19:53
Sam:

Oh dear, it sounds like you are confusing precision, variance and bias.

Precision greater than the *variance* of a parameter is pointless. But the variance of the demographics is nothing like 10%, so it is simply wrong to assert a lack of precision on the order of 10% is small compared to the variance of the system under analysis.

It is just an excuse to try and claim the data fits your theory - if you allow imprecision far greater than the statistical resolution the data allows in order to ensure your theory fits with the empirical data, that's called scientific malpractice.

Sam Adams
Member
Tue Apr 30 15:42:57
I said nothing of bias recently.

Meanwhile 10% is much smaller than the massive variance of crime data, and is thus an acceptably small error to be neglected.

This is the most fundamental stats theory that you keep butchering. In short, your mind is weak and you are beneath me.
Senor Marquez
Member
Fri May 03 22:18:00
Is it safe for me to pretend to be a lawyer again here?
Dukhat
Member
Sat May 04 00:38:58
Don't you get tired of arguing with an idiot Seb? Pry deeper and there's nothing there.

Kindergartener, incel logic. It's actually qutie boring.
Seb
Member
Sat May 04 06:36:47
Dukhat:

Arguing with Sam is what I do when I'm bored.

Sam:
Crime data in a region doesn't randomly vary by 10% in reporting periods. Those kinds of changes are driven by causal factors.

show deleted posts

Your Name:
Your Password:
Your Message:
Bookmark and Share