Author Topic: Episode #606  (Read 2973 times)

0 Members and 1 Guest are viewing this topic.

Offline werecow

  • Cryptobovinologist
  • Frequent Poster
  • ******
  • Posts: 3426
  • mooh
Re: Episode #606
« Reply #30 on: February 24, 2017, 11:39:34 AM »
Hmm I'm trying to interpret the Pew graph and from what I understand it seems to show that, if we take a slice of an age group say 16 year olds, 25% of the criminals are native born, 23% are second gen. immigrants and ~17% are first gen. immigrants. So that makes up 65% of the criminals. Who are the rest 35%? Tourists?

That's not what the graph is showing.

According to this graph, if you take a slice of 16 year olds, 25% of the native born 16 year-olds have been involved in a crime, 23% of the 2nd gen 16 year-olds have been involved in a crime, and 17% of the 1st gen 16 year-olds have been involved in a crime.

It is not a pie chart of all crimes committed.  I would imagine that the 25% of native born 16 year-olds who have been involved in a crime would be a MUCH larger proportion of the number "all crimes committed by 16 year-olds", as I don't imagine that the ratio of all native born to 2nd gen to 1st gen 16 year-olds is 1:1:1.

Hah I think you're right. I originally interpreted it similarly but got confused when someone else interpreted the graph the other way because they thought that it was ridiculous that 25% of 16 year olds were criminals. It could be that most people interpret 'crime' and 'criminal' as something terribly violent which a lot of it is but I bet that a lot of the 16 year olds get their sentences for smoking weed or something like that.

25% is a lot, though... I don't live in the U.S., so it's probably different over there, but I don't think I know a single person who has ever been convicted of a criminal offense. Am I that sheltered?
Mooohn!

Offline amysrevenge

  • Baseball-Cap-Beard-Baby Guy
  • Stopped Going Outside
  • *******
  • Posts: 4987
  • The Warhammeriest
Re: Episode #606
« Reply #31 on: February 24, 2017, 11:53:53 AM »
Hmm I'm trying to interpret the Pew graph and from what I understand it seems to show that, if we take a slice of an age group say 16 year olds, 25% of the criminals are native born, 23% are second gen. immigrants and ~17% are first gen. immigrants. So that makes up 65% of the criminals. Who are the rest 35%? Tourists?

That's not what the graph is showing.

According to this graph, if you take a slice of 16 year olds, 25% of the native born 16 year-olds have been involved in a crime, 23% of the 2nd gen 16 year-olds have been involved in a crime, and 17% of the 1st gen 16 year-olds have been involved in a crime.

It is not a pie chart of all crimes committed.  I would imagine that the 25% of native born 16 year-olds who have been involved in a crime would be a MUCH larger proportion of the number "all crimes committed by 16 year-olds", as I don't imagine that the ratio of all native born to 2nd gen to 1st gen 16 year-olds is 1:1:1.

Hah I think you're right. I originally interpreted it similarly but got confused when someone else interpreted the graph the other way because they thought that it was ridiculous that 25% of 16 year olds were criminals. It could be that most people interpret 'crime' and 'criminal' as something terribly violent which a lot of it is but I bet that a lot of the 16 year olds get their sentences for smoking weed or something like that.

25% is a lot, though... I don't live in the U.S., so it's probably different over there, but I don't think I know a single person who has ever been convicted of a criminal offense. Am I that sheltered?

Again, though, there's so much leeway available in "involved in" a crime that goes beyond "committed, arrested, and convicted of" a crime.

Underage drinking is a crime, isn't it?  Or a least the providing of alcohol to minors is, so you're by even the strictest definition involved in some sort of crime if you've had the tiniest sip of beer at any point before reaching your local minimum drinking age (18 where I live, 21 in many places in North America).  I did a looooooot of underage drinking myself...  If you include underage drinking as "involved in" a crime, in my locale in 1991 (age 16 lol) it would have been more like 95% than 25%.

I had been in the presence of someone smoking weed by age 16, in an area where possession of any amount of MJ was a crime - does that count as "involved in" a crime?  You could describe "involved in" to include that or to exclude that, depending on how you want your results to look.

There's just so much about regular everyday life that is adjacent to minor crimes.
Big Mike
Calgary AB Canada

Offline werecow

  • Cryptobovinologist
  • Frequent Poster
  • ******
  • Posts: 3426
  • mooh
Re: Episode #606
« Reply #32 on: February 24, 2017, 12:04:05 PM »
Hmm I'm trying to interpret the Pew graph and from what I understand it seems to show that, if we take a slice of an age group say 16 year olds, 25% of the criminals are native born, 23% are second gen. immigrants and ~17% are first gen. immigrants. So that makes up 65% of the criminals. Who are the rest 35%? Tourists?

That's not what the graph is showing.

According to this graph, if you take a slice of 16 year olds, 25% of the native born 16 year-olds have been involved in a crime, 23% of the 2nd gen 16 year-olds have been involved in a crime, and 17% of the 1st gen 16 year-olds have been involved in a crime.

It is not a pie chart of all crimes committed.  I would imagine that the 25% of native born 16 year-olds who have been involved in a crime would be a MUCH larger proportion of the number "all crimes committed by 16 year-olds", as I don't imagine that the ratio of all native born to 2nd gen to 1st gen 16 year-olds is 1:1:1.

Hah I think you're right. I originally interpreted it similarly but got confused when someone else interpreted the graph the other way because they thought that it was ridiculous that 25% of 16 year olds were criminals. It could be that most people interpret 'crime' and 'criminal' as something terribly violent which a lot of it is but I bet that a lot of the 16 year olds get their sentences for smoking weed or something like that.

25% is a lot, though... I don't live in the U.S., so it's probably different over there, but I don't think I know a single person who has ever been convicted of a criminal offense. Am I that sheltered?

Again, though, there's so much leeway available in "involved in" a crime that goes beyond "committed, arrested, and convicted of" a crime.

Underage drinking is a crime, isn't it?  Or a least the providing of alcohol to minors is, so you're by even the strictest definition involved in some sort of crime if you've had the tiniest sip of beer at any point before reaching your local minimum drinking age (18 where I live, 21 in many places in North America).  I did a looooooot of underage drinking myself...  If you include underage drinking as "involved in" a crime, in my locale in 1991 (age 16 lol) it would have been more like 95% than 25%.

I had been in the presence of someone smoking weed by age 16, in an area where possession of any amount of MJ was a crime - does that count as "involved in" a crime?  You could describe "involved in" to include that or to exclude that, depending on how you want your results to look.

There's just so much about regular everyday life that is adjacent to minor crimes.

Sure, I get that. If they went by conviction rates, for 25% to be convicted of a crime is still absurd to me. So that raises the question of how they arrived at that number; did they poll people aged 16 to ask them if they'd ever done something punishable by law?
Mooohn!

Offline amysrevenge

  • Baseball-Cap-Beard-Baby Guy
  • Stopped Going Outside
  • *******
  • Posts: 4987
  • The Warhammeriest
Re: Episode #606
« Reply #33 on: February 24, 2017, 12:09:41 PM »
I get that. If they went by conviction rates, for 25% to be convicted of a crime is still absurd to me. So that raises the question of how they arrived at that number; did they poll people under 16 to ask them if they'd ever done something punishable by law?

Polling has to be how, there's no other way to get at "involved in" than polling.  We don't track "involved in" (I think you're still hung up on conflating "committed a crime" and "involved in a crime").

Another anecdote: in high school I was in the car of a person I only barely knew well enough to get a ride from, since we were both leaving one place and heading to the same next place he offered the ride.  On the way, he stopped to fucking throw some eggs at a mosque.  I yelled at him and got out of his car and walked the rest of the way (and was too much of a coward to ever report it; if you're going to judge me do it quietly - 41 year old Mike has more balls than 17 year old Mike did).  Nobody was ever arrested or stopped or anything, and yet I would answer a pollster that I had been involved in a crime.
Big Mike
Calgary AB Canada

Online Ah.hell

  • Poster of Extraordinary Magnitude
  • **********
  • Posts: 11433
Re: Episode #606
« Reply #34 on: February 24, 2017, 12:13:56 PM »
Yep, how was the data collected?  "involved in a crime" is a nebulous term.  It could mean victim as well as perpetrator.  I'd like to know who this data was collected?  Self reporting?  I'd think the answer should be 100% if it just means adjacent to someone who committed a crime and well above 505 50%.  if it meant committed any crime no matter how trivial.  Jay walking, drinking under age, giving booze to someone underage(no matter what age you were at the time), drugs of any kind, drunk sex, trespassing, vandalism, etc.   All the crimes attorneys and cops know you committed that you don't even know are crimes. 

Side note: because I've changed my social circle quite a lot over the years.  There are people who don't know anyone who hasn't been to jail and who would be surprised to find out you haven't been and there are people who don't know any one who has been to jail and who would be surprised to find out you have been.  I like to remind my current associates of that occasionally.

Edit, I was going to change the typo but its redundant so....Also other typos.

« Last Edit: February 24, 2017, 12:21:00 PM by Ah.hell »

Offline amysrevenge

  • Baseball-Cap-Beard-Baby Guy
  • Stopped Going Outside
  • *******
  • Posts: 4987
  • The Warhammeriest
Re: Episode #606
« Reply #35 on: February 24, 2017, 12:17:21 PM »
Yep, how was the data collected?  "involved in a crime" is a nebulous term.  It could mean victim as well as perpetrator.  I'd like to know who this data was collected?  Self reporting?  I'd think the answer should be 100% if it just means adjacent to someone who committed a crime and well above 505 if it meant committed any crime no matter how trivial.  Jay walking, drinking under age, giving booze to someone underage(no matter what age you were at the time), drugs of any kind, drunk sex, trespassing, vandalism, etc.   All the crimes attorneys and cops know you committed that you don't even know are crimes.

And the problem is one that I already hinted at - even if they have a rock-solid perfectly valid objective and real definition of "involved in" in their methods in the study, this graph is what's floating around news sites and social media with no definition of "involved in" attached to it.  So people can use it to support their pre-existing notions (even notions that I agree with).
Big Mike
Calgary AB Canada

Offline Soldier of FORTRAN

  • Too Much Spare Time
  • ********
  • Posts: 6759
  • Cache rules everything around me.
Re: Episode #606
« Reply #36 on: February 24, 2017, 03:02:35 PM »
Reminds me of this humdinger:

Quote
NEW YORK (AP) — Nearly 50% of black men and 40% of white men are arrested at least once on non-traffic-related crimes by the time they turn 23, according to a new study.
From: USA Today (January 2014)
Every soup ladled to the hungry, every blanket draped over the cold signifies, in the final sense, a theft from my gigantic paycheck.

Offline werecow

  • Cryptobovinologist
  • Frequent Poster
  • ******
  • Posts: 3426
  • mooh
Re: Episode #606
« Reply #37 on: February 24, 2017, 03:57:57 PM »
Reminds me of this humdinger:

Quote
NEW YORK (AP) — Nearly 50% of black men and 40% of white men are arrested at least once on non-traffic-related crimes by the time they turn 23, according to a new study.
From: USA Today (January 2014)

That's crazy. I wonder what the statistics are over here.
Mooohn!

Offline Rien

  • Brand New
  • Posts: 2
Re: Episode #606
« Reply #38 on: February 25, 2017, 10:24:22 AM »
Those of us who are white and middle-class in America have a very different experience with the police than poor people and people of color do. Working with the homeless for about 5 years I saw the police treat homeless guys and Native Americans very differently than they treat middle-class white folks. Even well-meaning cops treated these guys very differently. It's hard to believe that the nice cops who arrested me would treat Indians so differently, but they did.

Either the cops are (obviously) "bad cops", or they have a reason for their behaviour.
I would tend to believe the later, and am very curious as to what that reason could be.
Is there any research in that direction?

Offline Rien

  • Brand New
  • Posts: 2
Re: Episode #606
« Reply #39 on: February 25, 2017, 10:45:19 AM »
How do the numbers look if you don't correct for the socioeconomic factors, regarding immigrants and crime?

Indeed. I had that very same objection.

The thesis that they spoke about was the question if immigrants are more often involved in crime, but then they subtly redefined it as "when compensated for socioeconomic factors".

However they then did not discuss the average socioeconomic status of immigrants.

Offline LaPalida

  • Frequent Poster
  • ******
  • Posts: 3689
  • To thy own self be true.
Re: Episode #606
« Reply #40 on: February 27, 2017, 04:48:16 AM »
Hah I think you're right. I originally interpreted it similarly but got confused when someone else interpreted the graph the other way because they thought that it was ridiculous that 25% of 16 year olds were criminals. It could be that most people interpret 'crime' and 'criminal' as something terribly violent which a lot of it is but I bet that a lot of the 16 year olds get their sentences for smoking weed or something like that.

I think the biggest factor in catching more people is the definition of "involved in" a crime.  That is a MUCH broader net than "committed" a crime would be.

I would think that you could word "involved in" loosely enough that it would be shocking not to have a much higher number - having been in a car where the driver exceeded the speed limit could, under some definitions of "involved in a crime", count as Yes for instance.  So it ends up being a major researcher degree of freedom to let you introduce whatever bias you want - all that most people are going to see is the graph, not the detailed minutiae of your methods.

Yes this is true as well. Involved in a crime may mean that as well. I don't think that speeding is a crime though. Not unelss you go to jail for it. I assume that crime actually means criminal offense and not a civil offense. Speeding is a civil offense and so cannot be a crime by that definition.
« Last Edit: February 27, 2017, 04:56:21 AM by LaPalida »
"First get your facts; then you can distort them at your leisure."
- Mark Twain

Offline LaPalida

  • Frequent Poster
  • ******
  • Posts: 3689
  • To thy own self be true.
Re: Episode #606
« Reply #41 on: February 27, 2017, 04:54:17 AM »
Yep, how was the data collected?  "involved in a crime" is a nebulous term.  It could mean victim as well as perpetrator.  I'd like to know who this data was collected?  Self reporting?  I'd think the answer should be 100% if it just means adjacent to someone who committed a crime and well above 505 if it meant committed any crime no matter how trivial.  Jay walking, drinking under age, giving booze to someone underage(no matter what age you were at the time), drugs of any kind, drunk sex, trespassing, vandalism, etc.   All the crimes attorneys and cops know you committed that you don't even know are crimes.

And the problem is one that I already hinted at - even if they have a rock-solid perfectly valid objective and real definition of "involved in" in their methods in the study, this graph is what's floating around news sites and social media with no definition of "involved in" attached to it.  So people can use it to support their pre-existing notions (even notions that I agree with).

Yes exactly. Personally I think it's a very misleading graph. I had trouble interpreting it, not even sure if I'm doing it right still. I did a search by image and yes a lot of news sites use the graph in their reporting to support their partisan view. I wish that people that did these studies (including PEW) were clearer on what they are actually saying.
"First get your facts; then you can distort them at your leisure."
- Mark Twain

Offline brilu34

  • Off to a Start
  • *
  • Posts: 19
Re: Episode #606
« Reply #42 on: March 09, 2017, 08:17:31 PM »
Here is an episode of Star Talk: Playing with Science from 3-8-17 titled How Technology is Changing Football. Neil deGrasse Tyson asks Jets QB Ryan Fitzpatrick if momentum in sports exists, Then Chuck Nice asks Dartmouth Football Coach the same question. Both of them agree that momentum in sports is a real thing. Here is the link the question begins at 52:37. https://www.startalkradio.net/show/technology-changing-football/ 

 

personate-rain