4 ms·
> Every assault makes the news, but cities are safer than they've ever been Reminds me of one of my favorite pieces of data.[0] Pew tracks both crime rates and
by godelski 5y ago
> Every assault makes the news, but cities are safer than they've ever been
Reminds me of one of my favorite pieces of data.[0] Pew tracks both crime rates and perception of crime rates. There's been an overwhelming declining trend crime (especially violent) in the past 30 years. But if you look at perceptions of crime, people think year over year it is on the rise. There's even a bias that it is on the rise more nationally rather than locally. What I like about this is how it shows the biases that we develop. You're right that _every_ assault makes the news (also there's 350m people in the US, even a 1 in 10k event is likely to occur frequently at the national level). Also, pareto plays a role. Most crime is limited to a few states, and really just a few cities in those states, and even more so specific areas within those cities (pareto is fractal in this way). We humans are typically bad at risk analysis.
I should add to this that I grew up and walked to school in the 90's. So it is safe to say that me walking to school as a kid was statistically substantially more dangerous than it is for kids to walk to school today.
[0] https://www.pewresearch.org/fact-tank/2020/11/20/facts-about-crime-in-the-u-s/ https://www.pewresearch.org/fact-tank/2020/11/20/facts-about...
- wolverine876 5y agoWhile I generally agree ... > Most crime is limited to a few states, and really just a few cities in those states, and even more so specific areas within those cities (pareto is fractal in this way). What is that based on? Some tangential data leans in the other direction: The most recent FBI/DoJ report (that I'm aware of) said recent increases in gun crime were spread pretty evenly, urban and rural, across the country - though that's just one kind of crime, and only the increases. Another report said crime is worse in 'red states' than 'blue states', and the latter tend to have larger cities (I think). Hardly conclusive data, I know, and I'd love to see some.
- godelski 5y agoThis is pretty common in pareto distributions. Here's a heat map from 2020 data with state level resolution[0]. You'll notice that crime is in fact not evenly distributed (my claim). Here's a similar map but with county level precision[1]. And here's a heat map for crime in San Francisco[2]. Nothing I've stated is really that controversial and is pretty well known by anyone that looks at crime statistics. I think it is also relatively easy for the average person to buy because everyone knows where "the bad part of town" is. Really that's all I've said. That "the bad part of town" exists fractally. There's the "bad states," "bad cities," "bad counties," "bad towns," and "bad parts of towns." I think this all should be rather unsurprising. To clarify on your FBI/DOJ statement, this doesn't have to be in disagreement with what I said. You can have increases in violent crime over the last year (or even 5) and a dramatic decline in the last 30. The Pew study that I cited does specifically look at violent crime and uses FBI statistics. But let's add more. Here's a DOJ 1993-2011 showing a large decrease in both fatal and non-fatal gun violence[3]. Here's a Reuters article that shows homicides (and the pareto effect I mentioned above) from 1980 to 2019[4]. Notice the trend towards declining (I never said it _strictly_ declined). [0] https://upload.wikimedia.org/wikipedia/commons/5/51/Violent_Crime_Rate_by_U.S._state.svg https://upload.wikimedia.org/wikipedia/commons/5/51/Violent_... [1] https://www.washingtonpost.com/graphics/national/crime-rates-by-county/ https://www.washingtonpost.com/graphics/national/crime-rates... [2] https://www.researchgate.net/figure/Crime-heat-map-and-overview-of-the-considered-zip-codes-in-San-Francisco_fig1_337585003 https://www.researchgate.net/figure/Crime-heat-map-and-overv... [3] https://info.publicintelligence.net/DoJ-GunViolence1993-2011.pdf https://info.publicintelligence.net/DoJ-GunViolence1993-2011... [4] https://www.reuters.com/world/us/biden-plans-new-steps-combat-us-gun-violence-violent-crime-climbs-2021-06-23/ https://www.reuters.com/world/us/biden-plans-new-steps-comba...
- wolverine876 5y agoThanks for doing all the research. I will keep this tab open and catch up on it later. One comment: > Nothing I've stated is really that controversial and is pretty well known by anyone that looks at crime statistics. I think it is also relatively easy for the average person to buy because everyone knows where "the bad part of town" is. 'What everyone knows' is not knowledge; that's why we need data, research, etc (which you have amply provided). Look at what BS people 'know' about all sorts of things. IME the 'bad part of town' is usually not so bad, if at all, and it's often conflated with race.
- kkielhofner 5y agoThis is one of favorites too! I was recently chatting with a friend who (like me) grew up in the 90s and they were reminiscing on how great walking to school was because “you could do that back in the day.” They then went on to say that they would NEVER allow that for their children “these days”. I pulled up the Pew study and it blew their minds. That said the news cycle, perception, parental instinct, etc is too powerful and fear (however irrational) is extremely powerful. After the initial shock they thoughtfully considered it and essentially said “Well, whatever. The kids are still never going to just walk around outside.” Those kids are never going to be out “in the wild” even though it’s substantially safer today than it was in the 90s when we were doing it.
- godelski 5y agoIt is an interesting response to me (without kids) because I think this doesn't give kids the chance to explore. Kids SHOULD be doing dangerous things, just not TOO dangerous. That's how I learned a lot of valuable lessons. It is also how you learn independence.
- Spooky23 5y agoCrime stats are a difficult thing to make comparisons with between jurisdictions and over time. I grew up in 1980s NYC. They flat out didn’t respond to take reports for certain types of crime, and may deliver a beat down to some weirdo following kids around. Unless it’s a clearly serious felony like a murder or burglary, you aren’t really getting any information. I would say that what people worry about is random freakshow people and the petty crime form other kids or teens. Folks are more aware of the freaks, and the evidence of in front of them that there are consequences for most petty crime. We walked to school about 12 blocks in those days. It was fine. I live in a small city today; my kid won’t bike to school alone until about 12, mostly due to traffic issues and idiots riding ATVs in the afternoon.
- godelski 5y ago> Unless it’s a clearly serious felony like a murder or burglary, you aren’t really getting any information. Good news! The numbers reported in my citation are specifically about: Larceny/Theft, Burglary, Aggraveted Assault, Motor Vehicle Theft, Robbery, Rape, and Murder/Non-Negligent Manslaughter. Every single one of these has drastically decreased since the 90's (90's increased in crime since the 80's but current levels are below that). I don't mean to sound sarcastic, but I can't but help but feel that this could have better been resolved by clicking on the citation I included.
- adrianN 5y agoI interpreted the comment you're replying to such that while the statistics for the serious crimes show a marked decline, parent are not only concerned about those crimes when letting their children walk alone to school and we have poor data about less serious crimes.
- godelski 5y agoMy comment was two parts. 1) that crime overall is decreasing. As far as I can tell this is true for every measurable metric. The only two I'm aware of that aren't are opioid use and sexual assault. Though the latter one isn't commonly believed to actually be increasing but just higher reporting. 2) that locality of crime is non-uniform. As a kid we knew which areas of town not to go to alone. Stay out of those parts of town and you'll be fine. In addition to this, most crime is not evenly distributed across the country and your suburban town isn't that dangerous. I'll strengthen my claim even more though, though this still isn't controversial. The richer your zipcode, the lower the crime rate. I think what people in this thread are more discussing is wealthy suburbanites picking their kids up in Teslas saying it is too dangerous to walk. No one is questioning the same sentiment for kids in Detroit or the bad parts of Chicago.
- aksss 5y agoI think the standard has changed, right? People want everything nerfed up. We have helmet laws where there used to be none, you can't use the Uber app without being bombarded with messages about 'safety', kids aren't left on their own outside. The mentality is that the world is scary - if not outright ending - constantly a threat. We wonder why we're raising kids with anxiety problems.