Why Can't Anyone Agree On The Crime Rate?
Shirin Ghaffary looks into why the crime rate in America can be such a confusing, and often misleading, topic to read and write about.
SUBSCRIBE to Above the Noise: [kzread.info...]
ABOVE THE NOISE is a show that cuts through the hype and takes a deeper look at the science behind controversial and trending topics in the news. Hosted by Myles Bess and Shirin Ghaffary.
NEW VIDEOS EVERY OTHER WEDNESDAY
The news media is chock-full of contradictory stories about crime in the United States. Are murders on the rise, or at remarkable lows? A skim of the headlines might not give you a clear answer. So why is there room for disagreement about what should be a very basic statistic?
The answer isn’t really about the data itself, but how we slice and dice that data. It’s about how we determine trends, what we’re comparing, and sometimes, what answer we want to find.
In this video, Shirin figures out how to get beyond the headlines and understand what’s really happening to the crime rate in America. And she looks at why all this data wrangling has given the American public some serious misunderstandings about crime.
CHECK OUT THE FBI’S CRIME DATA YOURSELF:
www.ucrdatatool.gov/Search/Cr...
OTHER SOURCES:
How politicians claim opposite things about crime:
www.factcheck.org/2016/07/duel...
What most Americans think about the crime rate:
www.gallup.com/poll/186308/ame...
Why just reading about something makes you believe it:
www.morilab.net/gakushuin/Zajo...
Follow KQED:
KQED: www.kqed.org/
Facebook: / kqed
Twitter: KQED?lang=en
Teachers follow KQED Learning
KQED Learning: ww2.kqed.org/learning/
Facebook: / kqedlearning
Twitter: KQEDedspace?lang=en
About KQED
KQED, an NPR and PBS affiliate in San Francisco, CA, serves Northern California and beyond with a public-supported alternative to commercial TV, Radio and web media.
Funding for Above the Noise is provided in part by S.D. Bechtel, Jr. Foundation, David Bulfer and Kelly Pope, Horace W. Goldsmith Foundation, The Dirk and Charlene Kabcenell Foundation, The Koret Foundation, Gordon and Betty Moore Foundation, Smart Family Foundation, The Vadasz Family Foundation and the members of KQED.
Пікірлер: 86
Annual crime rates rise despite decades-long decline
30-Year Crime Drop Stalls
"Unexplained uptick in historically low crime rates"
@Darasilverdragon
6 жыл бұрын
This is probably the best one I've seen so far. It informs both sides of the issue with an eyecatching short and punchy title, while also baiting the reader in with the mystery of the word 'unexplained' It's like some kind of bizarre 'good' clickbait.
"Video games cause drop in crime rates! subtitle: Since the release of the Nintendo NES crime has been halved!" Am I doing it right?
@Armendicus
7 жыл бұрын
Hell yes. All hail the nerd master race!
@gregsquires6201
6 жыл бұрын
You're doing it right according to the media. Combining misleading headlines with causation vs. correlation errors is standard practice for most media sources. Nicely done!
This is exactly the sort of information I'm looking for, great video :-) Keep up the good work!
@multiplio2924
7 жыл бұрын
xisumavoid Hello X! I think one fact we can all agree on is how awesome the tone of your new guitar sounds!
@Darasilverdragon
6 жыл бұрын
Good to know one of my favorite youtubers likes to stay well informed :D
@AboveTheNoise
6 жыл бұрын
Wow, somehow we missed your comment way back when, but here's a belated thanks for watching!
@theHusky2490
6 жыл бұрын
Wow! You are the first KZreadr I'm subscribed to that I've found in the wild. Bravo!
Here's a headline "IS CRIME UP OR DOWN? LOOK AT THE DATA" or something along those lines
"Crime is lower than it was when your parents were your age, but higher than last year"
Crime rates - a complicated subject.
"Blame Chicago"
@DyxoXinoro
7 жыл бұрын
Underrated post
This is exactly why this is an awesome channel
@AboveTheNoise
7 жыл бұрын
DJBsLectures thanks for the support! We're glad you like the channel - help us spread the word!
I love this channel! I get both sides of the story! You guys earned a new sub
The Roller Coaster of Crime rates-both ups and downs
Great video, thanks
Crimes are at a low yet have also slightly rised
Nice video the channel is growing slowly and remember me when the channel gets popular
All I have to say, is thank you for taking the time and energy to not flow with the media.
@AboveTheNoise
6 жыл бұрын
Thanks for taking the time to watch our video and share your feedback!
Very informative
Crime rate decreases with occasional spikes in crime doing little to budge the rising peacefulness.
B-b-but that's Google Sheets 😂
Good vid.
I love how you are explaining policy not in simple terms, but how complicated it... Using psychology, data analysis, policy.. I am definitely going to use your channel as a teaching tool....My headline: Question the crime rate data, don't be a fool... This phenomenon took place in the early 90s when crime rates were high, and later it dropped, researchers think it was due to abortion and not policing techniques like Guiliani believed.
Honestly what you just said would make a great headline
"I can't believe it's not statistically significant - These crime rate statistics will leave you uncertain"
Nice video!
@AboveTheNoise
7 жыл бұрын
Richsombie thanks!
You have my sub, thanks to xisuma who tweeted about you, nice channel.
@AboveTheNoise
7 жыл бұрын
Thanks for the sub! We really appreciate it!
Confusing crime statistics, but we know they're good! *headline*
"While crime rates in the US have dropped over the past several decades, we've not had a historically new low in several years"? Although that's a little too long for anyone to want to read and could mean different things to different people so let's just go with "crime rates in America?" And let the article do the non bias full picture speaking
@RimaNari
7 жыл бұрын
Agreed. Away with these "let's summarize a several page long article in one half sentence and all the while be perfectly aware that 90 % of people don't read further" headlines.
Lowest Murder Rates In Thirty Years Do The Macarena
"Murder, it still happening after all these years"
What about thefts? Or carjacking? And sexual assault? What's the trend with the non-murder crime?
@tomcmustang
7 жыл бұрын
Violent crime has a specific definition not limited to murder. ucr.fbi.gov/crime-in-the-u.s/2011/crime-in-the-u.s.-2011/violent-crime/violent-crime Here is Property Crime as an example of another category of crime: ucr.fbi.gov/crime-in-the-u.s/2010/crime-in-the-u.s.-2010/property-crime Also interesting, the clearance rates based on arrest (Percent of crimes closed because of arrest): ucr.fbi.gov/crime-in-the-u.s/2010/crime-in-the-u.s.-2010/clearances Long story short, all crimes are down with violent crimes being particularly down, while clearance by arrest is up.
A 30-Year Crime Drop Halts Are YOU and Your Loved Ones Still Safe? Or Is This all a HUGE Media Made Overreaction? The Answer: ¯\_(ツ)_/¯
crime is rising in an elevator going down
"Despite a 10% increase from 2015 to 2016 overall crime rates still at 30 year lows"
It's an easy answer as to why crime is at an all time low compared to 30 years ago... because technology has advanced along with DNA testing, to a point where it's almost impossible to get away with murder. Therefore, those bold individuals aren't risking those chances, although I'm sure those urges are still there for whatever individual reasoning they may have.
Murder is a very definable statistic you can compile, happiness is not. Any change every year is very important to determine the effectiveness and avoid future bumps, so they should not be considered inconsequential. I had hopes for this, so far not so good.
@6ThreeSided9
6 жыл бұрын
"Any change every year is very important to determine the effectiveness and avoid future bumps" Do you have evidence for this claim? In case you haven't noticed, statistical charts tend to not be a convenient straight line in one direction. When you flip a coin you will occasionally get 5 tails in a row, but that doesn't mean that the probability of the coin getting tails has changed. To get a real understanding of a trend you have to measure results over an extended number of data points. Every little bump doesn't matter, since that bump alone isn't enough to represent statistical significance that accurately reflects reality.
Well, it still depends. There is no true objectivity in one-line punch lines. I can write "Murder rates are drastically rising after decade-long decline" and it would not only be true, but also it would not be bending truth or be overly sensational. When looking at a data curve it is often a good idea to look at the derivative of the data as well, i.e. the increase of the value measured. In this case the rise in murder rate in recent years is very steep which is much more troubling then a steady rise over a longer period of time, even if the overall increase is the same. Judging from the graph in the video such a steep murder rate increase has not been observed in the 21st century before and has overall only been observed once or twice since records started. This would make my headline perfectly valid. On the other hand the trend could go down in the next year just as well, making this just an anomaly in the data. Leaving out this anomaly, murder rate certainly is at its lowest point, so I could write "Murder rates are as low as never after decade-long decline". Note the difference: Here I say "are" - absolute value - and before I said "are " - change of value. Even more interesting it gets if one were to overlay a graph with "amount of people dying in mass murder" per year. Maybe the sharp increase can be explained by a single event where a lot of people died, like a mass shooting. Or alternatively, just plot data from murders where less then 3 people died, and everything may look drastically different and the bump at the end may be gone entirely. Statistics is a fun and fascinating thing, but also dangerous. You're doing a great job with this video, take my sub ;-)
That’s only if these numbers are actually correct
Contradictory data has to do with your point of view and which figures you are comparing. If you are comparing numbers from 1978 to current then you are a following a trend of decline in violent crime. If you are selectively choosing another different time line then you can say they are up because from one certain point to another violent crime rates fluctuate and you are taking note of a small up tick compared to the national long term trend of violent crime hitting general lows.
'Crime rate down 50% from 1980s, but rising in past few years'
What the smile
Sounds like p-hacking that happens i scientific research. Depending on how you slice the data, you can get statistically significant answer. Suppose you have a significance of 5%. If you slice the data 20 different ways, one of those will return significant even when there would be no reason to suspect otherwise. See xkcd.com/882
@RimaNari
7 жыл бұрын
Good point. The binning might be interesting too. Binning the data with double precision might reveal something completely different in the last couple data points. Or binning with half precision might look the rise in the end like being constant.
Data Analysis: It's Murder
you guys should try to make a video about hackers are/or hacking......If you remember CityKid2100 from your other videos it's still the same person but a different account because the old CityKid2100 got hacked.....
@AboveTheNoise
7 жыл бұрын
Sorry to hear that you got hacked! You're on to something though, cybersecurity is definitely a topic we've been tossing around and brainstorming about. Is there anything in particular about hacking you'd like to know about it?
@Julian.BrawlStars
7 жыл бұрын
Above The Noise No there's nothing in particular....well actually...what about major hacking hotspots and how to avoid getting hacked?
IS CRIME UP OR DOWN? DECIDE FOR YOURSELF.
@AboveTheNoise
7 жыл бұрын
Alex Ozzie Ha! I like where you are going with this.
I noticed that your murder rate statistic was based on per 100,000 ppl if you were to remove that filter wouldn't it indicate crime rate is increasing (due to population increase)? If so, that would mean the amount of murders committed per year has increased since 1985, but you are still statistically less likely to be murdered correct?
👏👏👏
If you call the thing a "Bible for crime rates", how is it you're wondering how people who look at it come up with completely different interpretations? It would be contradictory in itself, if it's anything like the Bible, no?
"Rise in Crime after Historic Low"
Pretty much, crime is going down, unless you manipulate the data.
@Stars-Mine
7 жыл бұрын
2 points of data does not a trend make
i'm happy to live in a place where we worry about people robbing using toy guns.. cause real guns are not really accessible. at least not to the average junk.
Headline: "Crime rate is both raising and fall"