Oh no!  Look, violent crime is the highest it's been in 10 years.  Well, it's not 2020 anymore, and this data is only through 2020, but you get the point... 

But in reality, when you look at the full graph.....

Violent crime remains at historic lows, effectively "flat" (relatively speaking) at about 400 or so incidents per 100,000 people for each of the past 10 years as compared to the prior 25 (1985-2010).

It's easy to lie with data, when you (or your audience) have a pre-determined outcome, narrative, or bias in mind.  Also, when they can't be bothered to research, think critically, look at the full data picture, spend 5 seconds Googling, etc. etc. 

Basic data literacy is something we really need to get better at as a country.  It won't solve all of our problems, or even many of them.  But, it can certainly help move us back towards all living in the same shared reality - even if we continue to differ on what problems need tackling, and how to tackle them.

Sidenotes:

There was a major data system transition in 2021, and not all participating agencies fully made the switch.  As noted in the CDE site linked below, this means 2021 data (not shown in these graphs) isn't comparable to pre-2021 data.  I do not know if this holds true for 2022, but that annualized data isn't available yet, anyway. Changes in reporting systems, standards for what counts in certain categories, and all of those things can make a 15 second soundbite or summary hard to parse, especially without the full backstory and picture of all the inputs.

In addition, the obvious disruptions and atypical world we lived in due to the COVID pandemic in 2020, and to a lesser extent 2021 and 2022, obviously may have created statistical aberrations, both positive and negative, which can either be understood in that context, or used for nefarious purposes by those with an agenda.  Think hypothetical headlines like "Murders Up 250%" comparing 2021 to 2020, when, for instance, everyone being at home, a lot of social functions being shut down, etc. resulted in an abnormally low murder rate in 2020.  Or the other way - a lot of people out of work, economic pain finally caught up, people realized they weren't getting rehired in 2021...and got more desperate as the pandemic dragged on.  These are all reasons why data from the last 3 years is inherently going to be a bad source to draw overall trendline examples from.  Assuming anecdotal/talking points is actually pointing to real evidence at all, and not just dogwhistles designed to rile up the base of one political party just before the midterms.
Data Source:  FBI via https://crime-data-explorer.fr.cloud.go … rime-trend

Last edited by DmtDusty (2022-10-29 13:36:12)