There's a couple things that need to be considered when discussing "has crime gone up or down", as people have been leveraging a "creative use of data" to convey a particular message.
A) it largely operates off of arrest and conviction data. Which as I'm sure we all know, can be a challenge when involving a cohort of people for whom there is no official paper trail.
B) They confined it to a particular time window, which makes it easier to make misleading suggestions.
For Example:
If over a 3 year period -
If Dave increased his daily beer intake from 5 to 10 over a period of 2 years, and then in the final year, scaled it back down to 8.
It'd be, albeit statistically true, but a little misleading to say "Dave's drinking has actually gone down 20% in the past year" as a means of implying "people who are concerned about Dave's drinking are just overreacting"
C) They used to consume 2 data sources to build their reports, now they've consolidated into only one, and it happens to be the one that a large number of LEO agencies aren't reliably reporting to.
FBI crime reports (that indicate a drop in crime) coincide with them switching to using the solitary data source called the NIBRS...a source for which nearly a third of the nation's law enforcement entities aren't reporting data to.
Nearly one-third of law enforcement agencies are missing from the FBI's 2022 crime statistics. Use our tables to check on your state and local agencies.
www.themarshallproject.org
as opposed to before, where they used the NIBRS in conjunction with the NCVS to get a bigger picture.
So it stands to reason that it would show an overall decrease if they keep the source that shows the numbers going down (and that have heavily varying rates of participation), and put a
3 year pause on the one that showed certain crimes going up.