Useful 2019 UCR and MSR purchase summary

Kind of surprised at the large number of firearms that aren't identified by type. Do you think those are mostly handguns?

hso said:

Unknowable. Conjecture isn't a substitute for data.

Click to expand...

True. Also, incomplete data isn't very useful.

Jeff White said:

You have to be careful drawing conclusions from the UCR. Reporting is not mandatory. Not all LE agencies send in their data. There are no checks on what's reported and cities have been caught fudging their data to make the crime rate look better.

Click to expand...

Graphs like these are a dime a dozen and do nothing to protect the 2nd amendment are further the cause. They are boring. What does make a movement are folks that make a stand and speak out. And so many of those are shut down across the board. Please no more graphs.

bearcreek said:

Kind of surprised at the large number of firearms that aren't identified by type. Do you think those are mostly handguns?

Click to expand...

NIBRS was aimed to address many of the issues with reporting with UCR. From minor problems such as incomplete reporting of what kind of weapon was used to major issues like determining the beginning and end of what classifies as a reportable incident. Unfortunately, it has taken decades to implement and some places are still dragging.

bearcreek said:

Also, incomplete data isn't very useful.

Click to expand...

Not so. There are almost no "complete" data sets outside of textbook examples; in the real world the vast majority of analysis is done on incomplete (sometimes severely so) data. This introduces uncertainty; it doesn't make the data useless.

bearcreek said:

Kind of surprised at the large number of firearms that aren't identified by type.

Click to expand...

The trick to incomplete data is to account for the bias that may be present. Is there a reason to think that the "type not identified" category is a product of selection bias, and is all handguns? All rifles? All blunderbusses?

Is it reasonable to think that all the keyboard gnomes who produced that data imposed a consistent bias on the information? I think probably not. More likely the police report said, "Beretta M9" or "Glock 17", and the gnome checked the "type not specified" box, because it's true.

Contrast this to a classic case of widespread consistent bias; ask 1000 people to rank themselves, and you'll find Lake Wobegon, "all the women are strong, all the men are good-looking, and all the children are above average."

edwardware said:

Not so. There are almost no "complete" data sets outside of textbook examples; in the real world the vast majority of analysis is done on incomplete (sometimes severely so) data. This introduces uncertainty; it doesn't make the data useless.

Click to expand...

Didn't say it was "useless", I said it wasn't very useful, largely for the reasons pointed out in post 4 and 7.

edwardware said:

The trick to incomplete data is to account for the bias that may be present. Is there a reason to think that the "type not identified" category is a product of selection bias, and is all handguns? All rifles? All blunderbusses?

Is it reasonable to think that all the keyboard gnomes who produced that data imposed a consistent bias on the information? I think probably not. More likely the police report said, "Beretta M9" or "Glock 17", and the gnome checked the "type not specified" box, because it's true.

Click to expand...

Could be. As HSO pointed out in post 3 though, "conjecture isn't a substitute for data".

Ernie Bass said:

Graphs like these are a dime a dozen and do nothing to protect the 2nd amendment are further the cause. They are boring. What does make a movement are folks that make a stand and speak out. And so many of those are shut down across the board. Please no more graphs.

Click to expand...

This. They don’t care about graphs. They know the stats as well. That’s why they use fear, emotion, and misinformation to get what they want, because they know the facts don’t back them up.