Comments on: A geographic analysis of local crime data

By: Seth

Seth — Wed, 03 Nov 2010 03:09:50 +0000

The property crime data in my hometown of Bellport is a prime reason to lock my door at night.

http://www.homefacts.com/crime/New-York/Suffolk-County/Bellport.html

By: Ron

Ron — Sun, 09 Sep 2007 10:57:33 +0000

I like to go for a long walk in my small town (pop 7,000) every night. During the summer nights there tend to be many “characters” wandering around and gathering socially in certain locations I try to avoid. In the winter, which can be very cold here (northern New Hampshire), these same characters are nowhere to be seen and most of the people who are out seem to be those who enjoy the outdoors and keeping healthy – which usually are not these same “characters” I mentioned earlier.

My theory is that many muggers, drug users, and the like are not about to be inconvenienced by the cold of winter so they stay inside. Anyone who has the strength of character to enjoy being outdoors in the winter weather (which really is very beautiful) probably also has the strength of character to hold down careers and work on self improvement. In effect, winter is a great filter of sorts that separates types of people.

By: rajsingh.org blog » Check out John Udell

rajsingh.org blog » Check out John Udell — Sat, 11 Aug 2007 06:06:18 +0000

[...] A geographic analysis of local crime data [...]

By: The police station effect « Jon Udell

The police station effect « Jon Udell — Thu, 02 Aug 2007 18:56:58 +0000

[...] Filed under: Uncategorized — Jon Udell @ 1:56 pm Thanks to some really great comments on yesterday’s item I’ve taken another pass through the spreadsheet I got from the police department1. It looks [...]

By: duncan

duncan — Thu, 02 Aug 2007 08:10:24 +0000

Is it possible that there are fewer muggings during winter because there are fewer people on the streets to mug (i.e they’re all indoors so there is less trade for the muggers)?

By: Jon Udell

Jon Udell — Wed, 01 Aug 2007 21:44:05 +0000

“Is crime up in the area around the new station?”

Heh. Great question, I’ll look into it.

“Another possible explanation for the dip in winter recorded crime: the police don’t like getting out of their cars.”

:-)

“the full dataset available? or are you not allowed to publish that data?”

There’s a possible privacy concern about some of it, so I’m going slowly. But the data behind this analysis is just dates and locations, so I can put that up, and will.

“the station was the place of the crime report and there was often no specific address associated with the crime incident itself”

In this case the addresses are all real street addresses…

“Your analysis may be reflecting a similar quality issue in the underlying data.”

…but, great point. There are all kinds of questions that can and should be asked about the nature and quality of officially-reported data. It’s only once you start to explore the data, in these ways, that you can even begin to ask those questions.

By: David French

David French — Wed, 01 Aug 2007 21:01:13 +0000

An early effort at producing crime hotspot maps in NZ flagged up police stations as being crime centres because the station was the place of the crime report and there was often no specific address associated with the crime incident itself. Your analysis may be reflecting a similar quality issue in the underlying data.

By: Nathan

Nathan — Wed, 01 Aug 2007 19:35:41 +0000

maybe i’m missing it, but is the full dataset available? or are you not allowed to publish that data?

i remember my adviser doing something with the lapd and he had to sign some privacy contract. they then handed him the data on a memory stick that required a key.

By: Hamish Harvey

Hamish Harvey — Wed, 01 Aug 2007 17:14:10 +0000

Hi Jon,

You want, using this data, to answer the question “Is there more crime in the past few years?” But can you? What you have is data on _recorded_ crime, so you can surely only answer questions on that.

So you may be able to find trends in the data (are they statistically significant?) on recorded crime, but how do you move from that to actual crime? How many laws have creating new crimes, or guidelines which encourage the police to resource and record differently?

Another possible explanation for the dip in winter recorded crime: the police don’t like getting out of their cars.

Several causes may combine to generate a trend or pattern, of course. E.g. there may be less mugging _and_ the police stay in their cars more.

Cheers,
Hamish

By: Chris Anderson

Chris Anderson — Wed, 01 Aug 2007 17:09:54 +0000

My dad (a lawyer from Austin, TX) always regards police stations as the area of town with the most crime. “They don’t police their own backyard,” he says, “and this is where you end up when they let you out of the drunk tank.”

Maybe there’s some truth to that. Is crime up in the area around the new station?