Wednesday, April 23, 2008

Mapping fear of crime

Last month, I posted a graph and a blog about perceptions of safety. It was 13 years data from surveys of people who had contact with the police after being involved in traffic crashes. As part of a large ongoing survey process, we ask a few thousand people every year "How safe and secure do you feel in the neighborhood where you live?"

I used the data on overall perceptions of safety across time as a benchmark of a desired outcome in our proposed outcome-based budget presented to Mayor Chris Beutler last week. He asked an interesting question: "Can you break this down by area?" The answer was "Yes, we have these data broken out by five police team areas."

Back at the office, though, his question got me thinking about doing something at a finer level or granularity. A conversation with Jon Carlson, the Stronger Safer Neighborhoods coordinator, and Clair Lindquist, our Information Technology manager cooked up the idea. The result is this map (click to enlarge):

These are census block groups, assigned their color based on the average score of all the respondents within each are to the "How safe?" question: darkest red is least safe, darkest blue is most safe. Although there are some methodological issues I could bore you with (chiefly, the survey isn't random), I nonetheless think this is a pretty good indicator of where people feel safest in the neighborhood, and where they feel less safe.

The map is based on 2,517 responses for 2007 that could be geocoded to a specific location. The average scores within each census block group ranged from a high of 5 ("always safe and secure") to a low of 2.67 (between "usually not" and "sometimes safe and secure"). The overall mean score is pretty high, at 4.08, and keep in mind that of the total 1118 of the respondents were victims of crime within the recent past. Here's the statistics on the distribution:

I'm not aware of anyone who has mapped "fear of crime" like this before, but there may be other examples I don't know about. It's an intriguing use of this unique data set we collect. In the broadest sense, the map shows the same pattern that most crime displays in Lincoln: the people who responded to the survey are accurately reflecting the relative risk of crime, with a few exceptions. The area where perception of safety is lowest essentially overlays the city limits of Lincoln as it existed 60 years ago, further emphasizing the importance of taking care of the core.


Prairie Dog said...

Nice map. Can you show how many responses were obtained from each Team area after their taffic accidents?

Anonymous said...

That first map doesn't seem to be clickable for a larger version. I didn't look at the source code yet to try and figger out why that might be.

-JS- said...

Are the geocoded locations where the incident occurred or where the person physically resides?

Tom Casady said...


These are all respondents, not just drivers in crashes (victims, drivers, cited/arrested) and the rough numbers by Team are:

NW 424
SW 481
NE 527
SE 669
C 229

The remainder are outside the city, with a smattering of unmatched records. I'm going to working on the geocoding a little bit to improve the hit rate slightly, so that number is a moving target for now.


Thanks, problem with my .html code. Should be fixed now.


This is where the person resides, not where the incident occurred. Critical difference, given the question asked.

Anonymous said...

Here is a Census Poverty map of 2000. Quite a corelation on safe perception and the poverty rate even though there is a 8 year gap.

Anonymous said...

Why are thar so many 05000's on 4-23-2008
Must have been national macho man day?

Karin Dalziel said...

Is there a way to overlay this with actual crime statistics? Are the places where people feel the most safe actually the most safe?

I live in the "most unsafe" area from the looks of it. Funny thing- I feel pretty safe. But I think my view is tempered by the fact that I've lived in much bigger cities. I love when people who have never seen an inner city call my part of town the "ghetto." :)

Anonymous said...

I live in a relatively boring RD (105 I think), but even if were is surrounded by an army of bodyguards, one should never feel 100% safe, just 99% or thereabouts. "Condition White" is never, ever a smart way to be.

Karin is right, in that when you've lived in dicier cities, Lincoln seems relatively safe. I lived in one where a neighbor got all four of his wheels and tires stolen from his specially adapted minivan. He called the police, and since no one was hurt, they couldn't even dispatch a unit to take a report. Being in an unpowered wheelchair, he found it quite difficult to go to the closest precinct - 4 miles away - to fill out a report, so that he could submit an insurance claim. He sure couldn't drive a van with no wheels! He did get a ride from the neighbor, fortunately.

Yeah, that was a great city. It wasn't that the police didn't care, it was just that they had a lot more violent crime to deal with, and it was like trying to drink water out of a fire hose. Armed robberies galore, bank robberies, carjackings, and a whole host of things that are quite rare here.

Here, you call in to say your neighbor's music is too loud, and a unit probably shows up. There, you call and say you're a victim of a property felony, and no unit is dispatched.

Anonymous said...

I heard a Lincoln cop car get sent to an AM radio station around 3 am one morning because the station went off of the air and the caller was afraid something had happened to them.

There's no doubt at least one but probably two cars will get sent to the loud music call.

It's a pain to answer calls like that but if you take care of the small stuff the bigger stuff doesn't seem to grow as quickly. That's one reason Lincoln is so safe. I actually don't think there is any part of Lincoln that should have a mostly unsafe label.

Tom Casady said...


Yes, it could be overlaid, and would show that these perceptions of safety pretty accurately track the areas where crime occurs with the greatest frequency. People expressed the highest level of perceived safety in the safest places.


Keep in mind, though, that the mean score is still over 4--in between "Usually safe and secure" and "Always safe and secure." The use of a dichromatic blue-to-red symbolology on this map may imply that the red areas are high fear, when in fact none of the census block groups was below the median score of 2.5, and only two of 186 were below 3.