Pittsburgh interactive crime map (Manchester: crime rates, homes, neighborhoods)

[nuwaver88]

Neighborhood crime: 2009-2010 statistics

Click a neighborhood for an extensive breakdown of all crimes. This shoud be helpful for newcomers, but does not tell whole story. For example, decent amount of crime on the South Side, but still a great neighborhood to live according to many.

[BrianTH]

Would be a lot more useful if they included crime rates (e.g., per 1000 residents or something).

[nuwaver88]

I agree. It hardly gives an idea of quality of life for a particular neighborhood. One may get the impression that Manchester is a more desirable place to live than Squirrel Hill. Not necessarily so.

[Aqua Teen Carl]

A running total and per capita stats would be quite useful. Although, this is a pretty helpful if you're looking for neighborhoods in the same area of similar size.

[Hopes]

The murders reflected on the map validate my long-standing opinion of neighborhoods based on safety. I'm not picky about other crimes, but the neighborhoods with murders are the neighborhoods that have always required varying degrees of caution compared to the other neighborhoods that have no murders. Granted, there are a few murder statistics that are irrelevant, like the 3 muders is Stanton Heights were a one time thing and don't really accurately reflect the safety of Stanton Heights.

[Debbie1125]

Originally Posted by Hopes

The murders reflected on the map validate my long-standing opinion of neighborhoods based on safety. I'm not picky about other crimes, but the neighborhoods with murders are the neighborhoods that have always required varying degrees of caution compared to the other neighborhoods that have no murders. Granted, there are a few murder statistics that are irrelevant, like the 3 muders is Stanton Heights were a one time thing and don't really accurately reflect the safety of Stanton Heights.

With cities you are always going to have neighborhoods with what I call less serious crimes such as burglarly and car theft. However if I see an area with a substantial amount of murders, then it gets crossed off my list of places to live. Also if there are sexual assaults that are random (such as a stranger breaking into homes and committing rape), I will eliminate those neighborhoods as well.

Is Stanton Heights where the 3 police officers were shot?

[Hopes]

Originally Posted by Debbie1125

With cities you are always going to have neighborhoods with what I call less serious crimes such as burglarly and car theft. However if I see an area with a substantial amount of murders, then it gets crossed off my list of places to live. Also if there are sexual assaults that are random (such as a stranger breaking into homes and committing rape), I will eliminate those neighborhoods as well.

Murder is my main criteria. Sexual assault isn't a criteria for me. Many are date rapes. If there is a serial rapist, they eventually get caught. There's a certain degree of street smarts necessary. The murders are more a reflection of area violence. Rapes tend to be a reflection of where students live. Some of our most dangerous neighborhoods have low numbers for rape.

Originally Posted by Debbie1125

Is Stanton Heights where the 3 police officers were shot?

Yes. That's why those three murders aren't an accurate reflection of the safety of Stanton Heights. I used it as an example. There are a few other neighborhoods with a murder here and there that were one-time-incidents and not reprentative of the safety of the area. But for the most part, the murders on the map are a fairly accurate reflection of neighborhood safety. I wouldn't rule an area out for one murder because it could have been domestic murder. I'd research the actual details of a murder before I ruled out an area that showed one.

[BrianTH]

It would be great to see someone filter this data for stranger crimes only. A lot of violent crimes are committed by family members, criminal co-conspirators, and so on, and there isn't a lot of locational risk associated with those.

Absent something like that, you could make a good case for looking specifically at armed robberies, because those have relatively high stranger ratios. Unfortunately, you would then run into even more sample size problems.

Edit: Oh, and you would want to distinguish neighborhoods with substantial commercials districts from neighborhoods without, since the latter tend to have far fewer robberies.

[Hopes]

Originally Posted by BrianTH

It would be great to see someone filter this data for stranger crimes only. A lot of violent crimes are committed by family members, criminal co-conspirators, and so on, and there isn't a lot of locational risk associated with those.

Maybe Uptown will do it. The other maps have links to the news articles. It's very possible. Someone just needs to be inspired to do it.

I think there is locational-risk with co-conspirator murders. There is no doubt that it's more dangerous to live in neighborhoods with high gang murders.

[BrianTH]

Originally Posted by Hopes

I think there is locational-risk with co-conspirator murders. There is no doubt that it's more dangerous to live in neighborhoods with high gang murders.

It depends on what you mean by that. The ratio of innocent bystanders killed in gang wars compared to actual gang associates is really pretty low (although cases like that tend to get a lot of press attention). So if you were going to calculate a direct locational risk to innocent strangers from that factor, you would have to do it at a very heavy discount.

On the other hand, gang murders may correlate with other crimes that involve innocent strangers at higher rates. But even if that is true, if at all possible you should be using direct data on those crimes, and not gang murders as a proxy.

I might note that part of why I think this is important is to help people get a better sense of not just violent crime risks relative to each other, but also a better sense of violent crime risks relative to other locational risks.

For example, as I have pointed out before, a lot more innocent people are killed in car accidents than by violent crimes, and that is particularly true of children. So people should be taking that form of locational risk into serious account, but that tends not to be the thinking of most people. Of course it is likely many people can both avoid excessive violent crime risk and excessive car accident risk with a well-chosen neighborhood, and so in terms of eliminating the absolute worst neighborhoods for violent crime, this may not matter so much. But when you start to get to the more intermediate cases, there I think a better sense of the actual locational risk to innocent strangers would be helpful.