Back in July, the city opened the Micro Community Policing Plans website offering citizens a far more detailed look than was available previously into how criminal activities were distributed across neighborhoods and what specific crimes were being committed. This website breaks crimes down into 8 major categories (4 categories of crimes against persons and 4 of crimes against property) and 68 subcategories which include, for example, 8 different types of theft. The data goes back to 2008, and is downloadable as an Excel spreadsheet for those wishing to perform their own analysis. For those not so inclined, tools provided by Tableau are available to plot the data in various ways right there on the MCPP website.
The crime data lists the number of incidents of each crime type committed in each of 62 neighborhoods in the city. While these incident counts can be plotted in various ways, one thing that cannot be plotted is the crime rate for each neighborhood, i.e. the number of incidents per, say, 1000 residents. This makes it difficult to assess whether some neighborhoods are getting “more than their fair share” of crimes. To get a better insight into crime rates for the neighborhoods, I downloaded the data and computed rates using population information from the city Planning Department, which is derived from the 2010 US Census. In this article, I’ll discuss how Wallingford stacks up against some of the other neighborhoods, and in a future article, I’ll talk about Wallingford crime trends.
While I’m most certainly a numbers guy, I realize that not every reader is, so I’ll try not to get into too many details of the analysis. However, I have to say a word about the population data. Neighborhoods in Seattle are not particularly well defined, as we all know. While the MCPP data breaks the city into 62 distinct neighborhoods, the city planning website breaks us down into just 53. In some cases there is simple consolidation taking place. (MCPP lists crimes for a North and South Ballard, while the population data gives simply “Ballard.”) Other cases are much more unclear. (The population data lists a “Laurelhurst/Sand Point” while the MCPP data has no Laurelhurst, but there is a Sand Point. So are these the same areas?) Given these and similar possible pitfalls, I limited my analysis to 26 neighborhoods that are listed unambiguously in both the MCPP and population data as well as being somewhat clearly defined areas (at least to me). *
Crimes by Category
The two sets of barplots which, hopefully, appear somewhere in the vicinity of this text, show crime rates (incidents per 1000 residents) for 2015, the most recent complete year of data for our neighborhood and some nearby neighborhoods. For comparison, I’ve added the rate for the city overall, and thrown in the rate for Columbia City to add a little variety to the otherwise shipping-canal centric selection of neighborhoods. (Green Lake was one of those neighborhoods with problematic population data, so that’s not shown.) For the record, the population data I used lists Wallingford as having 16,014 residents. Columbia City, Fremont and the University District have populations similar to ours (16,883, 15,626, and 19,051, respectively) while Queen Anne has over twice as many residents (35,458). Seattle, as listed in these figures from the 2010 census, had 608,660 people.
Many people may wonder what makes burglary, theft, and robbery different. (I did!) Burglary is “the unlawful entry into a building or other structure with the intent to commit a felony or a theft,” while theft is unlawfully taking something from a person. Robbery is theft “under confrontational circumstances” or with “threat of force or violence.” More definitions are here.
When it comes to property crimes, the U District is the unfortunate winner among our near neighbors. As an employee of the UW who receives email notifications of every criminal incident reported to the campus police, I can attest that the U is a great place to have your phone stolen. Citywide, of the 26 neighborhoods I was able to analyze, the burglary rate ranged from 69 incidents per thousand residents (in Ballard) to as low as 1.5 incidents (in the Mt. Baker/North Rainier neighborhood). Theft ranged from 750 in the Downtown Commercial area to 2.7 (in Mt. Baker/North Rainier again). Over all four categories, Wallingford ranks pretty much in the middle of the pack and close to the city average. Still, when you consider that I’m plotting incidents per thousand residents, that’s about 100 cars a year disappearing off the streets of Wallingford last year and almost 500 thefts!
Curiously, crimes against persons tell a bit of a different story with our neighborhood ranking much lower in all four categories than our near neighbors and well below city averages. So while we can feel secure in our persons, we better watch out for our stuff.
Crimes by Subcategory
As I mentioned, SPD identifies 68 subcategories of crime, and I won’t delve into all of them. More importantly, rates have less meaning here because numbers of incidents in each subcategory can be fairly small. To ameliorate this problem somewhat, I averaged the number of incidents over the 8 full years of data (2008 to 2015) for each of the 26 neighborhoods. So the plots below represent the average annual incidence over that time period.
Some subcategories have an alarmingly large number of incidents and strike a chord among residents because they seem part of an endless drumbeat of more minor crimes that, seemingly, can never be abated. I speak here of bicycle theft and car prowls. As a bike commuter, I pay particular attention to incidents of bike theft, so the plot of bike theft rates is kind of alarming to me. One would expect the U District to have a high rate simply because there are so many bicycles there, but Wallingford’s rate isn’t too small, either! With regard to car prowls, we all know that Seattle is doing a terrible job handling that problem, and the numbers certainly reflect that.
Other subcategories enumerate serious crimes that could happen to any one of us. For this reason, I looked at home invasion robberies and robberies of persons on the street. SPD splits each of these out depending upon whether the crime was committed using bodily force, a gun or some other weapon. I added these together figuring that most of us don’t want to get robbed no matter what the weapon. There’s not much to say here other than that Wallingford has an enviably low rate for these crimes compared to our nearby neighborhoods.
* Those 26 neighborhoods are: Ballard, Belltown, Capitol Hill, Columbia City, Downtown Commercial, First Hill, Fremont, Georgetown, Greenwood/Phinney Ridge, Pioneer Square/International District, Judkins Park, Madison Park, Madrona/Leschi, Magnolia, Miller Park, Montlake/Portage Bay, Mt. Baker/North Rainier, North Beacon Hill/Jefferson Park, North Capitol Hill, North Delridge, Queen Anne, Rainier Beach, South Park, University District, Seattle overall, Wallingford. Well, okay, Seattle’s not a neighborhood.
Thanks for this excellent analysis!
I’m curious whether the “Wallingford” boundary includes Tangletown, or whether it stops at 50th St. Do you know? Sometimes Tangletown is considered part of Green Lake.
As defined by the MCPP, Wallingford runs from the ship canal north to about 60th and from I-5 west to about Stone Way. More info here: http://www.seattle.gov/seattle-police-department/mcpp/mcpp-location-map. Unfortunately, I can’t find any descriptions of the areas for the population data that I used.
If go to the link at the beginning of the article, you’d find the MCPP Location Map tab. Click that, and zoom into the map. Click on a location and it’d tell you the neighborhood name. By doing that you’d find Tangletown to be Wallingford. The immediate north of that would be the Roosevelt/Ravenna neighborhood. There is no Green Lake neighborhood in this database.
hmm a stolen car is same as a stolen bike, purse, phome, lunch, set of mittens?
Car theft gets a category all to itself, but otherwise, I think you’re right. I’ll bet most folks would just walk away from a stolen lunch, though, rather than report it. May depend upon the lunch!
I still do not see car theft in any graph
VEH = vehicular theft, I’m assuming
Yep, I think it is vehicle theft. What an amazing graph, and marvelous work. Thank you, Jack for this wonderful article, and easy to understand maps.
The caption under the first graph says “…. VEH is vehicle theft.”
Hey Jack, I’d love to chat with you about what you’ve done here with crime stats for an article I want to do for MyNorthwest / KIRO Radio. Please email me at [email protected]. -Dyer Oxley