AI News, Philadelphia Crime Science - A "data" investigation

Philadelphia Crime Science - A "data" investigation

In today's blogpost, we take a look at the crime statistics in Philadelphia, investigation from a 'data scientist' perspective, so as to speak.

This is my attempt at presenting crime in Philadelphia area from a data visualization perspective.

This dataset contains crime information over a 10-year period from 2006-2016, including location, police districts, type of offense, etc.

Red indicates high crime counts, green indicates low.  Maps for four different categories : thefts, residential burglary, narcotics and aggravated assault.

Thefts and burglaries of various kinds seem to be the largest segments, although vehicle-related thefts are decreasing steadily.

Crime by hour showed a very revealing picture, as seen in image below: (24-hour format)


One particularly expansive national source for neighborhood-level crime data is the National Neighborhood Crime Study (NNCS), which collects street crime data reported to police for the year 2000 from 9,593 nationally representative neighborhoods in 91 large cities.20 Considering NNCS data, Peterson and Krivo found striking racial inequality across neighborhoods in the average rates of violent crime: predominantly African-American neighborhoods (those that consist of more than 70% African-American residents) averaged five times as many violent crimes as predominantly white communities;

either street intersections or segments (two block faces on both sides of a street between two intersections).26 One study reviewed Boston police records from 1980 through 2008 and found that fewer than 3 percent of micro places accounted for more than half of all gun violence incidents.27 When gun violence increases, these hot spots account for most of the increase, and the same occurs when gun violence declines.

meaning that the victims and offenders of violent crime are often members of the same social network, and neighborhood context such as street culture might influence this phenomenon.35 One study found that in Boston, about 85 percent of gunshot injuries occur within a single network of people representing less than 6 percent of the city’s total population.36 Drawing on an array of research on networks, Papachristos argues that “gun violence is transmitted through particular types of risky behaviors (such as engaging in criminal activities) and is related to the ways in which particularly pathogens (e.g.

Numerous studies, for instance, show that neighborhoods with higher poverty rates tend to have higher rates of violent crime.42 Greater overall income inequality within a neighborhood is associated with higher rates of crime, especially violent crime.43 Sampson notes that even though the city of Stockholm has far less violence, segregation, and inequality than the city of Chicago, in both cities a disproportionate number of homicides occur in a very small number of very disadvantaged neighborhoods.44

appears to drive disparities in local crime rates between these neighborhoods.45 As Pattillo-McCoy writes, crime from disadvantaged areas in Chicago often spills over into middle-class, predominantly African-American neighborhoods.46 Moreover, the effects of citywide segregation extend beyond majority-minority neighborhoods: neighborhoods nationwide, regardless of their racial composition, tend to experience higher rates of violent crime when they are located in cities with higher levels of segregation.47

law emerged.49 This alternate law involves witnesses scared to testify, the formation of gangs for protection, and cascades of disputes and violent crime among interwoven communities.50 As Massey writes, “In a niche of violence, respect can only be built and maintained through the strategic use of force.”51 Evidence suggests that a greater propensity for arguments to escalate to lethal violence, combined with easier access to firearms, contributes to higher rates of homicide in the United States.52 As Leovy points out, the absence of law has fostered violent crime in communities throughout history.53 Source: Federal Bureau of Investigation.

hinder efforts to address violent crime.54 Concentrated disadvantage, crime, and imprisonment appear to interact in a continually destabilizing feedback loop.55 In disadvantaged, segregated neighborhoods, residents may also be more likely to be detached from social institutions and disregard the law,56 hampering crime enforcement and prevention.57 Evidence suggests that community policing can improve communities’ relationships with law enforcement and contribute to strategies such as hot-spot policing that seem to reduce violent crime.58

Collective efficacy, defined as social cohesion among neighbors and their willingness to intervene on behalf of the common good, appears to be an important determinant of violent crime in neighborhoods.59 Social cohesion measures ask, for instance, whether residents believe people in their neighborhood can be trusted.60 Across neighborhoods in Chicago and cities worldwide, Sampson and others have found that collective efficacy and violent crime are interrelated: violence can reduce collective efficacy, and collective efficacy can prevent future violent crime.61

for example, by encouraging people to move and stigmatizing a neighborhood.66 In fact, strong evidence indicates that shared perceptions of past disorder (that is, what people thought about a neighborhood years ago) are a better predictor of homicides in neighborhoods than are present levels of physical disorder.67 One study of violent crime in Chicago neighborhoods during the 1990s found that legal cynicism —

Numerous studies show that immigration is strongly associated with lower rates of violent crime.72 One rigorous study of neighborhoods in Los Angeles in the mid-2000s, for instance, found that greater concentrations of immigrants in a neighborhood are related to significant drops in crime.73 Similarly, Sampson, in analyzing data on Chicago neighborhoods, found that, after controlling for other factors, concentrated immigration is directly associated with lower rates of violence.74 One reason for this finding might be that people who immigrate have characteristics that make them less likely to commit crimes —

appears in some circumstances to be related to increases in violent crime.77 Research shows that residential instability might affect violence at least in part by, for instance, reducing community efficacy.78 Violent crime and residential instability appear to be interrelated: one study considering Los Angeles neighborhoods in the mid-1990s estimated that the effect of violent crime on instability was twice as strong as that of instability on crime.79

Multiple studies have found that foreclosures increase violent crime on nearby blocks.80 One study notes that because foreclosures appear to pull crimes indoors, where offenders are less likely to be caught, crimes resulting from foreclosures and subsequent vacant units could be underreported.81 On the other hand, foreclosures might just reshuffle crime at the local level.82

A study of Pittsburgh found that violent crime increased by 19 percent within 250 feet of a newly vacant foreclosed home and that the crime rate increased the longer the property remained vacant.83 In 2016’s Evicted, Desmond notes that Milwaukee neighborhoods in the mid-2000s with high eviction rates had higher violent crime rates the following year after controlling for factors including past crime rates.84 Desmond suggests that eviction affects crime by frustrating the relationship among neighbors and preventing the development of community efficacy that could prevent violence.85

Instead, they found strong evidence indicating that voucher holders tend to move into neighborhoods where crime is already increasing, perhaps seeking more affordable rents.93 Some other studies suggest that associations between increases of voucher holders and increases in crime could be limited to disadvantaged neighborhoods or neighborhoods where households receiving housing assistance are concentrated.94 Mast and Wilson considered this question in Charlotte-Mecklenburg County, North Carolina, from 2000 to 2009, finding that increases in voucher holders were associated with crime increases only in neighborhoods that exceed relatively high thresholds for poverty or concentration of voucher holders.95

Movers ended up in much safer neighborhoods, and parents and adolescent girls experienced significant improvements in health, including lower rates of obesity, linked to reductions in stress.99 In dangerous areas, people may avoid going outside, and a strong relationship exists between perceived neighborhood safety and obesity rates.100 In general, exposure to violence puts youth at significant risk for psychological, social, academic, and physical challenges and also makes them more likely to commit violence themselves.101 Exposure to gun violence can desensitize children, increasing the likelihood that they act violently in the future.102 One study found that children exposed to an incident of violent crime scored much lower on exams a week later.103 Another study focusing on Chicago in the 2000s considered children’s exposure to neighborhood violence over time, finding that, after controlling for differences between students, children living in more violent neighborhoods fall farther behind their peers in school as they grow older and that this effect is similar in size to that of socioeconomic disadvantage.104 At a larger level, Chetty and Hendren find that children who live in neighborhoods with higher crime rates for 20 years experience significant reductions in income as adults.105

Today, as Friedson and Sharkey point out, the recent decline of violent crime offers opportunities for “a virtuous cycle of declining crime and disorder, reinvestment, and greater integration of disadvantaged neighborhoods into the urban social fabric.”109 Taking advantage of these possibilities could reduce disparities and save more people, families, and neighborhoods from the impact of violent crime.

FBI’s Violent Crime Statistics For Every City InAmerica

The views expressed on this page are those of the author, not CBS Local Chicago or our affiliated television and radio stations.

My intention is to share what is essentially a sliver of information about our cities, a handful of numbers that contribute to a much bigger picture.

That includes a 2 percent drop in murders, a 6.7 percent drop in robberies and a .7 percent rise in aggravated assaults across America.

In 2014, America had a violent crime rate of 365.5 per 100,000 residents and a murder rate of 4.5.

Of America’s three biggest cities, Chicago had the highest violent crime rate at 884.26 violent crimes per capita.

Ohio 20.16 Chicago had a murder rate of 15.09, New York had a rate of 3.93 and Los Angeles had a rate of 6.66 (well, 6.65511 if you don’t round to the nearest hundredth).

American Cities With More Than 250,000 People Between 2013 and 2014, there was a 1.3 percent drop in violent crimes and a .4 percent rise in murders in Group I (what I’ve been calling big cities).

100,000 to 249,999 People Between 2013 and 2014, there was a .8 percent drop in violent crimes and a 1 percent rise in murders in Group II (what I’ve been calling medium cities).

10,000 to 99,999 People Between 2013 and 2014, there was a 3.13 percent drop in violent crimes and a 2.57 percent drop in murders in Groups III, IV and V (what I’ve been calling small cities).

Under 10,000 People Between 2013 and 2014, there was a 1.1 percent drop in violent crimes and a 5.8 percent drop in murders in Group VI (what I’ve been calling tiny cities).

With that said, while New York City (which has the largest population total), has the largest violent crime total, Los Angeles (second largest population total), has a smaller violent crime total than Houston and Chicago, despite the fact Houston and Chicago have fewer people.

instead used the violent crime rates per 100,000 people when looking at the numbers of the above cities. Rate per capita takes population into consideration, leveling the playing field between cities with drastically different population numbers.

They have the biggest numbers, yet, considering their populations, have a lower number of violent crime occurrences than the cities with the highest violent crime rates.

Crime rates may be better than totals, but a single murder in a small town yields a drastically different effect on a crime rate than a single murder in a large city.

Recently, the FBI changed their definition of rape to encompass more sex crimes than previously, meaning the number of rapes by the new definition is higher than the number of rapes by the old definition.

It looks like a great deal of states and cities have switched to the new definition, but a fairly large chunk have not.

The change wouldn’t likely drastically change the violent crime total or rate of a city, but it definitely changes the rape total and rate.

They do not represent the whole picture when it comes to a city’s crime, let alone the whole city, or how a city has dealt with its crime over long periods of time.

It takes a deep understanding of not only the statistics behind crime and poverty, but a solid grasp on the history of a place to begin to understand the true character of the place you live.

The Effects of Local Police Surges on Crime and Arrests in New York City

In 2003, under Operation Impact the NYPD began a major change in its deployment practices by implementing the concept of an “impact zone”–a high crime area with specific boundaries that was designated to receive additional police fresh out of the police academy.

Using street-level crime data presented on maps, police crime analysts produced detailed statistical reports and recommended ways of refining the targeted areas.

After discussions among local commanders and headquarters analysts, the Police Commissioner initially selected 24 areas with the highest rates of crime to receive extra police officers [31].

We use census block groups as the unit of analysis because the police designed impact zones using a wide circumference spanning multiple blocks, as well as clusters of public housing projects, that accompany crime markets.

Aggregating the data to census block groups also minimizes NYPD mapping errors of placing crime, arrest, and stop locations at intersections or in the middle of the street when the street address is missing from officer reports.

We also include separate counts for robbery, assault, burglary, weapons, misdemeanor offenses (e.g., criminal mischief, fraud, gambling, loitering, petty theft, and larceny), other felonies (e.g., escape 3 and forgery), drugs (e.g., dangerous drugs), property (e.g., grand larceny, burglary, and burglary tools), and violent felonies (e.g., homicide, rape, robbery, arson, felony assault, and kidnapping).

We do not calculate rates of crime per population because such rates will be distortedly high in business areas of NYC, such as Times Square or Wall Street, which have daytime populations that far exceed their residential population [32].

Under this specification, before and after changes in the monthly counts of crimes and arrests for those areas that become impact zones are compared to areas in the same police precincts that do not become impact zones.

where Yit ~ Poisson (λit) is the count of crimes or arrests in census block group i (n = 8,091) in month-year (= 2004–01,…, 2012–12) t of precinct p.

The DD estimates will understate the effect of impact zones if enforcement in those areas creates spillovers, and overstate the effect of impact zones if they generate displacement to nearby blocks.

If census block groups get assigned impact zones because of recent spikes in crimes or arrests, models (1) and (2) could overstate the effect of impact zones because of mean reversion.

(S) when police officers checked only the following indicators: (1) furtive movements, (2) fits descriptions, (3) carrying objects in plain view, (4) suspicious bulge, or (5) evasive actions [34].

The History of Crime Fighting in Black America, Khalil Gibran Muhammad

October 27, 2016 Framed by the polarizing 2016 presidential race, advocates of law and order stand on one-side against competing claims for systemic police ...

Quantifying Forgiveness: MLTalks with Julia Angwin and Joi Ito

Algorithms are increasingly arbiters of forgiveness. They are used to decide who will get let out of jail, the amount that people pay in auto insurance premiums, ...

Justice Involved Young Adults

A panel discussion of developmentally appropriate responses to justice-involved young adults.

Wheel of Justice - Torrence White

Help JSO get these wanted suspects off our streets.

Toronto and East York Community Council - May 2, 2018 - Part 2 of 2

Toronto and East York Community Council, meeting 32, May 2, 2018 - Part 2 of 2 Agenda and background materials: ...

Board of County Commissioners Douglas County Nebraska, April 24, 2018

Board of County Commissioners Douglas County Nebraska, April 24, 2018.

Auburn Coach Wife Kristi Malzahn Agrees with Match & eHarmony: Men are Jerks

My advice is this: Settle! That's right. Don't worry about passion or intense connection. Don't nix a guy based on his annoying habit of yelling "Bravo!" in movie ...

CBS News Live Stream HD

CBS News. Always On. CBSN is CBS News' 24/7 digital streaming news service. It's always on, always free, making CBS News' original, high-quality reporting ...