Visualizations On Crime In Los Angeles

UCLA DataRes
May 29 · 7 min read

By Boyang Yu and Radhika Ahuja

Motivation

Every city has an interesting relationship with crime, and so does Los Angeles with its list of street gangs and instances of under-reporting of crimes as “minor offenses.” Despite this, there is a lot of data to understand the underpinnings of criminal trends throughout the years to help us better understand and visualize crime in the city. We will walk through some general trends and suggest a hypothesis for why they each emerge.

The Dataset

We worked with a dataset from Kaggle that describes arrests made for different criminal activities in Los Angeles, based on time, age, gender, ethnicity, and location. We cleaned the dataset to add any missing values and then selected only the instances that occurred in 2019 for our analysis. It is a robust dataset for us to work with, as an official authority provides it. The only things that add bias to the data are:

The Top 5 Crimes in Los Angeles in 2019

We created a word cloud to give you a broad view of the most common crimes that happened in Los Angeles in the past few months. A word cloud is a collection of different words used in a document (for us, charge group description) with the size of the word corresponding to its frequency, i.e., the bigger the word, the more likely that crime happened in Los Angeles during January 1st, 2019 to May 4th, 2019.

Fig 1: Word cloud representing most common crimes in Los Angeles

Based on the word cloud, we can deduce the top 5 crimes (in order) in LA as:

  1. Drugs
  2. Aggravated assaults
  3. Drive under influence (DUI)
  4. Larceny
  5. Traffic

According to a 2017 study from PolicyLink and the USC Program for Environment and Regional Equity, with support from the Weingart Foundation, Los Angeles comes seventh in income inequality out of the largest of 150 metro regions. This fact, coupled with our proximity to the southern border and a healthcare system that is not fully equipped to deal and educate the general public about addiction perhaps contributes to drugs and DUIs being in the top 5.

Larcenies and assaults are also probably common due to similar reasons. We see a concentration of wealth in some regions of Los Angeles (like Bel Air and Beverly Hills) and a vast wealth gap. A more significant wealth gap in a high-earning economy means there is a higher number of people who fall in the edges of the income distribution. With a large low-income population who have an incentive to steal and a high-income population who are not just wealthy but considerably affluent, there would be a higher number of people who would commit larceny. Assaults may happen for a similar reason as cities with a considerable wealth gap also often have larger homeless populations, and Los Angeles is no different.

Los Angeles is a populous city and a buzzing economy and without the best public transport to support it. As inhabitants are forced to turn towards cars, Los Angeles sees around upwards of 7 million vehicles registered in her county. Traffic making the top 5 is no longer that surprising, and with the sheer volume of drivers on the road at any time, DUIs will tend to increase as well.


Crime By Age

Criminal activities center around the ages between 20 and 40 and peak between the years of 35 and 35. While this is to be expected, we still see a considerable amount of crime occurring even after the age of 40, right up to the age of 65 and a relatively low number between the ages of 15 and 20, perhaps saying something about the teenage criminal stereotype.

Fig 2: Histogram representing the number of instances of crime by Age

To better analyze the relationship between crime and age, we factor the ages into different groups and examine which crimes are more common for specific age groups. Here, we used the following age groups

  • teenager (10–18)
  • young adult (19–35)
  • middle age (36–55)
  • elderly (56 above)
Fig 3: Crime committed by different age groups according to the type of crime

We can still find that people between the ages of 19 and 35 commit much more crimes than other age groups. Miscellaneous offenses include public intoxication, disorderly conduct, etc. are the most common for all the age groups. However, compared to other age groups, young adults and middle-aged people are most likely to drive under the influence (of alcohol or other drugs) and violate narcotic drug laws. Middle age and older adults are more likely to be involved in drunkenness and violate the liquor laws while teenagers and young adults do not.

Crime By Gender

If we visualize the frequency of crime based on gender, we find that men commit more crimes than women do.

Fig 4: Male vs. Female instances of crime

We can then examine the types of crimes by gender.

Fig 5: Male vs Female instances of crime by category

Since in general crime is associated with men, it is standard for women to have a smaller proportion of the crime share as opposed to men. However, we can see three notable disparities:

  1. Most traffic violations are committed by men
  2. Most weapons are carried by men
  3. Most prostitution is committed by women.

While #2 and #3 may not be very surprising, #1 points against the stereotype that women are not good drivers.

Crime By Day

Another pertinent question about crime is about its frequency given a certain period. When is a crime most likely to happen?

Fig 6: Arrests made for every day of the year until May

We can see above that there is a noticeable peak in April and the number of instances inch higher than the months earlier in the year (the sharp drop in May represents the current date, with no data on recorded cases available yet). The trend of crime rate increasing as summer approaches is quite common. People tend to go out more in the summer, and there is an influx of tourists and people who travel during the summer, both resulting in higher instances of crime.

Crime By Area

We can also visualize the frequency of crime according to the different areas of Los Angeles. We see very high instances of offenses in Central LA and a significantly lower number in West LA. It might have something to do with income distribution in the areas.

Fig 7: Criminal Instances by Areas in Los Angeles

Mapping Crime

We also map the crime. In this section, we will just focus on the top crimes “Narcotic Drug Laws violation”, “aggravated assaults”, and “drive under influence (DUI)”.

Fig 8: Geoplot of Narcotic Drug Laws violation in Los Angeles

From this geo plot, Narcotic Drug Law violations are more likely to happen in Downtown Los Angeles.

Fig 9: Geoplot of DUI violations in Los Angeles

Drive under influence (DUI) violations are most likely to happen in USC, downtown LA, and southern LA.

Fig 10: Aggravated Assaults in Los Angeles

Aggravated Assault violations are more likely to happen in downtown LA and MacArthur Park.

It is reasonable to find that most crimes happen in downtown Los Angeles, which is why the downtown area has been recognized as one of the most dangerous places in Los Angeles for a long time. College students are more likely to get involved in driving under the influence (DUI) instances as that is a tendency supported by a college lifestyle of partying and drinking. MacArthur Park (west of downtown Los Angeles), once a gem of the city’s park system, has changed a lot since the early 1970s. Nowadays, it is one of the city’s most densely populated areas and consists mostly of low-income families. With poverty comes crime. Thus, it is also reasonable to find that aggravated assault is more likely to happen in both downtown LA and MacArthur Park areas.


Conclusion

For the most part, crime in Los Angeles tends to be related to substance abuse (drugs, alcohol), traffic (general violations and DUIs), assault, and larceny. Factors that influence this crime are typical to urban metropolises with a large population and an enormous wealth gap. An element unique to Los Angeles is its proximity to the southern border, and criminal activities tend to increase during the summer months.

In general, the issue of crime can be better tackled in LA by having a better public transportation system and having a better healthcare system that provides education and remedial care.

UCLA DataRes

Written by

UCLA’s first Data Science Club. datares.github.io/, facebook.com/ucladatares, linkedin.com/company/ucla-datares

Welcome to a place where words matter. On Medium, smart voices and original ideas take center stage - with no ads in sight. Watch
Follow all the topics you care about, and we’ll deliver the best stories for you to your homepage and inbox. Explore
Get unlimited access to the best stories on Medium — and support writers while you’re at it. Just $5/month. Upgrade