Selecting the Sources

To investigate how crime patterns across Los Angeles changed between 2021 and 2025, we relied primarily on publicly available datasets from municipal open data portals and law enforcement agencies, particularly the Crime Data from 2020 to Present dataset released by the Los Angeles Police Department (LAPD) through the City of Los Angeles Open Data Portal.

These datasets contain detailed records of reported incidents, including the offense type, date and time, geographic location, victim demographics (age, sex, race, ethnicity), and indicators of crime severity, including weapons involved. Because each incident is geocoded and tied to one of the city’s 21 LAPD patrol divisions, the dataset allows us to analyze both demographic patterns of victimization and the spatial distribution of crime across Los Angeles during the post-COVID period.

Using official datasets ensured that our analysis was based on consistently collected records of reported crime across L.A. This structure also enabled us to compare changes in various crime factors such as crime type, severity, victim demographic, and location between 2021 and 2025, simply following the COVID-19 pandemic.

At the same time, we felt it was vital to approach the dataset critically and consider potential gaps or biases. Not all crimes are reported, and differences in policing practices, reporting systems, community trust in law enforcement, or mere data collection methods can influence how crime appears in official records. Being conscious of these limitations allowed us to interpret our data more critically and gain a fuller picture of crime in Los Angeles. More information about the limitations of our data can be found on our Data Critique page.

In addition to the LAPD dataset, we consulted peer-reviewed research and credible policy and news sources on crime trends, policing practices, and structural inequality in urban environments, all factors that may influence crime rates. These sources were vital, as they contextualized the quantitative data and helped us understand the broader social dynamics affecting different areas/communities in Los Angeles. More information regarding our sources can be found on our Annotated Bibliography page.

Processing the Data

The LAPD crime dataset was generally well organized, though some minor cleaning was required before analysis. Using tools such as Excel and OpenRefine, we reviewed the data for duplicates, missing values, and inconsistent category labels. We also standardized variables such as crime types and date formats to ensure accurate analysis.

To create visualizations, we used tools including Tableau, R, and Python libraries such as Matplotlib and NumPy. Each team member chose to utilize different visualization tools based on their familiarity with the language and platform. Nonetheless, we all created our visualizations with careful attention to the type of data presented (qualitative or quantitative), the number of variables chosen to display, and how each graphic supported our overall narrative.

In addition to our visualizations, we used Timeline.js to present the importance of events and policy developments related to crime and Los Angeles’ political and social climate during the post-pandemic period. This interactive timeline provides greater historical context for the dataset by highlighting the social and policy changes that may have influenced the crime patterns depicted in our visualizations from 2021 to 2025.

Presenting Our Narrative

Our findings are presented on an interactive WordPress website to make the crime data accessible and easy to explore. The site uses a clean layout with clear navigation, including ‘Up Next” buttons at the bottom of each page, allowing users to move easily between sections that explain our research question, methods, and results.

Accessibility was also an important consideration in the design process. We wrote descriptive text for each visualization, chart, map, and image used so that screen readers could interpret the visual content. We also embedded hyperlinks within our narrative to make the site easier to navigate for assistive technologies. Finally, all section titles were formatted as proper HTML headers rather than just bold text, so screen readers can easily navigate the page structure.

We also used interactive visualizations that allowed viewers to explore crime patterns across neighborhoods, time periods, and offense categories. These visual tools help highlight trends that may not be immediately visible in raw data tables. Together, the narrative text and visualizations provide a comprehensive overview of crime patterns and the social context surrounding them.