Nicole Rogers

Author: Nicole Rogers | Major: Data Science | Semester: Spring 2023

In the Fall of 2022, I had an internship working with the Crime and Security Data Analytics Lab (CASDAL) under the Terrorism Research Center at the University of Arkansas. CASDAL works to train students in areas and methodologies of crime analysis and is led by Dr. Grant Drawve. Through my internship I discovered data science and analytics could be used to gain insights into crime and security issues. My work at CADSAL incorporated data visualization, communication, data cleaning, and statistical and geospatial analysis. Inspired and excited by work at CASDAL I knew I wanted to apply my new skillsets to a research project.

My research is a Spatiotemporal Analysis of Violent Crime in Little Rock, Arkansas from 1999-2022. Few police departments have historical data, so the longitudinal design of this study is unique as I am working with over two decades worth of crime data. My research mentor, Dr. Drawve, has a memorandum of understanding with the Little Rock Police Department which will allow me to access historic crime data from 1999-2014. Data from 2015-present is publicly available through the city’s data portal. The focus of study is on violent crime, specifically aggravated assault, and homicide, across 23 years (approximately 45,000 incidents) and the potential spatial stability of violence.

To provide a more precise description of crime occurrence across the city, my study focuses on a micro-unit of analysis: street segments. Yearly violent crime data were aggregated to the street the incident occurred on to identify patterns over the 23 years. Prior studies typically focus on larger metropolitan cities making the results less comparable to smaller cities. Additionally, they often examine larger spatial units such as neighborhoods.

The results of the spatial analysis determined that crime is extremely concentrated in Little Rock. Crime concentration is often looked at in relation to the 80/20 rule which is a theoretical concept in which a large majority of the incidents occur at a small minority of locations. In the current study, we would expect 20% of the streets to be considered risk prone and account for about 80% of the crime. My findings indicate that the concentration of violent crime is even more defined with 80% of historical violent crimes occurring on only 14% streets. The high concentration of violent crime is further proved with the fact that 1% of the streets account for 27% of the violent cries. In fact, within the top 1%, only 8 street segments are responsible for 5% of the violent crimes. These percentages speak to the longitudinal concentration of violence in Little Rock, Arkansas.

My research has been possible with the funding of the Honors College Research Grant. It has given me an opportunity to expand my portfolio by working with stakeholders in the real-world. As a data science major, it also gave me an avenue to expand my skillset and apply my knowledge into a different field. It will be a combination of my new skills, continued learning on new tools, and my continued expansion into data science that will help me to expand upon this semester’s research into the next.