This is part two of our conversation with Walter Palmer and Dr. Grant Drawve, PhD, of CAP Index. In this discussion, we delve deeper into the impact that “place” has when analyzing crime risk and crime trends. Find part one here.
LPM: During our last conversation, you made the point that crime trends do not all necessarily move in the same direction and that it’s critical to consider the location and nature of crimes. Can you elaborate on that?
Dr. Grant Drawve: Depending on the crime types being analyzed, there will often be a mix of different targets, different offenders, and different opportunities as well as varying levels of guardianship (e.g., security, bystanders) at the location. For instance, during the COVID-19 shutdowns, people’s routine activities naturally shifted to spending more time at home. Not surprisingly, in residential neighborhoods, especially during the day, burglary decreased dramatically as there was an increase in home guardianship during the pandemic. On the other hand, in neighborhoods where there was mixed land use (e.g., residential, commercial, industrial), there is evidence of crime increasing as non‑residents continued to access those areas even with limited routine activities due to the pandemic.
We recently studied this concept using crime statistics from Charlotte, North Carolina, over the last three years (2021-2023). We selected Charlotte for a) its unique positioning as a larger city that isn’t often featured in the crime headlines, b) Charlotte’s mix of residential areas combined with a strong presence of both small and large businesses from varying industry sectors, and c) the availability of rich crime data from the Charlotte Mecklenburg Police Department. The dashboard can be viewed by visiting capindex.com/charlotte-crime-trends.
LPM: CAP Index often references how “place matters.” Could you provide some additional examples of how “place” creates differing levels of crime even within the same crime type?
Drawve: The Charlotte crime data provides great insight into this discussion with the ability to compare crimes using multiple attributes. When looking at citywide crime change in this area from 2022 to 2023, we see that robbery decreased (7%) while simple assault and burglary, for example, increased (4% and 2% respectively). At a quick glance, this is good information and provides an overview of crime occurrence; however, as we drill down deeper into the data by location type, we find greater variation depending on the location. For instance, retail locations (convenience stores, department/discount stores, grocery/supermarkets, liquor stores, service/gas stations, shopping malls, and specialty stores) experienced a decrease in burglary and robbery with an increase in simple assault, while other commercial and office buildings had a decrease in simple assault and burglary with an increase in robbery. This underscores the fact that citywide trends may be misleading. It’s necessary to consider the specific types of locations to get an accurate assessment of the crime risk in an area.
Walter Palmer: Another caution is that using percentages can be interesting but may not really illustrate material differences. For instance, while retail robbery was down by 2.7 percent, it only represents a decrease of eleven incidents (393 vs. 404) across the entire retail estate in Charlotte. Is that material enough to change any particular retailer’s security posture? Probably not unless the changes are very skewed from one retailer to the next.
LPM: What if organizations analyze the crime trends on a neighborhood-by-neighborhood basis?
Drawve: This adds another layer to the discussion. For example, we often look at crime trends by census block groups, a common unit of analysis which is often used as a proxy for neighborhoods. In this analysis, let’s say we combine aggravated assault and simple assault into the category of “assault offenses.” Across the city of Charlotte, there was a 3.6 percent increase in assault offenses during 2023, but it was not equally distributed across neighborhoods. Of the 484 block groups in Charlotte, only eighteen experienced significant increases in assault offenses, thirteen had significant decreases, while the remaining block groups remained closer to their 2022 counts. This pattern is similar whether we look at this group of crimes as a whole or by location type for assault offenses at residences, retail establishments, or other commercial sites. This data set illustrates the general concept that while neighborhoods (block groups) may remain relatively stable for crime occurrence in larger cities, there will often be significant changes in a select subset of neighborhoods.
Drawve is the VP of Research and Innovation for CAP Index—the pioneer and leader in crime risk forecasting—and has over ten years of experience in applied environmental criminology and crime analysis research. Palmer is CAP Index’s COO/EVP, with over thirty-five years of hands-on experience in the AP industry. Drawve and Palmer can be reached at gdrawve@capindex.com and wpalmer@capindex.com