Marijuana and Crime in Colorado

Marijuana is something that has been historically associated with crime. This is because of its association with immigrants, who also have been historically attributed with crime. These times, however, are much different and Marijuana is now some form of legal in 23 states including our nation’s capital. With marijuana expanding across the states and being in many neighborhoods and having a big impact on the economy as a whole, I thought it would be interesting to take a look at how, or if, marijuana sales are correlated with crime. I also wanted to look at specific crimes in this regard to see if marijuana has an effect on certain types of crime.

In terms of the summary statistics, I found that the max for the crimes would indicate that crimes against person and property are the two most popular, with narcotics following and drug equipment tailing as the least committed. Also we can see that there are more medical

First, I decided to use 2 data sets: sales of marijuana by county, crimes by county. First I looked at the plotted points between crimes and marijuana sales. and I found some interesting stuff. The one thing that definitely stood out was the fact that retail marijuana had higher positive correlation, meaning as sales increase of marijuana, so too does The other thing that popped out was that more violent crime was attributed to marijuana sales rather than petty crimes and narcotic possession.

(Here are some scatter plots showing that crimes against person have the tightest and most linear distribution meaning strong positive correlation)

For the uninitiated, correlation simply measures how positive or negative the relationship is between all data points between two sets. A 1 is 100% positive and a -1 is total negative correlation. Notably, drug equipment had the least correlation with weed sales(.22 for medical and .27 for retail).

Here is a heat map showing correlation, all you need to look at is the bottom left corner for this one where we can see the correlation between crimes and marijuana. You can see that yes, for sure there is a positive correlation between these factors, and all evidence points to this. Honestly, a lot of this could be due to factors including population and strong crime counties. Hence, I will be continuing to investigate this further to see what in specific might be affecting crime.

To corroborate some of these findings I read an article posted by CU Denver where they studied the same question I have.

https://news.ucdenver.edu/do-marijuana-dispensaries-increase-neighborhood-crime/

To quote, “Our findings have important implications for the marijuana industry in Denver and the liberalization of marijuana laws nationwide,” suggests Lonnie M. Schaible, PhD, associate professor in the School of Public Affairs at the University of Colorado Denver, who co-authored the study. “Although our results indicate that both medical and recreational marijuana dispensaries are associated with increases in most major crime types, the weak strength typical of these relationships suggests that, if Denver’s experience is representative, major spikes in crime are unlikely to occur in other places following legalization.” In this regard I think my findings are something that has been observed by other studies, but I do agree in the fact that they are right, there isn’t anything too damning to show that marijuana sales have a strong correlation with an increase of crime.

As shown by this graph, crime is just generally on the rise, and so are marijuana sales. To think that one factor like marijuana sales could have that large of an impact is obviously just not logical. Yes, crime is rising and so too is marijuana, but we know that there are so many factors responsible for this (homelessness, poverty, population surging). I will note one notable outlier, which is Adam’s county which has more crime than even the largest counties, and it is the 5th most populous. It also is the 5th highest selling for marijuana in Colorado. So maybe that indicates something about the culture in that county.

To end, I think there is some effect on crime because of what we see between medical and recreation correlations. The effect however is generally small. I am going to forge ahead and continue my work analyzing this because I think it is fascinating to look deeper into, I also want to see some effects of alcohol sales and compare the correlations between that and marijuana.