MAPPING CRIMINAL ORGANIZATIONS
The maps and visuals in this page are produced using our 2007–2015 state-month dataset, which combines and processes information from government and expert sources with a nationwide scope. It is the result of work on two ends: First, hand-coding data found in over 60 documents from 11 sources, including Mexican and U.S. government agencies, specialized sources, and experts. And second, carrying out complementary research on the composition and evolution of groups through time.
​
For access to our dataset and updates, sign up for our newsletter by clicking here. For a summary of our methodology, please scroll down.
​
Note: The maps and visualizations presented here are optimized for desktop viewing and may not function correctly on mobile devices. They may be shared and embedded with attribution.
The visualizations above are based on our dataset on organized criminal group (OCG) presence by state and month from January 2007 to December 2015 (State Panel V1.0).
​
Our dataset combines and processes information from government and expert sources with a nationwide scope. It is the result of work in three parts.
​
First, we hand-coded data on territorial presence found in over 60 documents from 11 sources, including Mexican and U.S. government agencies, specialized sources, and experts. We selected our sources carefully to set a uniform standard for inclusion of organized criminal groups (OCGs) and minimize sources of bias. Then we extracted all of the relevant data points from those documents, including the names of the groups, factions, or cells involved; the territory in which they were reported to be present; how that presence was characterized (activities, level of influence in illegal markets, etc.); and the relevant period.
​
Second, we carried out supporting research to identify unique groups and track their composition and evolution through time. This was necessary to organize the “raw” data extracted from our sources, which included 290 unique group names, many of which referred to the same core group, factions of a larger group, or a conglomeration of independent groups.
Third, we transformed our raw dataset into one with uniform and stable group names and four categories of presence intensity (none, minor, significant, major) and used this “cleaned” data to fill in monthly time series for every group present in each state.