DC's fragmented crime data landscape
DC's fragmented crime data landscape
Siloed, delayed and ignored data
DC’s criminal justice “ecosystem” produces lots of data but can be incredibly difficult to navigate. Currently, one is largely unable to follow how crimes interact with arrests, cases, sentences or any of DC’s violence prevention programs. For example, the “Justice Statistical Analysis Tool” has 12 different dashboards and only 3 of them have timely data:
The lack of timely public data lets some policymakers pick and choose when they drop out of context data to bolster their arguments or evade meaningful follow up after announcing a new initiative. While DC should strive to make its fragmented data more accessible (like this excellent Manhattan dashboard), this post is a guide to where to find different types of data across these topics:
Crimes
Police & investigatory actions
Prosecutions
Courts & sentencing
Post-sentencing outcomes
Prevention
I’ll try to keep this post updated if additional useful data sources become available so that it can be an ongoing resource.
Crimes
DC Crime Cards is a great resource for crime data and is refreshed daily. It has all reported property and violent crimes back to 2008. Each crime is coded by its type, time of day and location (to include geographic divisions like ward, ANC, BID etc.), method (gun/knife/other). The default options to “cut” the data in the user interface are pretty useful and one can click the Download this Data button to get the raw data for analysis.
DC Crime Cards is far more timely and accessible than most of DC’s crime data sources. The main limitations are that it doesn’t include any information about victims or suspects or which crimes are “closed” by arrest or other means. For simple “is crime up or down?” style questions MPD also has a “Crime Data at a Glance” page. Open Data DC posts monthly data files for moving violations issued in DC.
Police & investigatory actions
The Metropolitan Police Department (MPD) does share quite a bit of data about operations though it’s often very infrequent:
Arrests: MPD releases relatively detailed arrest data for adults annually and for juveniles every 6 months. Adult arrest data includes the arrest location, date, hour, charge and defendant race, ethnicity and gender. Juvenile data is less detailed with only arrest date, charge, the defendant’s home Police Service Area (PSA) and the PSA where the crime occurred. Neither adult nor juvenile arrests data is otherwise linked to specific crimes or subsequent prosecutorial decisions.
Stops: Stop data is published infrequently (MPD has even been sued for not publishing it on time) and the latest dataset is through 2022. Stop data has dozens of columns noting the time, location, stated reason, arrest charges, and the stopped person’s race, ethnicity and gender.
Staffing: MPD updates the number of officers assigned to different areas or departments every month. This data does not detail where officers actually work or how they spend their time. MPD has separate reports for hiring and attrition that are supposed to be monthly but are often delayed. Lastly MPD appears to post its overtime reports every quarter. Collectively these reports provide some transparency into how MPD is assigning its officers but none of this data is tied to a transparent staffing model.
Discipline: MPD posts annual data files of “Adverse Actions” that include the category, brief description of the conduct and the discipline imposed for MPD officers violating policy/orders. These are anonymous but include the officer’s rank, race and gender. The Office of Police Complaints posts their decisions here.
Beyond these data sources the best source for information on MPD operations are often their responses to Council oversight questions. These are often massive documents that are created annually. MPD also produces Annual Reports and Specialized Reports that sometimes have information that isn’t published anywhere else.
The Department of Forensic Sciences (DFS) plays an important role but apparently only publishes operational data in response to Council oversight questions and in incredibly delayed Annual Reports. These reports give snapshots into evidence collection and outsourcing volumes but almost no granular detail. More on DFS’ operations in this post:
Prosecutions
Before this year the only public documents about adult prosecutions in DC were the USAO’s “Annual Statistical Reports” where page 67 was a simple data table that covered the matters in DC Superior Court. This includes the numbers of cases reviewed, presented, declined and volumes of dispositions (guilty pleas, trial outcomes etc.) but no granular data. In response to public demands for more transparency, the USAO now reports monthly updates of year-to-date volumes of cases presented, active prosecutions, trials and guilty pleas. However these updates do not include the number of cases they decline to prosecute or the cases they have dismissed. This is a common tactic of only presenting large numbers of “positive” metrics without any context or trend.
Juvenile prosecution data is even more limited. While the Washington Post received data from DC’s Office of the Attorney General (OAG, which prosecutes juveniles in DC) that the OAG only declined to prosecute 26% of cases (vs 67% for the USAO), that is the only snapshot I am aware of into the OAG’s prosecutions.
Courts & sentencing
The DC Courts publish a lot of data in their Annual Reports but that is infrequent and this year the 2022 report only came out in September 2023. These reports cover the volumes of cases filed, disposed and method of disposition (dismissed by prosecutor, plea, trial verdict etc.) but at very aggregated levels. There are no granular breakdowns more specific than broad categories like “Felony” or “Traffic.”
Pretrial assessment and release is run by the Federal Pretrial Services Agency (PSA) and they tend to only publish data annually as part of their Congressional Budget Justifications. They do report on a number of performance metrics like re-arrest rate, court appearance rate and service provision but only at program-wide levels without any granular detail. The Department of Corrections (DOC) has relatively up-to-date and comprehensive data available though most of their reports combine the pre-trial detainee and post-sentencing population as the DOC is responsible for both.
The DC Sentencing Commission does provide relatively granular sentencing data and solid analyses in their Annual Reports. The Sentencing Commission deserves praise for being one of the only parts of the criminal justice ecosystem to produce reports on how cases flow through the system from arrest all the way to sentencing:
Post-sentencing outcomes
The Federal Bureau of Prisons has monthly counts of the number of DC residents and those that committed crimes in DC and who are currently in a federal prison. Their dashboard also provides useful yearly trends in the prison population, admissions and releases:
The federal Community Supervision Program (CSP) is responsible for the ~6,500 Washingtonians on parole, probation or supervised release. This program is part of the (also federal) Court Services and Offender Supervision Agency (CSOSA) which tends to only release aggregated annual data as part of the Congressional Budget Justification process. These reports detail how many people are in each program, under what level of supervision, re-arrest rates and GPS monitoring rates. They also go into solid detail about the challenges this population faces like rates of unemployment, housing instability, substance use disorders etc.:
The main limitations here are that this is annual and aggregated data. There’s no way to analyze subgroups or know if things are improving/degrading until months after the fact.
For juveniles the Department of Youth Rehabilitation Services (DYRS) is all over the place in terms of data. On the one hand, they update their census of detained juveniles daily and also offer a monthly trend view. On the other hand, most of the data and reports on their primary website (including important stats like re-arrest and re-conviction rates) haven’t been updated since before 2020.
Prevention
All of the things that make for a stronger, healthier community contribute towards crime prevention but for this post we’ll focus on some of the major risk factors for kids from the CJCC’s reporting:
The Child and Family Services Agency (CFSA) maintains a quarterly dashboard that tracks the volumes of children and families served in different settings and the outcomes of abuse and neglect investigations. Homelessness and housing instability is difficult to measure and often relies on annual point-in-time counts. Student truancy and similar metrics are usually only reported annually.
The OAG’s Cure the Streets program reports outcome and process metrics in a dashboard that appears to be updated monthly. The similar program at the Office of Neighborhood Safety and Engagement (ONSE) does not report process metrics though the Office of Gun Violence Prevention has an excellent dashboard tracking outcome metrics (shootings) in both the ONSE and Cure the Streets areas. As far as I can tell prevention programs like Pathways and People of Promise only report any data during Council oversight.
Unfortunately, even when DC government analysts overcome this incredibly fragmented data landscape and create comprehensive overviews of crime in DC they are often ignored. The NEAR Act mandated the Deputy Mayor for Public Safety and Justice (DMPSJ) produce an annual report on felonies. This report actually identified the USAO’s low (and decreasing) prosecution rates years before it got media attention. Unfortunately, today’s Auditor report on the NEAR Act contained this disheartening passage:
“The DMPSJ’s annual reports on felony crime data provide a wealth of information intended to provide a more comprehensive view of outcomes in the District’s bifurcated, local-federal criminal justice system, tracing crimes from their occurrence through their adjudication in court. Nevertheless, ODCA interviews with D.C. policymakers in the executive and legislative branches indicated that the reports are not widely used; one senior official involved in criminal justice policy stated that she had not looked at the reports.”
This kind of finding is incredibly frustrating. Crime matters and so many of DC’s political leaders have built their brand around taking crime seriously. But people who actually took this seriously would want solid, timely data (rather than just vibes) to guide their decisions. People who took this seriously would find the minutes needed to read the reports that analysts spent dozens of hours preparing. Instead we have a fragmented, delayed mess of siloed databases that aim to achieve the barest minimum of transparency mandated by law and too much of our best analysis is ignored by policymakers.