By The Philadelphia District Attorney's Office, Data Lab

The Conviction Integrity Unit (“CIU”) was established in 2018 by District Attorney Larry Krasner. The CIU’s predecessor, the Conviction Review Unit (“CRU”), which was established in 2014, had operated for a number of years with only a small staff and a narrow mandate. The CRU only reviewed claims of actual innocence, and rarely undertook investigations into whether new evidence existed that could prove those claims. Cases where the defendant had confessed were largely excluded from consideration, as if false confessions (which occur in a quarter of DNA exonerations nationally) were always reliable. Today, the CIU is an independent unit within the Philadelphia District Attorney’s Office, reporting directly to the District Attorney, and involved in one out of every ten homicide exonerations in the country. When District Attorney Krasner transformed the unit from the CRU to the CIU, he immediately tasked it with a broader mandate: not only to review past convictions for credible claims of actual innocence but also to review claims of wrongful conviction and secondarily to consider sentencing inequities. Early in his first term, District Attorney Krasner merged the CIU with the Office’s Special Investigations Unit (“SIU”). The two units share a common focus on investigating official misconduct, and their cases frequently overlap. However, as the CIU and SIU personnel have grown and expanded their caseloads, the units were separated in the summer of 2020 to better accommodate each unit’s mission

The CIU’s mission is to ensure that justice is served by prosecutors at the Philadelphia District Attorney’s Office and to remedy the Office’s wrongful convictions. Pennsylvania prosecutors have limited post-con viction discretion in general and they have no legal authority to set aside convictions in the interest of justice. Since CIU prosecutors cannot unilaterally dismiss an existing conviction or free anyone we believe to be wrongfully incarcerated, the CIU makes a recommendation to the court that the petitioner be granted a new trial whenever its independent investigation leads it to conclude that a conviction lacks integrity. If warranted, the CIU will move to withdraw the charges against the petitioner or reduce the charges so that an equitable sentence can be imposed. In cases that are ultimately withdrawn or dismissed, the CIU will investigate and prosecute the actual perpetrator where feasible. However, given the inherent difficulties involved in investigating decades-old crimes where the original investigation was either botched or inadequate, identifying the real perpetrator and bringing that person to justice may be impossible. To date, the Philadelphia Police Department has declined to re-open and re-investigate old cases following exonerations. For example, Walter Ogrod was exonerated of a 1988 murder in 2020. While investigating the case, the CIU identified two alternate suspects. As of almost a year after Ogrod’s exoneration, however, police had not even begun the process of re-opening the underlying murder case. Additionally, the CIU believes that conviction integrity is more than simply fixing past mistakes and exposing misconduct. It also requires policies and processes to prevent future injustices. With this aim, the CIU helps craft office-wide policies and trainings designed to reduce the number of future wrongful convictions.

This report encompasses exonerations, commutations, and sentencing adjustments from January 1, 2018 through June 15, 2021. This report includes data on cases submitted to the CIU, active investigations, cases declined or closed, and cases awaiting review that are accurate as of May 31, 2021. Experts who have opined on the issue of best practices for conviction integrity units agree that in order to increase public understanding of and trust in such units, offices should publish annual reports detailing the results of their conviction and case reviews and actions taken. This report is the first report issued by the CIU under District Attorney Krasner and is a first-term report, rather than an annual report. Although annual reports were contemplated, they were postponed as a result of multiple factors ,including lack of resources, internal technology deficits, case load, and the COVID-19 pandemic.

Philadelphia District Attorney's Office, Data Lab. 2021. 47p.

By Emma Zürcher, Lana Eekelschot, Annalena Wolcke, Lucy Strang

Police performance measurement is a complex activity that encompasses considerations of the social, legal, institutional and political contexts in which a police force operates. The definitions and methods used need to be regularly updated and adapted to reflect the constant changes in these contexts. This study follows from a broader desire, articulated by the national police force and the Ministry for Justice and Security, to improve how police performance is measured in the Netherlands. Indeed in 2014, Tilburg University and KU Leuven conducted a study to develop a new framework for productivity measurement of the Dutch police.1 While this study provided useful insights, approaches to performance measurement adopted by police forces abroad may also offer valid examples for the Netherlands and provide opportunities for general learning. Research design The study had two main goals: 1) To gather insights into how different police jurisdictions have approached performance measurement. 2) To assess what lessons these approaches can offer for improving police performance measurement in the Netherlands. A series of research questions were developed to meet these goals (see Section 1.1.2.). These questions focused on the methods and indicators used to measure performance in a selection of ten policing jurisdictions,2 including the Netherlands; the stated purposes of the performance measurement; the reliability of the approaches; identified or potential adverse side effects of measuring performance; and examples of good or innovative practice in these areas. In addition, several research questions focused on the methods and indicators used to gain insight into the performance of the Dutch police; the aspects of policing not currently captured in this framework; the lessons that approaches to performance measurement in other jurisdictions may offer the Netherlands; and the transferability of these approaches to the Netherlands. To address these research questions, the study team conducted a targeted literature review and expert and stakeholder interviews in each of these jurisdictions, including the Netherlands. The team then selected five case study jurisdictions for more in-depth data collection and analysis: England and Wales, Israel, North Rhine-Westphalia (Germany), Seattle (United States) and Sweden. Using the resulting findings, the study team focused on extracting current practices from the case study jurisdictions that had potential to offer  starting points for improving police performance measurement in the Netherlands. These are presented in the main body of this report. For the countries not selected as case studies (Australia, Canada, Finland, New Zealand and the United States), the study team prepared high-level summaries of current approaches and highlighted several relevant practices that may be of interest to policy makers for the Dutch National Police. These are presented in Annex A. To provide focus and enhance the feasibility of the research, this study looks exclusively at how police performance measurement has been approached by stakeholders who can directly influence policing policy, such as the police force and national- and local-level government stakeholders. For each of the case study jurisdictions, the research team has analysed the practices used by the relevant police force(s) and the administrative and/or political authority under which they sit. We are confident that the findings will therefore be of particular interest to decision-makers for the Dutch National Police, including police leadership at both national and regional levels and the Ministry of Justice and Security, and provide leads for improving how police performance is currently measured.        

Cambridge, UK: RAND Europe, 2023. 182p

By Olivia  Elizabeth Vere Taylor, Richard Philpot, Oliver Fitton, Zoë Walkington, Mark Levine

Introduction: Recent high-profile cases of police misconduct have revealed that officers were often aware of misconduct, but remained silent, compromising public trust in law enforcement. Here, we systematically review ‘police whistleblowing’ literature to identify barriers and facilitators to officers challenging misconduct. Methodology: Employing PRISMA guidelines, we systematically reviewed 118 relevant papers, extracting data and coding key variables including who the ‘target’ of the research was; whether reporting practices were studied, and whether practical solutions were offered. A reflexive thematic analysis then assessed consensus among researchers within the literature. Results: Five themes - 1) knowledge and rules, 2) consequences, 3) interpersonal relations, 4) responsibility, and 5) police culture and group relations – emerged as barriers and facilitators to whistleblowing. The review revealed relatively poorer representation of internal police reporting structures and limited practical solutions, with only 40 papers proposing strategies, predominantly centred on training and education. Discussion: This review highlights methodological limitations in existing research, with an overreliance on survey methods and a dominant focus on the characteristics of individuals over the structural constraints of reporting. The positive impacts of whistleblowing on policing as an institution and the development of practical strategies to overcome officers’ reluctance to report misconduct remain largely unexplored. 

Journal of Criminal Justice, 2024, 15 pages

By Arthur H. Garrison

This article uses five years of data, 2015–2019, from the Washington Post dataset on police use of deadly force and enhances the data with 21 situational and 8 police perceived threat measure variables to put in context the use of force and the disproportionate incidents involving Blacks compared to other races. Rather than comparing percentage race outcomes of police use of force to general population or behavior proxy measures the benchmark of situational contexts of police use of force is used to interpret race percentage distributions by race within the same context. Under this analysis, the top three situations that result in police deadly force involved 1) an assault or civilian call for help, 2) a call for domestic violence, and 3) a police officer being attacked. When viewing police shootings by situation and race percentage distribution, the data shows that police use of force is differentiated. Within the same situations Blacks were more likely to be shot and killed than Whites. Blacks were more likely to be shot by the police in a traffic stop, were more likely to shot by the police mistaking them as armed, were more likely to be shot if they are perceived to be suffering mental illness and were more likely to be shot if the police are responding to call for illegal drug activity than other races in the same situations and contexts that resulted in police use of deadly force.

Journal of Race and Policy 16(1) 2022