Prosecutors have immense power in the criminal justice system. They decide whom to charge with crimes, what punishment to seek and what evidence to use at trial. State and national legal and ethical standards serve as a check on that power. These rules are meant to deter prosecutors from withholding exculpatory evidence, failing to disclose expert witnesses and making inappropriate trial statements, among other improper acts.

In Ohio, prosecutorial misconduct happens when a prosecutor violates a law meant to guarantee the right to a fair trial, even unwillfully, and that improper conduct prejudices a defendant's constitutional due process protections, requiring the court to overturn the conviction. Prosecutors can also engage in improper conduct that violates those laws but does not require reversal of the conviction. This may also involve not taking adequate measures to be certain that other elements of the justice system, including the police, are also meeting their obligations to provide information related to the prosecution to the defendants and their counsel.

In other states and in academic articles, improper conduct is often also referred to as prosecutorial misconduct, but in Ohio the latter term specifically implies that the case was reversed.

To investigate the impact of this improper behavior on defendants, their families and Ohio's justice system, Columbia Journalism Investigations, NPR, member station WVXU in Cincinnati and The Ohio Newsroom built a first-of-its-kind database by reading and analyzing thousands of pages of state appellate decisions from 2018 to 2021. We consulted legal experts who study prosecutorial misconduct and improper conduct by prosecutors to develop a database methodology. They offered guidance throughout the data analysis. A legal researcher from Columbia Law School also helped our team of reporters categorize the state appellate rulings.

We gathered appellate opinions from the Ohio Supreme Court online docket using specific keywords from landmark court decisions related to prosecutorial misconduct and expressions commonly used by appeals courts when they address alleged errors. Chosen keywords include:

—Harmless error
—Harmful error
—Prosecutorial misconduct
—Prosecutor misconduct
—Improper remark
Brady v. Maryland
Doyle v. Ohio
Griffin v. California
Batson v. Kentucky
Darden v. Wainwright
Napue v. Illinois
Giglio v. United States
Slagle v. Bagley
Oregon v. Kennedy
Booth v. Maryland
—Crim.R. 16
—Criminal Rule 16

We logged more than 1,600 allegations of prosecutorial misconduct using Airtable, a relational database platform. Roughly 450 claims made it into the final database. Based on our methodology, we excluded any claim originating from criminal trials prior to 2001; originating in civil court, juvenile court or grand jury proceedings; not handled by county prosecutors; or that had been procedurally barred.

We classified the alleged misconduct and the court rulings according to the categories used by the American Civil Liberties Union (ACLU) of New Jersey in its exhaustive 2012 report on prosecutorial conduct in that state.

The types of misdeeds catalogued in the Ohio database include: 1) failing to disclose exculpatory evidence (known as a Brady violation), 2) using improper arguments either in closing arguments or opening remarks, 3) discriminating in jury selection (Batson violation), 4) failing to disclose evidence or expert witnesses (discovery violation), 5) asking leading questions to a state's witness or letting them introduce false evidence, 6) errors in plea deals and 7) other.

We catalogued appellate rulings on the misconduct allegations according to one of six categories, depending on how the alleged misconduct might have affected the defendant's right to a fair trial. The categories include:

  • No error — court found no improper conduct.
  • Did not reach error — court did not directly address the alleged misconduct.
  • Harmless error — court found improper conduct but did not reverse the case.
  • Reversible error — court found prosecutorial misconduct, reversing the conviction.
  • Error but reverses on other grounds — court found improper conduct, but the conviction is reversed for other reasons, such as technicalities.
  • Did not reach anything — the alleged misconduct is rendered moot. 

Our dataset coded those claims as improper-conduct cases only if they ended in either a reversible error, whereby the leading appellate judge granted a new trial after ruling that the state's misdeeds had tainted a defendant's criminal proceeding, or a harmless error, whereby the judge found an impropriety in either the prosecutors' actions or, by extension, the state's actions, but ruled it did not warrant a new trial.

Based on these classifications, appellate judges found some kind of improper conduct in 104 total criminal cases. When appeals courts rule on these claims, their decisions almost always shield the prosecutors' identities. The team then worked to identify the prosecutors involved in our dataset. Prosecutors' identities were uncovered by either cross-referencing appellate opinions and briefs with the original trial transcripts or reaching out to trial and appellate defense attorneys, as well as clerks of court. We identified 109 prosecutors responsible for the improper conduct in 100 cases. In four cases, we were unable to verify the prosecutors responsible for the improper conduct.

For Brady and discovery violations, the first- and second-chair prosecutors who handled the original criminal trials were considered responsible for the misconduct. For cases involving improper or inflammatory remarks, only the prosecutors who made those comments were identified.

Throughout this two-year investigation, legal experts have graciously given their time and shared their expertise to advise the team on its data methodology and analysis. They include Alex Shalom, senior supervising attorney and director of Supreme Court advocacy at the ACLU of New Jersey, who authored the 2012 ACLU of New Jersey study on prosecutor conduct in that state; Samuel Gross, University of Michigan law professor and co-founder of the National Registry of Exonerations (NRE); Margery Koosed, University of Akron School of Law professor who specializes in Ohio criminal procedure; Maurice Possley, NRE senior researcher who co-authored a 2010 report on prosecutorial misconduct in California and an NRE 2020 report on prosecutorial misconduct in wrongful conviction cases; James Liebman, Columbia Law School professor who specializes in capital and federal habeas appeals; and Marissa Bluestine, assistant director of the Quattrone Center for the Fair Administration of Justice at the University of Pennsylvania Carey Law School, who co-authored a recent report on prosecutorial misconduct in Pennsylvania.

Taylor Eldridge, a former data fellow at PBS' Frontline, contributed to the initial design of the database methodology. Additional contributors include Columbia Journalism School students enrolled in Jelani Cobb's Covering Race class and Tom Meagher's Criminal Justice Reporting class in the spring of 2022: Julian Abraham, Marielle Argueza, Joshua Arshonsky, Jessica Blaeser, Samantha Chaney, Ilyssa Daly, Jessy Diamba, Jaden Edison, Naomi Feinstein, Susanna Granieri, Isabelle Hamilton, Sirena He, Colin Hogan, Jasmine Hyman, Jehangir Irani, Elizabeth Janowski, Rachyl Jones, Tanya Kaushal, Meg Kelly, Shakti Langlois-Ortega, Tandy Lau, Chantelle Lee, Tiamen Montgomery, Evelyn Nam, Ramisa Rob, Marcela Rodrigues and Sofia Saric.

Copyright 2023 NPR. To see more, visit https://www.npr.org.