Forensic science, or more specifically, problems in forensic science, contributes to many wrongful convictions, as seen in in nearly half (45%) of DNA exoneration cases and one-quarter (24%) of all exonerations in the United States.

Through the examination of more than 360 DNA exonerations, the Innocence Project has seen many examples of problems in the foundation or application of forensic science, including:

• Unreliable or invalid forensic discipline. Studies have demonstrated that some forensic methods used in criminal investigations cannot consistently produce accurate results. Bite mark comparison is an example of an analysis that is unreliable and inaccurate.
• Insufficient validation of a method. Some of the forensic disciplines in use may be capable of consistently producing accurate results, but there has not been sufficient research to establish validity. Accuracy of a method should be established using large, well-designed studies. Without these studies, the results of an analysis cannot be interpreted.  Analysis of shoeprints as a basis of identifying the unique source of a print is an example of a method that has not been sufficiently validated.
• Misleading testimony.
  • Sometimes forensic testimony overstates or exaggerates the significance of similarities between evidence from a crime scene and evidence from an individual (a “suspect” or “person of interest”), or oversimplifies the data. Examples include testimony that suggests a collection of features is unique or overstates how rare or unusual it would be to see these features, implying that it is quite likely that the suspect is the source of the evidence, and testimony that doesn’t convey all possible conclusions, as can arise with masking in serology testing.
  • Sometimes forensic testimony understates, downplays, or omits the significance of an analysis that establishes that an individual should be excluded as a possible suspect. An example is testimony that an analysis is “inconclusive” when in fact, the analysis excluded the suspect.
  • Sometimes forensic testimony fails to include information on the limitations of the methods used in the analysis, such as the method’s error rates and situations in which the method has, and has not, been shown to be valid.
• Like everyone, forensic practitioners can make mistakes, including mixing up samples or contaminating specimens. These can occur in any type of science or laboratory testing, even in well-developed and well-validated fields.
• In some cases, forensic analysts have fabricated results, hidden exculpatory evidence, or reported results when testing had not been conducted.

The first major scientific institution to investigate this problem across the board was the National Academy of Sciences (NAS) in its report, Strengthening Forensic Science in the United States: A Path Forward, released in 2009. This report noted that “imprecise or exaggerated expert testimony has sometimes contributed to the admission of erroneous or misleading evidence.” It also found that some forensic techniques, particularly those that deal with comparing patterns or features (such as tire tread impressions, bite marks, fiber, or hair), have not been subjected to sufficient scientific evaluation, and noted that the scientific basis for arson investigations should be strengthened.

The same concerns were reiterated and expanded upon in the 2016 report, Forensic Science in Criminal Courts: Ensuring Scientific Validity of Feature-Comparison Methods, and addendum by the President’s Council of Advisors on Science and Technology (PCAST). This report examined the research underlying specific forensic feature comparison disciplines, evaluated their accuracy and reliability, and made recommendations to various federal agencies to strengthen these disciplines. Among the recommendations was the need for better resources to support judicial training given the changing landscape in the evaluation of forensic evidence and state of validation of various forensic techniques.

Reforms and Solutions

The serious problems identified in forensic science and its practice in the United States can only be addressed by a commitment to reform at multiple levels. The Innocence Project is working at the national level to support efforts to improve forensic science disciplines through research and standards setting, including:

• Ensuring that the National Institute of Standards and Technology (NIST), a preeminent, independent science agency with measurement expertise, conducts scientific evaluations of the validity of the forensic disciplines.
• Increasing funding for research at science-based agencies and institutions, in accordance with a well-developed strategic plan, to establish or strengthen the fundamental science underlying forensic science disciplines.
• Developing rigorous national standards, recommendations for documentation of forensic sciences, and guidelines for reports and testimony for those forensic disciplines that have been shown to be based on robust and reliable science.
• Supporting judicial training and other efforts to ensure that future decisions in admissibility consider the validity of a forensic test in general, and the validity of the test as applied in the specific case at hand.

From 2013 to 2017, the National Commission on Forensic Science (NCFS) advised the nation on this path forward, and provided recommendations to the attorney general and the director of NIST on several important topics, including a uniform code of professional responsibility, requirements for report contents, discontinuing the use of the phrase “reasonable degree of scientific certainty”, and the necessity of independent scientific evaluation of the validity of forensic science disciplines.

To improve the quality of forensic science practices in laboratories, the Innocence Project supports efforts to promote error identification and prevention, improve transparency of information, and ensure just outcomes, including:

• Adopting of procedures that reduce the influence of “cognitive biases” on an analysis, such as preventing access to extraneous and potentially biasing information.
• Adoption by every forensic science service provider of a root cause analysis procedure, as recommended by the NCFS, the results of which should be transparent and available to the public.
• Continuation of federal funding for the Coverdell Forensic Science Improvement Program to support state and local forensic science service providers and to ensure independent external investigations when allegations of negligence or misconduct are made.
• Retrospective review of cases when problems in forensic science are identified. Examples include the joint microscopic hair comparison review undertaken by the Federal Bureau of Investigation, the National Association of Criminal Defense Attorneys, and the Innocence Project, and reviews conducted by the Texas Forensic Science Commission in response to complaints filed in the state.
• Establishment of a duty-to-correct mistakes and a duty-to-notify affected defendants when errors or significant breaches of scientific principles or ethics are identified.

Lastly, as a matter of fairness and justice, the Innocence Project supports judicial examination (post-conviction review) in cases in which unsound science may have contributed to a conviction. Statutes addressing this issue have recently been passed by Texas and California.