Forensic science in criminal justice is the use of scientific methods to analyze physical evidence from a crime, producing results that police, prosecutors, defense attorneys, and juries rely on to decide what happened and who was responsible. It covers dozens of specialized disciplines, from DNA profiling and toxicology to fingerprint and firearms comparison. Some of those disciplines rest on solid laboratory science. Others have been found by federal scientific reviews to lack the validation their courtroom use has long assumed. Knowing the difference is now central to how forensic evidence is presented, challenged, and admitted.
The Main Forensic Disciplines and How Reliable Each One Is
Forensic science is not one field. It is a collection of specialties built around different types of evidence, and courts increasingly treat them differently depending on how well the underlying method has been scientifically tested.
DNA Analysis
DNA analysis identifies genetic markers in biological samples such as blood, saliva, or skin cells and uses them to link or exclude a person as the source. Single-source samples and simple two-person mixtures are widely regarded as the most reliable identification tool in forensic science. Complex mixtures from three or more contributors are harder to interpret, and the 2016 President’s Council of Advisors on Science and Technology (PCAST) report concluded that current methods for analyzing those complex mixtures have not been established as scientifically valid.1The White House. Forensic Science in Criminal Courts: Ensuring Scientific Validity of Feature-Comparison Methods
Toxicology and Pathology
Forensic toxicology detects drugs, alcohol, and poisons in blood, urine, or tissue and measures their concentration, because the difference between a therapeutic and a lethal dose often determines whether a death is ruled accidental or intentional. Forensic pathologists perform autopsies to determine the cause of death (the specific injury or disease) and the manner of death (homicide, suicide, accident, natural, or undetermined).2National Institute of Justice. Overview of Forensic Pathology During the autopsy, the pathologist can recover physical items like a bullet and document wound patterns that may be matched to a weapon.
Firearms, Fingerprints, and Pattern Comparison
Firearms examiners compare microscopic marks left on bullets and cartridge cases to decide whether they came from a particular weapon. Latent fingerprint analysis compares friction-ridge patterns from a scene with known prints. These are pattern-comparison disciplines, and their scientific standing varies.
PCAST found latent fingerprint analysis to be “foundationally valid” but noted that its false-positive rate is “substantial and is likely to be higher than expected by many jurors.”1The White House. Forensic Science in Criminal Courts: Ensuring Scientific Validity of Feature-Comparison Methods Firearms analysis, by contrast, “currently falls short of the criteria for foundational validity,” with insufficient empirical studies to establish reliable error rates. Microscopic hair comparison and bite mark analysis fared worse still, and were found scientifically invalid as identification tools.
Trace and Digital Evidence
Trace evidence covers microscopic transfers like fibers, soil, and glass. Digital forensics recovers and analyzes data from computers, phones, cloud accounts, and network traffic, with the main scientific concern being proof that the data has not been altered between seizure and trial. Investigators generate a cryptographic hash at collection that functions as a digital fingerprint of the file; comparing that hash to one generated at trial shows whether anything changed.
How Courts Decide Whether Forensic Evidence Is Admissible
Judges serve as gatekeepers for scientific evidence. Before a jury hears a forensic expert, the court applies a legal test to decide whether the method is reliable enough to be presented. Two tests govern U.S. courts.
The Frye Standard
The older test comes from the 1923 case Frye v. United States, which held that a scientific technique must be “generally accepted” within the relevant scientific community before its results can come into evidence.3Cornell Law Institute. Frye Standard A handful of states still use this test, particularly in criminal cases.
The Daubert Standard
Most jurisdictions follow the framework established in Daubert v. Merrell Dow Pharmaceuticals, Inc., decided by the Supreme Court in 1993.4Justia. Daubert v. Merrell Dow Pharmaceuticals, Inc. 509 U.S. 579 (1993) Under Daubert, the judge evaluates the science itself using several factors:
- Whether the theory or technique can be and has been tested.
- Whether it has been subject to peer review and publication.
- The known or potential error rate.
- Whether maintained standards control how the technique is performed.
- Whether the method is generally accepted in the relevant scientific community, now one factor among several rather than the sole test.
The Supreme Court later clarified in Kumho Tire Co. v. Carmichael that these factors are illustrative, not a rigid checklist, and their relevance depends on the nature of the testimony.
Federal Rule of Evidence 702
These admissibility principles are codified in Federal Rule of Evidence 702, which governs expert testimony in federal courts and has been adopted in substance by most states. The rule allows a qualified expert to testify if the party offering the testimony shows it is “more likely than not” that the expert’s knowledge will help the jury, that the testimony rests on sufficient facts or data, that it is the product of reliable principles and methods, and that the expert has reliably applied those methods to the facts of the case.5Legal Information Institute. Federal Rules of Evidence Rule 702 – Testimony by Expert Witnesses
A 2023 amendment to Rule 702, effective December 1, 2023, explicitly added the “more likely than not” language to clarify that the party offering expert testimony bears the burden of establishing reliability by a preponderance of the evidence. The amendment responded to concerns that many courts had been applying a more lenient standard that let questionable expert testimony reach juries too easily.
How Forensic Experts Testify
Before a forensic scientist can present findings to a jury, the court must formally qualify them as an expert witness. That happens in a preliminary proceeding called voir dire, where the judge reviews the person’s education, training, and professional experience against Rule 702’s requirements. Once qualified, the expert can offer opinions based on specialized knowledge, something ordinary witnesses are generally not permitted to do.
On direct examination, the expert explains their findings in language a lay jury can follow. Oversimplifying can distort the science; too much jargon leaves jurors unable to evaluate what they have heard. On cross-examination, the opposing attorney probes the limitations of the method, the possibility of human error, and whether alternative explanations fit the data. Defense attorneys commonly challenge the analyst’s error rate, the laboratory’s accreditation status, and whether proper contamination controls were followed.
Experts are legally bound to stay within the scope of their qualifications and cannot claim certainty the science does not support. They must acknowledge margins of error and avoid definitive statements that exceed what their testing can show. This is where many forensic disciplines have historically gotten into trouble, with examiners testifying to “matches” or “identifications” without disclosing the method’s actual error rate.
What Federal Reviews Have Said About Forensic Reliability
Two landmark government reports have reshaped how the legal system views forensic evidence, and both are routinely cited when courts weigh admissibility today.
The 2009 NAS Report
The National Academy of Sciences published “Strengthening Forensic Science in the United States: A Path Forward” in 2009. Its conclusions were blunt. The report found that the forensic science system “has serious problems that can only be addressed by a national commitment to overhaul the current structure,” with wide variability across disciplines in reliability, error rates, and research foundations. In most areas of forensic science, no well-defined system existed for determining error rates, and proficiency testing showed that some examiners performed poorly.6Office of Justice Programs. Strengthening Forensic Science in the United States: A Path Forward
The NAS drew a sharp distinction between laboratory-based methods like DNA analysis and toxicology, which rest on established scientific principles, and pattern-comparison methods like fingerprints, toolmarks, bite marks, and handwriting, which rely heavily on an examiner’s subjective interpretation. For the pattern disciplines, the report found that no studies of large populations had been conducted to establish the uniqueness of the marks or features being compared, yet examiners routinely made probabilistic claims based on personal experience.
The 2016 PCAST Report
Seven years later, PCAST evaluated whether those gaps had been closed. The report assessed individual disciplines against a specific standard: foundational validity, meaning reproducible procedures and empirical error rates established through properly designed studies. Its findings discipline by discipline:
- DNA analysis of single-source samples and simple mixtures: foundationally valid.
- Latent fingerprints: foundationally valid, but with a substantial false-positive rate likely higher than jurors assume.
- Firearms analysis: falls short of foundational validity due to insufficient empirical studies.
- Bite mark analysis: “far from meeting the scientific standards for foundational validity.”
- Footwear analysis: no appropriate empirical studies to support linking a print to a specific shoe.
- Microscopic hair comparison: demonstrated to be foundationally invalid.
Defense attorneys now routinely cite both reports when challenging forensic evidence, and some courts have begun excluding or limiting testimony from disciplines PCAST found invalid.
Cognitive Bias and Countermeasures
One persistent criticism of forensic science is that analysts face the same cognitive biases as everyone else. When an examiner knows the suspect confessed, or has read a detective’s notes pointing to a conclusion, that context can unconsciously shape how they interpret ambiguous evidence. The NAS report flagged this as a systemic problem in pattern-comparison disciplines.
The main countermeasure gaining traction is a procedure called sequential unmasking. The analyst examines and documents the crime-scene evidence before ever seeing the suspect’s reference sample. A case manager who knows the full context of the investigation controls the workflow, while the analyst stays shielded from details like the suspect’s criminal history or whether there was a confession. Only after the analyst has independently documented their findings on the unknown evidence does the reference sample get revealed for comparison.
Blind proficiency testing offers another layer. Instead of knowing they are being tested, analysts receive test samples inserted into regular casework, so they handle them exactly as they would any case. Both the NAS report and the American Society of Crime Laboratory Directors have recommended blind proficiency testing as a more realistic measure of examiner accuracy than open tests where the analyst knows the results will be evaluated.
Chain of Custody and Laboratory Standards
Even reliable methods can be undermined by sloppy handling. The chain of custody is a continuous documented record of every person who touches an item of evidence, including the date, time, and description of the transfer.7Global Forensic Justice Center. Chain of Custody Any gap gives the defense an opening to argue the evidence was tampered with or contaminated.
Inside the laboratory, technicians wear protective gear so their own biological material does not mix with samples, instruments are calibrated on a regular schedule, and bench notes record every chemical used and every parameter of every test. Many laboratories also maintain elimination databases containing DNA profiles of their own staff and investigators, so that a stray profile appearing in a sample can be checked against the people who handled it.
Laboratories seeking to demonstrate competence pursue accreditation under ISO/IEC 17025, an international standard covering procedures, equipment, and personnel qualifications. For DNA laboratories specifically, the FBI’s Quality Assurance Standards impose additional requirements, including external proficiency testing twice per year in each DNA technology an analyst uses in casework.8Federal Bureau of Investigation. Quality Assurance Standards for Forensic DNA Testing Laboratories
DNA Databases: How CODIS Works
The FBI’s Combined DNA Index System, known as CODIS, is the backbone of forensic DNA identification in the United States. At its top level sits the National DNA Index System (NDIS), which contains DNA profiles contributed by federal, state, and local forensic laboratories.9Federal Bureau of Investigation. CODIS and NDIS Fact Sheet The system compares unknown profiles from crime scenes against indexed profiles of convicted offenders and, in states that authorize it, arrestees. A confirmed match can establish probable cause to obtain an additional sample from the suspect for independent verification. CODIS also links unsolved cases to each other: when DNA from two scenes matches, investigators know the same person was likely involved, even before a suspect is identified.
No names or personal identifiers are stored in CODIS itself. The system holds the DNA profile and an identifier linking it back to the originating laboratory, which keeps the person’s identity separately.
Wrongful Convictions and Post-Conviction DNA Testing
The consequences of flawed forensic science are not theoretical. Misapplied forensic evidence has contributed to more than half of wrongful conviction cases overturned through DNA testing. These cases typically involved examiners who overstated the significance of pattern evidence, used methods that have since been discredited, or presented conclusions that went beyond what the science actually supported.
Federal law provides a route for convicted defendants to request DNA testing of evidence that was not tested at trial or to request retesting using newer methods. Under 18 U.S.C. ยง 3600, a person sentenced to imprisonment or death for a federal offense can file a motion for post-conviction DNA testing if they assert actual innocence under penalty of perjury, the specific evidence was secured in connection with the case, and the proposed testing could produce new evidence raising a reasonable probability that the applicant did not commit the offense.10Office of the Law Revision Counsel. 18 USC 3600 – DNA Testing The evidence must still be in the government’s possession with an intact chain of custody, and if the applicant was convicted at trial, the identity of the perpetrator must have been at issue. Most states have parallel statutes, though eligibility requirements vary, and access is generally more restricted for defendants who pleaded guilty.
Federal law also requires the government to preserve biological evidence secured during the investigation or prosecution of a federal offense for as long as the defendant remains imprisoned. The statute defines biological evidence broadly to include sexual assault examination kits, blood, saliva, hair, and skin tissue. Anyone who knowingly destroys, alters, or tampers with such evidence, intending to prevent DNA testing, faces up to five years in federal prison.11Office of the Law Revision Counsel. 18 USC 3600A – Preservation of Biological Evidence Without preserved evidence, post-conviction testing is impossible, which is why the law makes deliberate destruction a crime.