Loading
Legal Standards
8 min read
Published Date:
Last Modified Date:
“Reasonable degree of scientific certainty” is a common expert-witness phrase, but it has no universally accepted scientific definition and is not required by Federal Rule of Evidence 702. This article explains how courts evaluate certainty language, why the phrase has drawn criticism in forensic practice, how it differs from standards such as reasonable medical probability, and what attorneys should examine when assessing whether an expert’s stated confidence is supported by reliable methods and evidence.
"Reasonable degree of scientific certainty" is a phrase used in expert witness testimony to describe confidence in a scientific or technical conclusion. However, the phrase has no generally accepted scientific definition, and Federal Rule of Evidence 702 does not require it for admissibility.
Under Federal Rule of Evidence 702, courts evaluate an expert's qualifications, factual basis, methods, and reliable application of those methods. Simply describing an opinion as scientifically certain does not, by itself, establish that Rule 702's requirements are met.
This guide explains how courts treat reasonable scientific certainty, its limitations, and what attorneys should examine when evaluating an expert's stated confidence.
Federal Rule of Evidence 702 does not require an expert to express an opinion "to a reasonable degree of scientific certainty." The rule requires the witness to be qualified by knowledge, skill, experience, training, or education. The proponent must also demonstrate to the court, by a preponderance of the evidence, that the proposed testimony meets the following requirements:
The expert's specialized knowledge will help the factfinder understand the evidence or determine a fact in issue.
The testimony rests on sufficient facts or data.
The testimony is the product of reliable principles and methods.
The expert's opinion reflects a reliable application of the principles and methods to the facts of the case.
The December 1, 2023 amendment clarified the burden for admitting expert testimony and emphasized that an expert's conclusions must remain within what the factual basis and reliably applied methodology can support.
This preponderance standard concerns admissibility. It does not require every expert opinion to express a probability greater than 50%.
In Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993), the Supreme Court of the United States established the trial judge's gatekeeping role in assessing scientific reliability and relevance. The Court identified flexible considerations such as testing, peer review, error rates, standards, and general acceptance.
Neither Rule 702 nor Daubert makes a particular certainty phrase a substitute for reliable analysis.
Uncertainty quantification
Evidence Analysis
Forensic Analysis
Causation Analysis
Litigation Support
"Reasonable degree of scientific certainty" is a phrase used in expert witness testimony to describe confidence in a scientific or technical conclusion. However, the phrase has no generally accepted scientific definition, and Federal Rule of Evidence 702 does not require it for admissibility.
Under Federal Rule of Evidence 702, courts evaluate an expert's qualifications, factual basis, methods, and reliable application of those methods. Simply describing an opinion as scientifically certain does not, by itself, establish that Rule 702's requirements are met.
This guide explains how courts treat reasonable scientific certainty, its limitations, and what attorneys should examine when evaluating an expert's stated confidence.
Federal Rule of Evidence 702 does not require an expert to express an opinion "to a reasonable degree of scientific certainty." The rule requires the witness to be qualified by knowledge, skill, experience, training, or education. The proponent must also demonstrate to the court, by a preponderance of the evidence, that the proposed testimony meets the following requirements:
The expert's specialized knowledge will help the factfinder understand the evidence or determine a fact in issue.
The testimony rests on sufficient facts or data.
The testimony is the product of reliable principles and methods.
The expert's opinion reflects a reliable application of the principles and methods to the facts of the case.
The December 1, 2023 amendment clarified the burden for admitting expert testimony and emphasized that an expert's conclusions must remain within what the factual basis and reliably applied methodology can support.
This preponderance standard concerns admissibility. It does not require every expert opinion to express a probability greater than 50%.
In Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579 (1993), the Supreme Court of the United States established the trial judge's gatekeeping role in assessing scientific reliability and relevance. The Court identified flexible considerations such as testing, peer review, error rates, standards, and general acceptance.
Neither Rule 702 nor Daubert makes a particular certainty phrase a substitute for reliable analysis.
The phrase can suggest a recognized scientific threshold even though different experts may interpret it differently.
In 2016, the National Commission on Forensic Science recommended that the Attorney General direct DOJ attorneys to avoid the phrase when presenting forensic testimony. It also recommended that DOJ-employed forensic science and medical providers avoid it in reports and testimony. Both recommendations allowed exceptions where a court requires the language.
The Commission explained that the phrase lacks scientific meaning and can mislead factfinders about methodological reliability and limitations.
Following the recommendation, the DOJ directed its forensic laboratories to review their policies and procedures to ensure examiners avoided expressions such as "reasonable scientific certainty" in reports or testimony. DOJ prosecutors were also instructed to avoid that terminology when presenting forensic reports or asking forensic experts questions in court, except where a judge or applicable law required its use.
The DOJ instructions were departmental policies, not a new evidentiary rule governing expert testimony generally.
The 2023 Advisory Committee Note to Federal Rule of Evidence 702 also advises forensic experts to avoid claiming absolute certainty or a reasonable degree of scientific certainty when their methods are subjective and potentially subject to error.
Courts have distinguished between an expert's choice of words and the reliability of the underlying opinion.
In State v. DeLeon, 131 Hawaiʻi 463, 319 P.3d 382 (2014), the Supreme Court of Hawaiʻi reviewed the exclusion of expert testimony about the shooting victim's cocaine use.
The trial court required the expert to express an opinion to a reasonable degree of scientific certainty rather than scientific probability.
The Supreme Court held that the trial court plainly erred in excluding the testimony based on that requirement. Expert opinions did not need that particular formulation to be admissible under Hawaiʻi law, although speculative testimony could still be excluded.
In United States v. Glynn, 578 F. Supp. 2d 567 (S.D.N.Y. 2008), the United States District Court for the Southern District of New York considered firearms-comparison testimony.
The court determined that the methodology did not support the examiner's claimed degree of ballistic certainty. In its order governing the retrial, it limited the examiner to expressing an identification opinion as "more likely than not."
That restriction addressed the testimony before the court. It did not establish a universal probability threshold for forensic identification.
Reasonable scientific certainty and reasonable medical probability should not automatically be treated as interchangeable expressions or legal requirements.
In medical causation disputes, governing substantive law may require affirmative causation opinions to meet a reasonable medical probability standard.
California Evidence Code section 801.1 concerns general civil cases as defined by Rule 1.6 of the California Rules of Court. It applies when the party bearing the burden of proof offers expert medical-causation testimony and that party's expert is already required to express a causation opinion to a reasonable medical probability as a condition of testifying.
In those cases, the party not bearing the burden of proof may offer an expert who proposes alternative causes only if the expert can opine that each cause exists to a reasonable medical probability, subject to the exception in subdivision (b).
Subdivision (b) adds that subdivision (a) does not stop an expert from testifying that a matter cannot meet a reasonable degree of probability in the applicable field, so long as the expert explains the basis for that opinion.
These requirements apply only to the proceedings and medical causation opinions covered by section 801.1, not to all medical experts or opinions offered in litigation.
The 2023 Advisory Committee Note to Federal Rule of Evidence 702 also explains that the amendment does not itself prevent experts from expressing opinions to the degree of certainty required by governing substantive law.
An expert's stated confidence should reflect the evidence, methodology, and limitations underlying the opinion.
Depending on the discipline, a suitably qualified expert may examine laboratory findings, medical records, physical evidence, technical standards, scientific literature, or statistical data.
Attorneys reviewing the analysis should consider:
Supporting evidence: What facts, measurements, observations, or studies support the conclusion?
Methodological reliability: Has the technique been tested or validated for its intended purpose?
Error and uncertainty: Are relevant error rates, measurement limitations, or statistical uncertainties identified?
Alternative explanations: Has the expert addressed competing explanations where appropriate?
Scope of the conclusion: Does the opinion extend beyond what the methods and available information support?
For example, a toxicologist evaluating a disputed chemical exposure may review laboratory measurements and research examining exposure levels and health effects. Evidence supporting a conclusion about exposure does not necessarily support an opinion that the exposure caused a particular medical condition.
Similarly, a statistical expert may quantify sampling uncertainty when appropriate data and methods are available. That analysis does not automatically establish the probability that a disputed legal allegation is true.
Expert reports should explain the reasoning supporting an opinion rather than depend on an unexplained declaration of certainty.
In federal civil litigation, Federal Rule of Civil Procedure 26(a)(2)(B) generally requires written reports from retained or specially employed testifying experts and from employees of a party whose duties regularly involve giving expert testimony, unless otherwise stipulated or ordered by the court.
Those reports must contain a complete statement of all opinions the expert will express, their bases and reasons, and the facts or data considered, among other required information.
During depositions and trial, attorneys should clarify whether terms such as "highly probable," "consistent with," or "scientifically certain" reflect an empirically supported probability or a qualitative professional judgment. They should also ask the expert to explain relevant limitations.
Confidence alone does not establish reliability. Nor does acknowledging uncertainty automatically make otherwise reliable testimony inadmissible.
Does reasonable scientific certainty mean an expert is 100% certain?
The phrase does not, by itself, establish absolute certainty or specify a universally accepted numerical probability. Its meaning depends on how the expert uses it and what the supporting methodology establishes.
Can an expert testify without expressing scientific certainty?
In federal court, yes. Rule 702 does not require the phrase. Expert testimony must still satisfy applicable admissibility requirements, and governing state law or other substantive law may require an opinion to meet a specified degree of certainty or probability.
Can a court limit an expert's certainty language?
Yes. A court may restrict conclusions expressed with greater confidence than the underlying evidence and methodology reliably support. The determination depends on the applicable rules and the circumstances of the case.
Does acknowledging uncertainty weaken expert testimony?
Not necessarily. Identifying uncertainty can accurately communicate a method's limitations. Insufficient factual support or unreliable application of a methodology can prevent admission under Rule 702. Once the admissibility requirements are satisfied, remaining uncertainty may affect the weight assigned to the testimony.
The phrase can suggest a recognized scientific threshold even though different experts may interpret it differently.
In 2016, the National Commission on Forensic Science recommended that the Attorney General direct DOJ attorneys to avoid the phrase when presenting forensic testimony. It also recommended that DOJ-employed forensic science and medical providers avoid it in reports and testimony. Both recommendations allowed exceptions where a court requires the language.
The Commission explained that the phrase lacks scientific meaning and can mislead factfinders about methodological reliability and limitations.
Following the recommendation, the DOJ directed its forensic laboratories to review their policies and procedures to ensure examiners avoided expressions such as "reasonable scientific certainty" in reports or testimony. DOJ prosecutors were also instructed to avoid that terminology when presenting forensic reports or asking forensic experts questions in court, except where a judge or applicable law required its use.
The DOJ instructions were departmental policies, not a new evidentiary rule governing expert testimony generally.
The 2023 Advisory Committee Note to Federal Rule of Evidence 702 also advises forensic experts to avoid claiming absolute certainty or a reasonable degree of scientific certainty when their methods are subjective and potentially subject to error.
Courts have distinguished between an expert's choice of words and the reliability of the underlying opinion.
In State v. DeLeon, 131 Hawaiʻi 463, 319 P.3d 382 (2014), the Supreme Court of Hawaiʻi reviewed the exclusion of expert testimony about the shooting victim's cocaine use.
The trial court required the expert to express an opinion to a reasonable degree of scientific certainty rather than scientific probability.
The Supreme Court held that the trial court plainly erred in excluding the testimony based on that requirement. Expert opinions did not need that particular formulation to be admissible under Hawaiʻi law, although speculative testimony could still be excluded.
In United States v. Glynn, 578 F. Supp. 2d 567 (S.D.N.Y. 2008), the United States District Court for the Southern District of New York considered firearms-comparison testimony.
The court determined that the methodology did not support the examiner's claimed degree of ballistic certainty. In its order governing the retrial, it limited the examiner to expressing an identification opinion as "more likely than not."
That restriction addressed the testimony before the court. It did not establish a universal probability threshold for forensic identification.
Reasonable scientific certainty and reasonable medical probability should not automatically be treated as interchangeable expressions or legal requirements.
In medical causation disputes, governing substantive law may require affirmative causation opinions to meet a reasonable medical probability standard.
California Evidence Code section 801.1 concerns general civil cases as defined by Rule 1.6 of the California Rules of Court. It applies when the party bearing the burden of proof offers expert medical-causation testimony and that party's expert is already required to express a causation opinion to a reasonable medical probability as a condition of testifying.
In those cases, the party not bearing the burden of proof may offer an expert who proposes alternative causes only if the expert can opine that each cause exists to a reasonable medical probability, subject to the exception in subdivision (b).
Subdivision (b) adds that subdivision (a) does not stop an expert from testifying that a matter cannot meet a reasonable degree of probability in the applicable field, so long as the expert explains the basis for that opinion.
These requirements apply only to the proceedings and medical causation opinions covered by section 801.1, not to all medical experts or opinions offered in litigation.
The 2023 Advisory Committee Note to Federal Rule of Evidence 702 also explains that the amendment does not itself prevent experts from expressing opinions to the degree of certainty required by governing substantive law.
An expert's stated confidence should reflect the evidence, methodology, and limitations underlying the opinion.
Depending on the discipline, a suitably qualified expert may examine laboratory findings, medical records, physical evidence, technical standards, scientific literature, or statistical data.
Attorneys reviewing the analysis should consider:
Supporting evidence: What facts, measurements, observations, or studies support the conclusion?
Methodological reliability: Has the technique been tested or validated for its intended purpose?
Error and uncertainty: Are relevant error rates, measurement limitations, or statistical uncertainties identified?
Alternative explanations: Has the expert addressed competing explanations where appropriate?
Scope of the conclusion: Does the opinion extend beyond what the methods and available information support?
For example, a toxicologist evaluating a disputed chemical exposure may review laboratory measurements and research examining exposure levels and health effects. Evidence supporting a conclusion about exposure does not necessarily support an opinion that the exposure caused a particular medical condition.
Similarly, a statistical expert may quantify sampling uncertainty when appropriate data and methods are available. That analysis does not automatically establish the probability that a disputed legal allegation is true.
Expert reports should explain the reasoning supporting an opinion rather than depend on an unexplained declaration of certainty.
In federal civil litigation, Federal Rule of Civil Procedure 26(a)(2)(B) generally requires written reports from retained or specially employed testifying experts and from employees of a party whose duties regularly involve giving expert testimony, unless otherwise stipulated or ordered by the court.
Those reports must contain a complete statement of all opinions the expert will express, their bases and reasons, and the facts or data considered, among other required information.
During depositions and trial, attorneys should clarify whether terms such as "highly probable," "consistent with," or "scientifically certain" reflect an empirically supported probability or a qualitative professional judgment. They should also ask the expert to explain relevant limitations.
Confidence alone does not establish reliability. Nor does acknowledging uncertainty automatically make otherwise reliable testimony inadmissible.
Does reasonable scientific certainty mean an expert is 100% certain?
The phrase does not, by itself, establish absolute certainty or specify a universally accepted numerical probability. Its meaning depends on how the expert uses it and what the supporting methodology establishes.
Can an expert testify without expressing scientific certainty?
In federal court, yes. Rule 702 does not require the phrase. Expert testimony must still satisfy applicable admissibility requirements, and governing state law or other substantive law may require an opinion to meet a specified degree of certainty or probability.
Can a court limit an expert's certainty language?
Yes. A court may restrict conclusions expressed with greater confidence than the underlying evidence and methodology reliably support. The determination depends on the applicable rules and the circumstances of the case.
Does acknowledging uncertainty weaken expert testimony?
Not necessarily. Identifying uncertainty can accurately communicate a method's limitations. Insufficient factual support or unreliable application of a methodology can prevent admission under Rule 702. Once the admissibility requirements are satisfied, remaining uncertainty may affect the weight assigned to the testimony.
Uncertainty quantification
Evidence Analysis
Forensic Analysis
Causation Analysis
Litigation Support
Legal Standards
Legal Standards