Concept
Relevancy and Fit
Reliability and relevancy
Relevancy and Fit
Rule 702(a) admits expert testimony only if the expert’s specialized knowledge “will help the trier of fact to understand the evidence or to determine a fact in issue.”1 Reliability and helpfulness test different things. Reliability measures the quality of the expert’s work; helpfulness measures whether that work connects to the lawsuit and whether the jury needs it. An opinion can be sound and still useless, and the rule screens for both defects. Two inquiries carry the helpfulness condition: fit, which asks whether the testimony speaks to the issue in dispute, and the helpfulness threshold, which asks whether the jury can use the testimony at all.
Fit to the Disputed Issue
Daubert read the helpfulness language as a requirement of fit: the expert’s knowledge must have a “valid scientific connection to the pertinent inquiry.”2 Fit stands apart from reliability because validity does not travel. Knowledge that is valid for one purpose may not be valid for another, unrelated purpose, and the mismatch is not always obvious.3
On remand in Daubert, the Ninth Circuit turned that point into a rule of exclusion. Federal judges must exclude proffered scientific evidence under Rules 702 and 403 unless they are convinced that it speaks clearly and directly to an issue in dispute and will not mislead the jury.4
The Supreme Court’s own example holds the science constant and changes only the question. Sound knowledge of the moon’s phases can show whether a certain night was dark, and if darkness is a fact in issue, that knowledge helps the trier of fact. The same knowledge does not help decide whether a person was unusually likely to act irrationally that night, because, absent credible grounds supporting such a link, a full moon says nothing reliable about conduct.5 The astronomer’s method is equally good in both settings. Reliability asks whether the method is good; fit asks whether the good method answers the question the jury must decide. The qualifier matters, too. A proponent who could supply credible grounds for the link would have a valid connection, and the testimony would fit. What fit screens out is the unjustified leap from an accepted premise to an unfounded conclusion, which later cases treat as the core of the inquiry.6
The Bendectin remand applied the same inquiry to a legal standard. Causation was the pertinent inquiry, and California tort law required proof that Bendectin more likely than not caused the injuries. Statistically, that means the drug must more than double the risk of the defect. A relative risk below two may suggest that a drug can cause birth defects, yet it tends to disprove legal causation, because it shows that the drug does not double the likelihood of the defect.7 The plaintiffs’ experts were unprepared to testify that Bendectin caused the injuries, and the remaining experts would say only that it is “capable of causing” birth defects.8 That is testimony about possibility, and the governing standard demanded probability. Sound methods could not repair the mismatch.
Fit also depends on what the record supports. Testimony does not help if it is unrelated to the facts at issue or rests on factual assumptions the evidence does not support. An economist’s opinion on lost future income in a wrongful death case, for example, is irrelevant if it assumes past earnings that have no basis in fact. Testimony fails for the same reason when its reasoning is so illogical that it cannot affect the probabilities of facts at issue.9 In General Electric Co. v. Joiner, Justice Stevens, writing separately, addressed one basis of the district court’s exclusion order: the absence of evidence that Joiner was exposed to furans and dioxins. If the evidence did raise a genuine issue of fact on that exposure, as the court of appeals held, then that basis was erroneous, but not because the district judge abused her discretion or misapplied the law.10 The exposure question is a question about the record. An opinion that assumes exposure fits the case only if the evidence supports the assumption.
A reliable technique, a well-credentialed expert, and a clean methodology answer only the reliability question. Before admitting the opinion, identify the fact the jury must find, including the standard of proof that governs it, and ask whether the testimony speaks to that fact.
The Helpfulness Threshold
The second idea inside Rule 702(a) is that the testimony must actually assist the trier of fact. No expert is needed to tell a jury that a person wearing gloves leaves no fingerprints, or that compulsive gamblers characteristically mismanage their finances.12 Older cases went further and often required that the subject lie outside the reach, or “beyond the ken,” of the trier of fact, which confined experts to highly specialized topics. Modern courts require only that the testimony be helpful, as Rule 702 provides.13
The subject therefore need not be completely unfamiliar to jurors. An expert often adds depth and clarity to matters the jury knows in part. If the jury is adequately equipped to decide an issue without assistance, however, expert testimony is inadmissible.14
In practice the standard is generous, and courts have admitted a wide range of testimony under it. Approved examples include testimony on the typical structure of mail fraud schemes, fire marshals’ testimony that arson started a blaze, and a DEA agent’s description of the purity of heroin commonly sold on the street.15 The ceiling shows in the opposite example: expert testimony explaining the enormity of a physician’s blunder in amputating the wrong leg of a patient might not be admissible in a malpractice case.16 The heroin and arson witnesses bring knowledge that jurors lack, while the wrong-leg witness offers nothing a juror cannot see unaided. When expertise is only marginally helpful because the subject is simple or familiar, a decision to exclude better rests on Rule 403 than on a finding that the testimony cannot help.17