Concept
Reliable Application to Facts
Reliability and relevancy
Reliable Application to Facts
A sound method does not guarantee a sound opinion. Rule 702 addresses that gap in its last condition, which requires that “the expert’s opinion reflects a reliable application of the principles and methods to the facts of the case.”1 Two earlier conditions examine different things. Subsection (c) asks whether the testimony is the product of reliable principles and methods,2 and subsection (b) asks whether it is based on sufficient facts or data.3 An expert can satisfy both, with a validated technique and a full set of data, and still fail subsection (d), either because the technique was not carried out properly or because the conclusion reaches further than the technique allows. “An unreliable application of a reliable methodology results in an unreliable opinion,” so the gatekeeper must look to how the expert applied the methodology, and not only to which methodology the expert chose.4
Application can fail in two ways, and each has its own subsection below. The expert may have performed the method incorrectly in this case, which raises the questions of what the field requires and who decides whether the expert met it. Or the expert may have performed the method correctly and then announced a conclusion stronger than the method supports. The first failure is a defect in the work; the second is a defect in the testimony.
Following Proper Procedure
A technique that works in the field’s hands can still fail in one expert’s hands. Proficiency studies show the pattern. It is dangerous to read too much into them, since their difficulty varies and analysts sometimes know beforehand that their work is being evaluated, but in some instances in which a proficiency test identified analyst errors, the follow-up investigation found that the root cause was faulty test procedure.5 A valid methodology and trustworthy information about the case do not, by themselves, ensure that the test was performed properly. DNA comparison makes the point concrete: a technician who puts DNA from a single sample in two lanes of the gel obtains a perfect but false match.6 The technique is accepted, and the mistake lies in the execution.
Kumho Tire offers a glimpse of the same concern. The tire expert there had based his initial conclusions on photographs, and the Court noted that its reservations about his methodology would have been further augmented by the fact that he first inspected the tire itself on the morning of his first deposition, and then only for a few hours.7 Hands-on work that late and that brief does not by itself prove an unreliable application, but it is the kind of fact a gatekeeper may weigh when asking whether the method was actually carried out on the object in dispute.
The judge, not the jury, decides whether the expert followed proper procedure. The question belongs to the judge under Rule 104(a), and the jury does not determine it as a matter of “weight.” The same issues still affect weight, however, so the jury can take another look at them once the testimony is admitted.8 That sequence is the point of gatekeeping. The judge screens first, before the jury hears the conclusion, and the jury then weighs whatever flaws survive the screen.
The field supplies the yardstick. One Daubert factor already asks whether there are accepted standards for using the methodology,9 and the practical question is where those standards live. For a commercial instrument such as a particular gas chromatograph/mass spectrometer, the manufacturer is usually the best starting place, because it usually prepares an operations and maintenance manual. The American Academy of Forensic Sciences has a Standards Board, and the International Standardization Organization (ISO) and the American Society for Testing and Materials International (ASTM) develop consensus standards for industrial and forensic techniques. ASTM committee E-30 is tasked specifically with standards for the forensic sciences, so an ASTM standard might apply.10
The requirement has a history of challenge. Just as the original provisions of Article VII were silent on a general acceptance test, they said nothing about a requirement for a showing of proper test procedure. The argument ran that just as Rule 702 displaced Frye, it impliedly superseded any common-law requirement for a showing of proper test procedure.11 The text of the rule now answers that argument. Since the 2000 amendment, which was meant to capture the approach of Daubert and Kumho Tire, an expert may testify only if the expert “has reliably applied the principles and methods to the facts of the case.”12
An opponent who contests procedure can offer evidence, including expert testimony, that the proponent’s expert employed an unsound methodology or failed to follow an otherwise sound protocol assiduously. If that raises a material dispute about the admissibility of the expert evidence, one court has held, the district court must hold an in limine hearing, a so-called Daubert hearing, to consider the conflicting evidence and make findings about the soundness and reliability of the methodology.13 The burden section below takes up how the burden and standard of proof operate at that hearing.
Exam stems on this topic usually pair an accepted technique with a lapse in performing it: a standard test run without a required control, an analyst who departs from a published standard, a gel loaded incorrectly. Two answers tempt. One admits the opinion because the technique is accepted, which answers the subsection (c) question and leaves subsection (d) untouched. The other excludes the opinion for any deviation at all, or sends every deviation to the jury as impeachment. The sound answer identifies the standard the field supplies, asks whether the method was actually carried out properly in this case, treats the judge as the decision-maker, and leaves the jury whatever imperfection survives that ruling.
Staying Within the Method
A method can be sound and its execution flawless, and the expert can still say more than the method shows. In many situations the problem is overstatement: the expert expresses a confidence the methodology does not support.15 A firearms examiner who testifies that his rate of error is “zero,” or that the bullet he tested “matches” the bullet found at the scene, overstates, because subjective judgment carries a rate of error and no one has established that each gun makes a unique marking on a bullet. The expert therefore cannot validly testify to a “match.”16 Likewise, a ballistics expert who calls himself a “scientist” overstates the support for his conclusion, because ballistics is not a scientific endeavor.17
Rule 702(d) is written to catch this, and its subject is the opinion. The question is whether “the expert’s opinion reflects a reliable application of the principles and methods to the facts of the case.”18 The text adopted in 2000 asked instead whether “the expert has reliably applied the principles and methods to the facts of the case.”19 Both versions demand reliable application, but the current wording makes the opinion itself the thing that must reflect it. A rule that asks whether the expert applied the method reliably can seem satisfied by proof that the steps were followed. A rule that asks whether the opinion reflects that application sends the judge to the statement the jury will actually hear. Because one expert often gives several opinions, the judge can test them one at a time, admitting those the method supports and limiting those it does not.
This is not a license to exclude a conclusion because it seems wrong. In Daubert the Court directed that the inquiry “must be solely on principles and methodology, not on the conclusions that they generate.”20 The comparison here runs between the opinion and the support the method supplies, and not between the opinion and the truth. An opinion can be correct and still claim more than the method showed, and an opinion can be modest, well supported, and later proved mistaken.
Joiner shows how an opinion can outrun its basis. Rather than explaining how and why the experts could have extrapolated their opinions from seemingly far-removed animal studies, the respondent chose “to proceed as if the only issue [was] whether animal studies could ever be a proper foundation for an expert opinion.” That was not the issue. The issue was whether the experts’ opinions were sufficiently supported by the animal studies on which they purported to rely.21 The lesson reaches beyond animal studies. Showing that a kind of basis is acceptable does not show that this opinion is supported by this basis, and a proponent who argues only the former leaves the real question unanswered.22
Courts often answer overstatement by confining the expert, not by excluding the whole field. Some federal trial courts have barred testifying handwriting experts from giving opinions about authorship and limited them to describing similar features and characteristics of the disputed writings, as in United States v. Hidalgo. Others have excluded handwriting experts altogether under Daubert, as in Almeciga v. Center for Investigative Reporting, which reviewed the history of handwriting analysis in the courtroom. This reexamination has been driven in part by a 2009 comprehensive report of the National Academy of Sciences that evaluated several fields of forensic science, including handwriting analysis.23 The narrower order separates describing similarities, which claims little, from naming an author, which these courts treated as beyond what the testimony could support.
The stated degree of confidence can also depend on the governing formula. Courts have traditionally required medical testimony to be stated with medical certainty, and some modern courts have backed away from that tradition. In People v. Ramirez, the Colorado Supreme Court abandoned the medical-certainty standard. Colorado Rule 702 governs and turns on reliability, so even an opinion stated with something “less than certainty” is admissible.24 The point for application is that the label matters less than the match between the confidence the expert states and the support behind it.
Two answers go wrong on this subject. One treats the court’s approval of a technique as approval of everything the expert says about it, and the other excludes the whole witness because a single statement overreaches. A sound answer takes each opinion, matches it to the support behind it, and limits whatever outruns that support. The next section takes up who must make that showing and by what standard.