Torts I, Pages 290–296

Daubert v. Merrel Dow Pharmaceuticals, Inc.

United States Court of Appeals, Ninth Circuit, 1995


Plaintiffs were born with limb defects after their mothers' took Bendectin while pregnant. Their expert witnesses claim this was the cause, but this is is not statistically proved, and no other scientists believe this.

Procedural History:

District court granted summary judgment, this court affirmed, and Supreme Court reversed with a new standard to apply.


Is expert testimony admissible when it runs contrary to popular science?


  1. Determine whether the experts' testimony reflects "scientific knowledge," whether their finding are "derived by the scientific method," and whether their work product amounts to "good science."
  2. Ensure that the proposed expert testimony is "relevant to the task at hand."
    • i.e. that it logically advances a material aspect of the proposing party's case.
    • The "fit" requirement.


Some factors to guide this with:

  1. Whether the research was conducted independently of the litigation.
  2. Whether it's been peer-reviewed.


Experts did not conduct it independently or peer review it, and research does not show that it would have made it more likely than not that the injury was a cause thereof.


Not if it fails the Daubert test and doesn't show a preponderance of the evidence. Affirmed.


Statistically evidence must show that the risk of the damage more than doubled to satisfy the "but for" test.