The Acceptable “Rate of Error” in Expert Witnesses’ Opinion

police officer showing DWI test device - error rate

“Error Rate” Sobriety Tests

The “error rate”, one of the Daubert factors, requires the trial judge to evaluate the error rate with regard to whether the method has been tested, and whether the expert’s opinion has an acceptable rate of error. If the test has a high error rate, the court will have to decide whether or not to admit the expert witness’s opinion or whether the evidence is of a limited use in a particular case. Thus, the court would like to know how to decide whether the test outcome was erroneous; and thus, of no help to the jury.

The expert witness must explain to the court the importance of the error rate. Experts have recognized acceptable error rates associated with statistics and the disclosure of the actual importance of those error rates to accept the results as reliable. The case Kumho Tire Co. v. Carmichael, 526 U.S. 137, 151 (1999) states that reliability in the form of a known and an acceptable error rate is a significant consideration for admissibility.

When a method is known to have a high rate of error and has being carried out without any regulatory model, it would be deemed to lack valid scientific knowledge and possibly would be of no help to the jury, as it would concern both the reliability and the relevance of the offered expert witness’s opinion.  The case Daubert v. Merrell Dow Pharmaceuticals, Inc., 509 U.S. 579, 593 (1993) identifies “the known or potential rate of error” as one of the factors in evaluating the scientific reliability or validity of the given expert witness’s opinion. Thus, evidence of a method that has a higher than accepted error rate may not be admissible as the only evidence, but it may be admitted as circumstantial evidence if the court can be persuaded that the limited use is relevant and helpful. In addition, admissibility should not be confused with credibility or weight because it determines only whether the jury would consider the evidence. On the other hand, credibility or weight affects whether the evidence is believed or accepted by the jury.

In the case U.S. v. Eric D. Horn where Defendant was charged with driving while intoxicated (DWI), defendant moved to exclude evidence of his performance on field sobriety tests. Due to the performed “tests” substantially fell below the accepted norm for such tests, the Judge did not allow the police officer to testify that a “test” had been performed and that the defendant had failed it because of reliability concerns about the test, including its high error rate. However, the judge held that the test results were admissible on the issue of whether there was probable cause for arrest, but not for purpose of proving that the defendant had a specific blood alcohol level. Also, the arresting officer could give lay opinion testimony that defendant was driving while intoxicated based on his own perceptions of defendant’s performance of the tests, such as the driver’s balance, coordination, ability to concentrate and to follow instructions, but could not suggest that tests were direct evidence of intoxication.

Therefore, the error rate is important to the extent that it impacts its evaluation of whether the method is reliable; and thus, its outcome is helpful to the jury. Likewise the other Daubert factors, “error rate” analysis may not be dispositive since it should be measured in conjunction with all the other evidence affecting admissibility.