# Statistical Extrapolation in Complex Litigation: Science or Guesswork?

## Article excerpt

Introduction

When used appropriately, statistical sampling can be a powerful weapon in a litigator's arsenal. In a complex case, statistical sampling may be used to draw scientifically-supported conclusions from data, and to quantify concepts that would otherwise be amorphous, with the added bonus that it may be utilized relatively inexpensively.2On the other hand, statistical sampling has its limitations: the science can be difficult to explain, the results can only be presented as ranges which may seem imprecise to fact-finders, and the scientific methodology can be difficult to apply correctly.3

When it comes to using statistical extrapolation to calculate damages, there is no clear-cut rule of common law allowing or disallowing its use across the federal courts of the United States. Some courts cite the Seventh Amendment as authority to disallow the use of statistics in calculating actual damages, since a statistical approach may deprive the defendant of their constitutional right to have every contended fact tried by a jury.4 Other courts opine that using statistics does not run afoul of the Seventh Amendment, so long as the defendant has the opportunity to challenge or rebut the statistical evidence.5The use of statistical extrapolation is already prevalent in several areas of litigation, and this paper will explore and analyze that use, and propose other areas of jurisprudence where the use of statistical sampling might be advisable. This paper will evaluate the differing treatment of statistical extrapolation between jurisdictions, the reasoning behind those opinions, and the public policies for and against using statistical extrapolation in this fashion. Finally, the paper recommends a course of action for the courts (or Congress) with regard to the use of statistical extrapolation to calculate damages.

Background on Statistical Extrapolation

Statistical extrapolation is "[t]he process of estimating an unknown value or quantity on the basis of the known range of variables" or "[t]he process of speculating about possible results, based on known facts."6 This process can be applied to any type of data, including calculation of damages in civil litigation. Courts have allowed sampling to calculate damages in a variety of circumstances, including political district apportionment cases, pornography prosecutions, drug prosecutions, labor and employment litigation and others.7 In cases that can include hundreds, or thousands, of potential claims within a single lawsuit, a claim-by-claim presentation at trial may simply be untenable and overly tedious, if not logistically impossible.8Alternatively, class actions that are comprised of a large number of class members and where compensatory damages need to be determined on an individual basis could be good candidates to allow sampling to determine the damages in a more efficient manner. In such instances, where the number of claims within a given case is simply overwhelming administratively and logistically, statistical sampling provides an alternative to claim-by-claim proof by providing a means of estimating with reasonable certainty the aggregation of large quantities of data without having to measure each individual data point. Accordingly, many courts accept and encourage the use of statistical sampling in certain types of cases.

Federal courts have repeatedly held that the government may use statistical sampling methods when calculating claims overpayments or actual damages in health care fraud and abuse cases. In holding that the U.S. Department of Health and Human Services enjoyed this authority, the Ninth Circuit Court of Appeals found extrapolation based on statistical sampling to be reasonable where the defendant provider had the opportunity to challenge the statistical sampling methodology.9 The other federal circuit courts have held similarly.10 These cases hold that a statistical sampling-based estimate of overpayment creates a rebuttable presumption that defendant providers can challenge through introduction of countervailing or claim-by-claim evidence if the provider chooses. …

Search by...
Show...

### Oops!

An unknown error has occurred. Please click the button below to reload the page. If the problem persists, please try again in a little while.