When can Utah courts exclude expert testimony based on unreliable data? California College v. UCN Explained

Utah courts serve as gatekeepers under Rule 702 of the Utah Rules of Evidence, screening out unreliable expert testimony before it reaches the jury. A recent Utah Court of Appeals decision in California College v. UCN demonstrates how data reliability issues can doom expert opinions, even when courts apply only a threshold standard for admissibility.

Background and Facts

InContact provided enhanced telephone services to California College and related educational entities. During a damage period from February 2006 to April 2007, the College’s offices allegedly suffered frequent telephone service disruptions that prevented inbound calls or caused improper routing. The College sued InContact for lost profits, claiming the telephone problems led to reduced student recruitment and enrollment. The College retained two experts: Ted Tatos, a statistician who used regression analysis to estimate lost student enrollments at 1,254, and Richard Hoffman, a CPA who calculated approximately $19.7 million in lost profits based on Tatos’s enrollment figures.

Key Legal Issues

The central issue was whether the experts’ opinions met the threshold reliability standard under Rule 702(b), which requires that expert testimony be based on sufficient facts or data that have been reliably applied. InContact’s rebuttal experts revealed that Tatos had relied not on actual business records, but on a “summary” prepared by one of the college owners specifically for litigation. This summary contained numerous inaccuracies and inconsistencies when compared to the underlying operational reports.

Court’s Analysis and Holding

The district court initially excluded the expert testimony as “flawed” and “insufficient” under Rule 702(b). However, after reassignment to a different judge, the court reconsidered and ruled that the data was “sufficiently reliable to satisfy the minimal threshold standard.” The Utah Court of Appeals reversed, finding this was an abuse of discretion. The court emphasized that even under Rule 702’s low threshold standard, expert opinions cannot be based on data that parties acknowledge contains “certain inaccurate information.” The court noted that regression analysis based on flawed data produces equally flawed results, as the statistical model’s “best fit” cannot reasonably approximate actual values when built on erroneous inputs.

Practice Implications

This decision reinforces that Utah courts must rigorously examine the foundational reliability of data underlying expert opinions, particularly in statistical analyses like regression modeling. Practitioners should thoroughly investigate data sources and document any inconsistencies or inaccuracies when challenging expert testimony. The decision also demonstrates that courts cannot simply assume reliability—there must be some evidentiary foundation supporting the data’s trustworthiness, even under Rule 702’s threshold standard.