What Happens When Experts Don’t Show Their Work

Many may remember early math teachers repeatedly instructing students to “show their work.”[1]  The concept is not limited to math.  Expert witnesses also need to show their work, assuming they did any.  Consider this example from a generic slip and fall case I once handled.

Plaintiff timely disclosed an initial expert to testify on her behalf.[2]  He tested the floor and asserted it was insufficiently slip resistant when wet.[3]  He “researched the locally adopted Building Codes as they relate to the incident,” but cited none because apparently none applied.  He concluded only that “Building Codes dictate that the floor should be slip resistant.”[4]  Without citing any standard, he concluded the “average results of the slip testing of 0.16 determined that the floor surface is well below the threshold of safety of 0.50 when wet for the XL VIT.”

  • Problem 1: The slip resistance opinions were unsupported.

A qualified witness may offer opinion testimony if “(b) the testimony is based on sufficient facts or data; (c) the testimony is the product of reliable principles and methods; and (d) the expert has reliably applied the principles and methods to the facts of the case.”[5] Rule 26(a)(2) requires a party designating a retained expert witness to provide “(i) a complete statement of all opinions the witness will express and the basis and reasons for them; [and] (ii) the facts or data considered by the witness in forming them…”  “It is the proponent of the expert who has the burden of proving admissibility.”[6]

Simply put, there was no way to assess whether the expert’s slip resistance opinions were the product of sufficient facts and data, the product of reliable principles and methods, or if he reliably applied those principles and methods.  Why?  He never identified the principles or methods he used.  He cited not even one standard that he followed to gather data or form his opinions.  The report’s statement that “Building Codes dictate that the floor should be slip resistant” was useless because he never identified an applicable code.

Data and opinions relying upon them cannot be reliable if there is no way to determine if the methods and principles utilized were proper and reliably applied.

  • Problem 2: 0.5 is a myth.

The expert claimed a floor with an average reading below 0.5 when wet is below “the threshold of safety.”  I encounter this argument frequently and have discussed it before.[7]  As usual, the expert never identified a standard supporting his claim, so the only thing making it a “standard” is the expert’s own opinion.  That itself is a problem.  Further, even if the expert’s statement that “Building Codes dictate that the floor should be slip resistant” was correct, his own measurements indicated the floor had slip resistance when wet, it was just more slip resistant when dry.

  • Problem 3: Even if supported, the slip resistance opinions were unreliable because the expert failed to verify his device was calibrated.

The expert tested the floor using an XL Variable Incidence Tribometer.  As to that device, he stated it “was calibrated on 06/20/2017 and was still within the timeframe of the annual manufacture calibration.”  That statement was inadequate to validate data this device gathered.

As a prior post discussed,[8] for a tribometer to be calibrated per ASTM F2508, (1) it must rank the coefficient of friction for each of four reference surface tiles in the correct order; and (2) it must produce statistically significant results, using the mean and standard deviation, for all adjacently ranked surface tiles.[9]  If the tribometer passes, the manufacturer then creates a validation report noting the 95% confidence interval for each reference surface tile.[10]  The device is then returned to the end user.  Before the end user uses the tribometer in the field, he must first verify the tribometer is still properly calibrated.  This requires re-testing the four reference surface tiles and comparing the results to the 95% confidence interval in the validation report.[11]  If the results are not within the accepted range, the device is out of calibration and unreliable.[12]

The expert’s assertion that his device “was still within the timeframe of the annual manufacture calibration” didn’t satisfy ASTM F2508’s requirements for tribometer calibration.  He gave no indication that he tested the tribometer against the four reference surface tiles in the field before then testing the floor.  The consequence was the expert had no reference to assess if the data he obtained from the floor were valid.

  • Problem 4: A supplemental report could not fix the problem.

After these problems were noted, Plaintiff disclosed a supplemental report.  It relied upon 1) a December 19, 2013 promotional newsletter; 2) a study published in 2006 about measuring the co-efficient of friction in young adults; and 3) a study published in 1999 about bias in tribometers.  Since this information was not “new” in any sense of the word, I first argued the report was an improper supplement.[13]

But even if considered on its merits, the materials the supplemental report cited did not fix the reliability problems.  It first quoted heavily from a December 19, 2013 promotional newsletter sent by the company that manufactured this tribometer.  Nearly three pages of the supplemental report were cut and pasted directly from the newsletter.  In a later motion, I created a table showing the cut/paste job. 

The expert gave no indication that the promotional newsletter was subjected to peer-review or is otherwise scientifically reliable.  The newsletter itself did not indicate reliability as pages 3-5 contained sales information about the company’s services and products.  Finally, one paragraph the expert specifically omitted from his rebuttal report acknowledges there is no definite standard.

Anyone who represents that a measured slip resistance of 0.48 is dangerous, or that a measured slip resistance of 0.52 will prohibit slip and fall injury events must re-educate themselves to understand that the concept of safety deals with probabilities and acceptable levels of risk, and is not absolute or finite. People will still slip, fall and be injured even if the available slip resistance exceeds the criteria for safety for the given circumstances, however rarely that might occur. Conversely, not everyone will slip, fall and be injured at values less than the criteria for safety, however too many will.

The expert then cited a 2006 article.[14] He did not provide this article, but his description of it indicated it did not apply.  He claims the study “demonstrated that knowledge of a person’s uCOF and the available slip resistance … can be used to predict the probability of a slip event during level walking in young adults.”  But the plaintiff in the case was born in 1964.  She was not a “young adult” when she fell, nor was she in 2006 when the study was published.  The expert did not explain how a study about young adults would apply to a female in her 50s.  By its own terms, the study had no application to the case.

The supplemental report concluded by citing a 1999 study about variability between tribometers.[15]  However, the expert did not assert the article established 0.5 as an industry standard.  He also cited the book published by the inventor of the tribometer he purchased,[16] but did not assert it established 0.5 as an industry standard either.

In effect, the supplemental report did nothing to demonstrate the reliability of the data, principles, or methods utilized to form the opinions.

  • What happened?

A motion in limine was filed, but the case resolved before a ruling.  Maybe I was right, maybe I was wrong.  The overriding point is experts need to show their work.  Some might have deposed the expert, but I didn’t.  Why take a deposition and give the opponent an opportunity to talk his way out of your motion?  Plaintiff had the burden to demonstrate admissibility.  If the report didn’t do that, then that was Plaintiff’s problem to fix, not mine.

[1] That phrase annoyed me at the time.  If I know the answer, and the answer is correct, then why do I need to show the work?  Obviously I get it now, but ah to be young and headstrong.

[2] As usual, the names of those involved have been removed because they are irrelevant to this blog’s purpose.

[3] I love that opinion: floors may be slippery when wet.  If that doesn’t earn a Nobel Prize, I don’t know what could.

[4] Id.

[5] Rule 702(b)-(d).

[6] Lust v. Merrell Dow Pharms., Inc., 89 F.3d 594, 598 (9th Cir. 1996).

[7] February 20, 2018.

[8] October 6, 2016.

[9] ASTM F2508 at § 9. 

[10] Id. at § 10.1.

[11] Id. at § 13.2.

[12] Id. at § 13.3.

[13] A series of posts explains why, Supplemental Expert Reports.

[14] J. M. Burnfield, et. al., Prediction of Slips: An Evaluation of Utilized Coefficient of Friction and Available Slip Resistance, 49 Ergonomics 982 (2006).

[15] Powers, C. et. al., Repeatability and Bias of Two Walkway Safety Tribometers, 27 Journal of Testing and Evaluation 368 (1999).

[16] William English, Pedestrian Slip Resistance: How to Measure It and How to Improve It (2d. ed. 2003).