EEOC Failed Court’s Science Test

By Ryan Thompson, Esq
In an unidentified room, perhaps in the middle of Pennsylvania somewhere, five academics gather to conduct a test. One has an advanced degree in cultural anthropology, another has an advanced degree in human development, and the others have advanced degrees in education, psychology and economics, respectively. All have experience with “multicultural, multiracial, treatment outcome research” but what exactly is “multicultural, multiracial, treatment outcome research?”  No one seems to know.
The task at hand is simple, though, despite the fact that none of the academics nor the designer of this test, has ever done this before. The five academics will now become “race raters.”
The race raters will be separated from one another and each given approximately 900 color photographs from various peoples’ driver’s licenses. Though these photos will include the pictured person’s name, the race raters have been asked to ignore the name. The race raters are simply asked to look at the photograph and determine what race the person is based only on what they look like.
They will be asked to determine whether the person in the photograph is 1) African-American, (2) Asian, (3) Hispanic, (4) White, or (5) Other. If someone appears “multi-racial,” then the race raters should mark “Other.”
If four of the five race raters independently agree on a person’s race, then that 80% agreement rate will be sufficient to correctly assess that person’s race as that which the race raters have determined.
From this data, the designer of the test, Pennsylvania State University psychology professor and expert witness Dr. Kevin Murphy, believes he will be able to accurately compile the racial statistics on these 900 or so people necessary for his analysis on race and racial disparity.
In retrospect, someone asked why, if the only point of this test was to determine what race a person is, why didn’t Dr. Murphy’s team simply call the individuals in question on the phone and ask them?
This Is a Test, This Is Only a Test
In today’s “multicultural, multiracial” society, such a test may raise some racially sensitive eyebrows as being oversimplified and even stereotypical. One may even wonder whether conducting such a test would garner the attention of the U.S. Equal Employment Opportunity Commission (EEOC).
Ironically, it was the EEOC that had conducted this race-rating test. Despite publicly discouraging American employers from identifying employees and candidates visually by race, the EEOC found itself recently in court arguing that visual identification of race is both accurate and scientifically reliable, specifically when done in the manner of Dr. Murphy’s race-rater team.
The courts, however, have disagreed.
“The EEOC brought this case on the basis of a homemade methodology, crafted by a witness with no particular expertise to craft it, administered by persons with no particular expertise to administer it, tested by no one, and accepted only by the witness himself,” the U.S. Court of Appeals for the Sixth Circuit wrote in a decision earlier this month.
In EEOC v. Kaplan Higher Education Corporation, et al., the EEOC had filed suit against Kaplan, claiming that Kaplan’s practice of conducting credit checks on many of its employee candidates who may be exposed to students’ financial data was done so in a way that creates a disparate impact against African-Americans and in violation of Title VII of the Civil Rights Act.  In order to prove this disparate impact, the EEOC hired expert witness Dr. Murphy to determine what the racial makeup of Kaplan’s applicants were, since Kaplan does not collect racial data.  
It was this directive that had Dr. Murphy design his race-rater test. It is this test that therefore must pass the court’s test — which, in the world of expert testimony, is known as the Daubert test.
Murphy’s Law
“The district court considered every one of the Daubert factors — and found that Murphy’s methodology flunked them all,” the Sixth Circuit wrote in its opinion upholding the U.S. District Court for the Northern District of Ohio’s decision to grant defendants summary judgment and exclude Dr. Murphy’s expert testimony.
Both the District Court and the Circuit Court examined “the checklist for trial courts to use in assessing the reliability of scientific testimony."
“The specific factors explicated by the Daubert court are (1) whether the expert's technique or theory can be or has been tested; (2) whether the technique or theory has been subject to peer review and publication; (3) the known or potential rate of error of the technique or theory when applied; (4) the existence and maintenance of standards and controls; and (5) whether the technique or theory has been generally accepted in the scientific community.”
The courts looked first to see if Murphy’s race-rater test “can be or has been tested,” concluding that Murphy had only “corroborated” a small sampling of race-rater results with actual racial identifications available at a later date. Murphy agreed that doing so in only 57 instances was not enough “to establish the reliability of [his] photo rating methodology.”
Even if the “anecdotal corroboration” was of a large enough sample, the rate of accuracy, which in one instance was 95.7% but in another was only “at least 80%,” was unlikely to satisfy the court’s requirement that the rate of error fall within acceptable parameters.
As for whether the technique has been subject to peer review and publication, the court found that “undisputedly there is none.”
Though the EEOC attempted to cite a heroin trafficking case in which a Kansas court referenced a study that stated that visually deciphering between African-Americans and Caucasians was reliable, the EEOC failed to mention that in that case, it was further stated that doing so for Hispanics was not proven reliable. Nevertheless, Murphy’s race-rater test wasn’t nearly so simple, but instead asked for a five-way racial categorization, including identifying multi-racial persons.
As for the fourth Daubert prong, the district court found a variety of faults in the race-rater standards and controls, including the failure to give the race-raters any proper criteria for racial identification, the fact that Dr. Murphy did some of the race-rating himself, and the inclusion of the pictured persons’ names. The court gave the example of how the surname “Gonzalez” could lead to an identification of that person as Hispanic.
Finally, the court finds that there is no evidence that “race rating” and determining race by visual means is generally accepted in the scientific community.
Do As I Say, Not as I Do
Like Kaplan, the EEOC also conducts credit checks on its employee candidates, because, according to EEOC’s personnel handbook, “overdue just debts increase temptation to commit illegal or unethical acts as a means of gaining funds to meet financial obligations.”
Kaplan agrees, having discovered some years ago that several of its financial aid officers had stolen student funds and several of its executives had engaged in self-dealing. To combat these issues, Kaplan instituted several new strategies, including credit checks on its potential employees.
Kaplan argued that these credit checks were a business necessity related to the candidates’ particular job requirements. Whether that argument would have persuaded the court to rule in favor of the defendants despite there being a disparate impact on minorities is a decision that will not be made in this case.  
Nor will a finding of whether these credit checks had any actual disparate impact on African-Americans or another protected class be made. Nor will the court weigh how much other non-discriminatory variables (such as socio-economic status) could account for any disparate impact.
Furthermore, the question of whether the EEOC will be estopped from filing a claim against an employer for engaging in a practice that the EEOC itself condones will also not be answered.
While employers across America kept a close eye on this case, wondering whether their own credit checks violate some civil rights law, they are still left sitting in suspense.
In a case alleging unfair treatment based on race, the one detail that will certainly be necessary to know is the actual race of the affected people.
This data can be obtained through a variety of ways, but it is now clear that having a group of academics eyeball driver’s license photos to visually determine race will not be an acceptable means of doing so.
Perhaps next time, as the defendants suggested, the EEOC should contact the people in question and ask them directly.
Is there any way expert witness Dr. Murphy could have conducted his race-rater test in a manner that was scientifically reliable and thus satisfy the Daubert standard?

Maggie Tamburro

Maggie Tamburro is an attorney and writer who holds a Juris Doctor from The John Marshall Law School and a Bachelor of Arts from the University of Texas. She was admitted to the Illinois Bar in 1994 and Florida Bar in 1999 and has significant experience in legal research, editing, and writing. Maggie is active her in local community, holding various publicly appointed civic board positions.

Get the best expert

Fill out the form and one of our representatives will be in touch with you shortly. Or, you can call or email us directly.