On September 5, 2014, Judge D.J. O’Toole of the District Court of Massachusetts delivered the latest ruling on alternatives to reduce adverse impact after a defendant has successfully proven the challenged test is job related and consistent with business necessity (hereafter referred to as simply “alternatives”). The issue is promotion to police sergeant and the case is Lopez v. City of Lawrence. On its face, it’s a loss for proponents of alternatives. However, there is more to this than a simple won/lost balance sheet. Therefore, we’ve decided to cover it in a four-part series. Part 1 below a brief introduction to what we feel is the essential ingredient in this case. Parts 2 through 4 will cover background issues and case law preceding the Lopez ruling, a detailed presentation of that ruling, and summary and discussion of what it all means.
My partner in this endeavor will be Jim Outtz, one of two experts in Lopez and a major contributor to the issue of alternatives in both case law and other venues. He served as expert for the defendant. The expert for plaintiff was Cassie Fields, who favored alternatives similar to what Jim recommended, but unlike Jim, believed that defendant’s failure to use those alternatives constituted noncompliance with the Uniform Guidelines. Although this is a collaborative effort, I (Art) will write this in the first person and refer to Jim’s contributions as they become relevant (which is frequently). Jim will write, edit, and correct where necessary. Our ultimate goal is to start a serious dialogue on the issue of alternatives and we hope that our audience chimes in.
At issue in this case were written exams for promotion to police sergeant administered in 2005 through 2008. Jim’s opinion in this case was that the written tests have utility because they capture knowledge required to be a police sergeant. However, he argued that standing alone, the tests were “minimally valid” because they cannot measure important skills and abilities also required to be a police sergeant (e.g., leadership & decision-making). Interestingly, Judge O’Toole agreed, opining “After consideration of the evidence as a whole, I find and conclude that Dr. Outtz's opinion rests on adequate grounds and is therefore correct: the exams in question were minimally valid”.
However, he also opined that:
The exams, together with an inventory designed to supplement them, satisfied the technical standards for content validity studies. 29 C.F.R. ß 1607.14(C). They addressed a representative sample of the KSAs of the sergeant position. Id. ß 1607.14(C)(1), (4). They were based on job analyses that considered the important tasks necessary to the successful performance of the job. Id. ß 1607.14(C)(2). They took account of prior relevant work experience as well as relevant training and education. Id. ß 1607.14(C)(6).
He concluded that:
There is no doubt that Dr. Outtz also thought that the content validity of the exams could have been improved by the use of additional test elements, such as an assessment center (see infra). However, for assessing validity, the fact that it could have been better does not mean necessarily that it was not good enough to be deemed sufficient. The key question regarding the content validity of a selection method is whether it reliably predicts a candidate's suitability for the job, such that persons who perform better under the test method are likely to perform better on the job. I am satisfied on the evidence that Boston carried its burden of showing that the exams in question satisfied that criterion.
I’ll leave you with a simple thought. How is it possible for a testing procedure not to have alternatives that make it more valid when it is “minimally valid?” It’s almost an oxymoron to suggest that a testing procedure that is minimally valid cannot be improved upon. The key question is whether there is evidence that such improvement would have less adverse impact.
End of Part 1
By Art Gutman, Ph.D., Professor, Florida Institute of Technology, and James L. Outtz
