« Physicians' non-compete clause OK in New Jersey | Main | Class Action Fairness Act does not apply . . . »
Four-fifths test is not a safe harbor
April 12, 2005 by Ross Runkel at LawMemo
"A selection rate for any race, sex, or ethnic group which is less than four-fifths (or eighty percent) of the rate for the group with the highest rate will generally be regarded by the Federal enforcement agencies as evidence of adverse impact, while a greater than four-fifths rate will generally not be regarded by Federal enforcement agencies as evidence of adverse impact." So says the EEOC regulation.
How does a Memphis police sergeant to get promoted to lieutenant? First pass a written test (whose cut-off score was never validated), and then move on to three other requirements.
The City had required a passing score of 70, but that resulted in a greater than 4/5 impact on African-American applicants, so the City reduced the passing score to 66. Black applicants who failed the test sued under Title VII claiming race discrimination, using the disparate impact theory.
The City's defense: No disparate impact proved because under the new cut-off score the impact was less than the four-fifths required by the EEOC regulation. In other words, there was some impact but not enough to trigger a requirement that the cut-off score be validated.
The trial court used other statistical methods as evidence of adverse impact. (a) The "T-test" which measures the difference in mean (i.e. - average) scores between minority and non-minority candidates. (b) The "Z-test" which measures statistical success among groups (i.e. - it computes the average scores of the passing candidates for each group and then compares those group averages).
The 6th Circuit (2-1) affirmed the trial court's finding of adverse impact, rejecting the City's argument that disparate impact could not be proved unless the impact satisfied the four-fifths rule. Isabel v. City of Memphis (6th Cir 04/11/2005).
The dissent was quite sure that previous authority in the circuit meant that no other statistical methods could be used if an employer satisfied the four-fifths rule. She thought the City properly "relied on the bright line four-fifths rule" and that plaintiffs should not be allowed to use "the newest statistical flavor of the month."
My view: It's too late in the day to argue that only one method of statistical analysis can be used to prove disparate impact. These cases are won or lost at the trial level, by persuading the court that one or another statistical method is truly indicative of whether the measured impact is merely a matter of chance.
|
Editor: Ross Runkel, Professor of Law Emeritus. email Ross@LawMemo.Com, Phone 503-399-8028. Copyright LawMemo, Inc.
|
