Employment attorney Jonathan Hyman writes an interesting blog post this week concerning statistics:  Why Statistics are So Important in Reduction in Force Cases.  

Attorney Hyman states that "the Older Workers Benefit Protection Act requires all releases and waivers of federal age discrimination claims provided as part of a severance program offered to a group of employees (such as in a reduction in force) to include a written disclosure of the job titles and ages of all eligible individuals selected for the program and all not selected for the the program."  

The EEOC, in its guidance on Understanding Waivers of Discrimination Claims in Employee Severance Agreements, provides the following example of what this disclosure should look like:

Job Title


# Selected

# Not Selected


























As attorney Hyman correctly notes, when "the lone 63-year-old employee in Job Title 1 . . . [decides] whether to sign the waiver or pursue an age claim, the only fact he and his lawyer will have to go on is that within his job grouping, 7 out of the 9 oldest employees were RIFed, including the oldest employee."

Thus, the "raw statistics" will likely be the critical piece of information on which that employee will base his/her decision on whether to sue or accept the severance.    

However, raw statistics alone are insufficient to make a case of disparate treatment or disparate impact.  According to the federal Uniform Guidelines on Employee Selection Procedures (1978), adverse impact occurs when the selection rate for a protected class is less than 80% of the rate for the class with the highest selection rate, which is commonly known as the 80% rule or the four-fifths rule.   

The 80% rule is a fairly rudimentary calculation, and is not the final word on whether unlawful discrimination occurred.  If the 80% rule has been violated, then the employer should use a more statistically significant methodology to determine if there really is a statistically significant adverse impact on the protected group and if the sample size is statistically significant.  A workplace investigator may need to consult a statistician if this issue arises.  

In general, however, when the 80% rule is violated, employers have four alternatives:

  1. Abandon the procedure;
  2. Modify the procedure to eliminate adverse impact;
  3. Validate the job-relatedness of the selection procedure (e.g., validation studies, detailed records, and fact finding about alternatives with less adverse impact);
  4. Justify the procedure as a business necessity.

Thus, a workplace investigator investigating an "adverse impact" claim should conduct some fact finding concerning alternatives 3 and 4–  if the 80% rule is violated.  If the 80% rule is NOT violated, the query may stop there.  However, an investigator could still decide to investigate alternatives 3 and 4 at his/her discretion.