LITIGATION PREVENTION FAQS

Frequently asked questions about our role as litigation-prevention and affirmative-action consultants to corporate management:

1. What if one or more protected classes are over-represented in terminations?

2. How can IRI help us minimize the risk of employment discrimination lawsuits in connection with reductions in force ("RIFs")?

3. What's the process like? What do you do and what do we do?

4. How much time does Integral require for pre-RIF testing?

5. Suppose the statistical results show disparate impact. What choices do we have?

6. Suppose the "two standard deviation" tests show no disparate impact. Does this mean that the RIF list would not support a disparate treatment charge either?

7. Can IRI determine whether the RIF list will support a disparate treatment charge?

8. What about the "four-fifths rule"? Is the "two standard deviation rule" enough or should we also be using the "four fifths rule" to test for adverse impact?

9. What if, in spite of all our efforts, complaints of discrimination arise?

10. Does Integral review and monitor hiring practices as well as reductions in force?

11. We'd like to know whether our performance-evaluation system has an adverse impact on protected classes. Does Integral do that kind of testing?

1. What if one or more protected classes are over-represented in terminations?

Over-representation may occur by chance. Unless it is statistically significant under the "two standard deviation rule", over-representation alone does not support a disparate impact charge.

2. How can IRI help us minimize the risk of employment discrimination lawsuits in connection with reductions in force ("RIFs")?

Before any pink slips go out, we determine whether or not the RIF you're planning will provide statistical support for a discrimination charge. We can do this because the Supreme Court has clearly specified the statistical criterion for an inference of employment discrimination. To prevail, plaintiff must demonstrate that the disparity or disproportion complained of is at least "two standard deviations" in width. Working with you, we run the required "two standard deviation" tests on the RIF list before you finalize it. With these results available to you in advance, you can ensure that the RIF you ultimately conduct is statistically problem free.

3. What's the pre-RIF testing process like? What do you do and what do we do?

First we meet with you to review your plans and goals for the RIF. For example, we need to know whether the RIF will be company-wide or limited to particular job groups, departments, divisions or facilities. It is best to include in-house counsel and the key RIF planners in this initial meeting. Second we arrange with your human resources people to obtain a copy of your company's electronic personnel database. Once we have your database on our system, we prepare it for the analyses decided on at our first meeting. Then as soon as the RIF list is ready, we run the tests and get the results back to quickly so you can manage the RIF process effectively.

4. How much time does Integral require for pre-RIF testing?

The actual running of the tests on the preliminary RIF list can usually be done very quickly, within a day or even less. This is because all the preparatory work has been done earlier on. We have met with you and gone over plans and goals you have for the RIF. We have transferred a copy of your electronic database to our system. And we have written the computer programs to do the analyses you decided on. However, the more time, the better. Depending on the results of our first round of testing, you may want to change the RIF list and have us re-run the tests. The only limit to the number of changes and re-runs we can do is time.

5. Suppose the statistical results show disparate impact. What choices do we have?

You have three choices. You can amend the RIF list. We will then re-run the statistical tests on the new list to ensure that no "two standard deviation" problems remain. Or we can proceed to a more advanced analytical technique (called "multiple regression') that allows us to control statistically for differences between employees in entirely objective characteristics, such as educational background or experience, that may have influenced management's selection decisions. Sometimes differences in such characteristics - rather than race or gender -- account for the gross disparities observed. Or, finally, you can proceed with the RIF as planned. With advance knowledge of potential problems, you have time to document the objective business criteria on which your decisions were based.

6. Suppose the "two standard deviation" tests show no disparate impact. Does this mean that the RIF list would not support a disparate treatment charge either?

No. The two charges are very different. An impact charge alleges only adverse results. It says nothing about the firm's intentions. A disparate treatment charge alleges discriminatory intent.

7. Can IRI determine whether the RIF list will support a disparate treatment charge?

Yes. Though the precise statistical procedures differ, courts use the "two standard deviation rule" to evaluate statistical evidence in disparate treatment cases as well.

8. What about the "four-fifths rule"? Are "two standard deviation rule" results enough or should we also be using the "four fifths rule" to test for adverse impact?

The Equal Employment Opportunity Commission and the Office of Federal Contract Compliance Programs in the Department of Labor will always regard "two standard deviation" tests as definitive. In other words, passing or failing the "four-fifths" rule criterion for compliance has no certain implications for these agencies' judgment about your compliance status. Any conclusion as to the existence of adverse impact based on "four-fifths" rule results may be set aside by the agencies if a "two standard deviation" supports different conclusions. And the two types of tests often do yield quite different results.

9. What if, in spite of all our efforts, complaints of discrimination in termination arise?

You are prepared. Prepared for settlement negotiations and prepared to litigate. IRI will assist you at every step.

10. Does Integral review and monitor hiring practices as well as reductions in force?

Yes, we do. Where application forms have been retained for both rejected and hired applicants, we use them for our analysis. Applying "two standard deviation" tests, we determine whether or not there is a statistically significant difference between the percent protected classes represent of those hired and the percent they represented of applicants. When the application forms of rejected applicants have not been retained, we must estimate the proportion protected classes represented of applicants. To accomplish this, we do is called a "labor pool" or "labor market" analysis.

11. We'd like to know whether our performance-evaluation system has an adverse impact on protected classes. Does Integral do that kind of testing?

Yes, we do. And if we find adverse impact, we help you determine whether observed disparities in performance score can be traced to differences in objective characteristics such as level of education or years with the company.

Page 1 of 1