At what point during the review of compensation matters will the OFCCP perform a regression analysis? According to the OFCCP’s FAQ’s, the OFCCP will conduct a review of compensation data based on a four tier process.
Source: http://www.dol.gov/ofccp/regs/compliance/faqs/emprfaqs.htm#Q18
“A regression analysis will be conducted when there are indicators of potential systemic compensation discrimination problems. OFCCP will perform a regression analysis and provide summary results to the contractor in support of a Predetermination Notice or a Notice of Violation alleging systemic compensation discrimination. The purpose of the regression analysis is to analyze the relationship between the compensation and the legitimate factors that the contractor uses to determine pay decisions, such as work performed, responsibility level, skills and qualifications, education, performance, or other relevant experience. The regression analysis will show whether or not race and/or gender have statistically significant effects on compensation.
OFCCP’s review of a contractor’s compensation practices generally will proceed through the following steps:
Step 1: Analysis of the contractor’s submission in response to Item Number 11 of the OFCCP Scheduling Letter.
Step 2: After assessing the contractor’s submission, OFCCP may ask the contractor for additional compensation or personnel data to determine whether further investigation is warranted.
Step 3: If OFCCP determines that further investigation is required, OFCCP will conduct an on-site review in order to obtain a full understanding of the company’s pay practices.
Step 4: After conducting the onsite review to obtain a comprehensive understanding of the contractor’s pay practices, OFCCP will perform a multiple regression analysis. This is a statistical analysis which will determine whether there are any significant differences in compensation by gender and/or race that are not explained by legitimate factors.”
The OFCCP will evaluate the Item Number 11 data submitted, and while Biddle can’t confirm the exact threshold the OFCCP uses to determine potential disparity, a general theory is that they look for 2% differences in mean pay and a minimum pay difference of $2,000. It’s important to note that any grouping found to fail the 2% and $2,000 threshold can trigger an OFCCP Step 2 request for additional compensation data. What most contractors are seeing is that the OFCCP’s assessments conducted in Step 2 include the requirement of addition data provided in MS Excel format.
Recent audit letters from the OFCCP indicate that contractors are expected to provide the following data:
- Employee ID Number (or a suitable ID for matching purposes, NO Social Security Numbers please)
- Job title
- The Job Title’s corresponding Job Group (for cross-reference purposes)
- Department
- Gender (Coded as either “Male” or “Female”)
- Race/Ethnicity
- Date of Hire
- Date of last change in grade/title
- Current annual base salary or base hourly wage (excluding overtime, bonuses, and incentives)
- Part-time (please indicate average hours worked in a typical week as readily available) vs. full-time status
- Work shift
- Exempt vs. non-exempt status1
- Grade level or salary bank classification
- Employee location 2
- All other factor not previously listed that impact compensation of your employees
The OFCCP may follow-up with additional questions prior to proceeding to Step 3 of this compliance review phase. Contractors should be prepared to address any and all questions, and should have solid explanations to describe differences in compensation.
If you are a federal contractor and are uncertain as to whether or not you are asking the right questions about your compensation practices, and how you will fare in an OFCCP audit, you would be well advised to seek professional assistance.
Additional Resources:
- COMPARE Compensation Analysis Software
- Compensation Analysis Consulting by Biddle Consulting Group
- Compensation Analysis: A Practitioner’s Guide to Identifying and Addressing Compensation Disparities by Dr. Patrick Nooren