Compensation Analysis

The regulations at 41 CFR 60-2.17(b)(3) require Federal contractors to

"… perform in-depth analyses of its total employment process to determine whether and where impediments to equal employment opportunity exist. At a minimum the contractor must evaluate (3) Compensation systems to determine whether there are gender, race, or ethnicity based disparities."

A complete compensation analysis actually consists of five separate analyses:

Median and Mean (In response to OFCCP median analysis)
T-Test Analysis
Regression Analysis
Cohort Analysis
Outlier Analysis

When the compensation analysis is being conducted in anticipation of an OFCCP audit, EEO Logic will also conduct a median or mean analysis of grade levels, to help identify potential areas that likely would surface from OFCCP's analysis. We will also prepare a response to Paragraph 11 of the scheduling letter.

OFCCP has been utilizing a median analysis of compensation to identify pay situations where protected group employees within a grade or unit are paid less than non-protected group employees within the same unit. OFCCP has been utilizing median analyses to identify their focus areas. In contrast with OFCCP's methodology, EEO Logic's analyses center on judicially recognized methods of evaluating compensation.

T-Test Analysis
The initial analysis of compensation differentials for protected (females and minorities) and non-protected (males and non-minorities) employees is conducted through the use of a t-Test statistical analysis for each grade level. This analysis establishes whether or not differentials in compensation between protected and non-protected employees within a grade level are statistically significant.
Regression Analysis
Regression analyses are completed for the same groupings that were identified for the t-Test. Such factors (variables) as time in service, time in grade, time in job, department, education and performance can be incorporated into the regression. The more variables that can be included in the regression, the more reliable the results will be.
The determination of which factors (variables) to include will depend upon how your company administers its compensation system, and the company's ability to extract the information from its database. The regression analysis is used to eliminate potential problem areas and to help further focus the investigation in those areas that have the greatest potential liability.
Cohort Analysis
A cohort analysis is a non-statistical review of comparators (i.e. similarly situated employees) within a focused area. For each area where it has been determined that the differentials in compensation are statistically significant, and where the regression analysis has not accounted for the differentials, a comprehensive cohort analysis can be conducted to identify potential comparators for further investigation. A potential comparator is a non-protected individual who is receiving greater compensation than the protected individual but who has what appears to be less qualifying experience, time in grade, education, etc.
Outlier Analysis
An outlier analysis is a non-statistical review of individuals within a focused pay grade. Normally pay grades have established minimums and maximums for the pay grade. An outlier analysis identifies those individuals who are above the established maximum of the grade or who are below the established minimum of the grade. OFCCP will seek to identify individuals who are outside established parameters for a grade as possible victims of disparate treatment.

Fees for Services

One of the difficulties in establishing a specific estimate as to how long a comprehensive compensation analysis will take (therefore its cost) is the fact that there are many potential variables. Since the t-tests are initially run on each grade level, the number or grades will affect the number of calculations required. Depending on the outcome of the t-tests an unknown number of primary and/or secondary regressions will have to be performed, and based upon the results of the regressions, subsequent cohort analyses may be required. However, a reasonable estimate would place the required time at between twenty and forty hours. If all goes smoothly it could be less; if there are complications (generally with the completeness and accuracy of the data), it could well be more. On average the basic analysis runs around $3,500. One of the keys to reducing your company's cost is to provided data for all of your locations, rather than running the analyses individually.

Additional Services:

View a sample Compensation Analysis Report.

Contact EEO Logic for professional help.