Over the last few weeks, DCI staff members have written a number of blogs related to OFCCP’s pay equity directive. These blogs have focused on some of the challenges federal contractors will likely face when conducting proactive pay equity research in the context of the Directive. As described earlier, one obvious challenge relates to unit of analysis considerations. Directive 307 endorses the use of pay analysis groups, which are broader than similarly situated employee groupings that were endorsed by the now rescinded Compensation Standards from 2006. As discussed in a previous blog (http://dciconsult.com/the-increased-likelihood-of-false-positives-under-directive-307/), this approach has implications for balancing the likelihood of false positives and false negatives. In addition, this approach may have implications for whether a reasonable regression-based analysis is feasible.
For example, in many organizations different jobs require very different knowledge, skills, abilities, and other characteristics. As such, reward factors that explain pay may vary substantially across different jobs. Perhaps for one job a specific education or certification is needed and rewarded, while for other jobs experience or sales performance are drivers of pay. This situation is a common one. In this situation, combining these jobs into one pay analysis group creates a problem in that some pay factors will make no sense for some jobs. In other words, the model would include pay factors that do not influence pay for specific jobs, and these factors would essentially add noise to the analysis. This situation would make it very difficult to have a properly specified model. This was an issue that the SSEG approach mitigated, because grouping similar jobs allowed for a standard set of pay factors to be included in a regression model to potentially explain pay. A one-size-fits-all regression model for broad groups of jobs simply does not mirror the reality of how most federal contractors pay their employees. It will be interesting to see how this issue plays out in audits. Stay tuned.
by Eric Dunleavy, Ph.D., Principal Consultant, DCI Consulting Group