On Thursday, October 26th, Jeff Davies and Phil Akroyd presented a webinar titled Disparity Analysis 101. The webinar laid out the basics of disparity analysis, the legal framework, the typical (and often incorrect) approach taken by the OFCCP in audits, the correct way to conduct the analysis, and some recommendations.
Federal Contractors are required to “ensure that all qualified applicants and employees are receiving an equal opportunity for recruitment, selection, advancement, and every other term and privilege associated with employment.” Given these Federal Regulations, organizations should analyze their Applicants, Hires, Promotions and Terminations to see if “a substantially different rate of selection in hiring, promotion, or other employment decision works to the disadvantage of members of a race, sex or ethnic group.” In other words, selection rates should be analyzed to see if one groups is adversely affected in relation to selection.
The presentation highlighted the fact that statistical differences in selection rates open the Contractor to additional questions related to their selections decisions, and when the OFCCP finds these statistically significant selection rates, the Title VII “Burden Shifting” model means that companies then have to explain selection processes and decisions. This often means additional work on the part of the Contractor and can lead to a long and drawn out Compliance Evaluation, or audit.
Biddle Consulting Group advises that analyses reflect reality and therefore the OFCCP and their Compliance Officers should be directed to analyzing the data that way. The OFCCP will often attempt to combine applicant data from separate hiring events in a job group to conduct analysis, as this often increases the sample size of the data, which makes it easier to detect statistical significance.
To reflect reality, it is recommended that data is presented and analyzed in a way that reflects the practices of the organization. This often means analyzing the data by requisition or by job title. If statistically significant results are still detected, the next step is often to conduct component Step Analysis, which can identify a specific section within a selection process that may be generating the adverse impact.
To conduct this analysis, it is critical that Federal Contractors maintain accurate and up to date applicant data, with complete data for each applicant, as well as accurate disposition codes. Disposition Codes not only allow for identification of who is a genuine applicant and should be included in the analysis, but also at which step they fell out of the application process to allow for good quality step analysis, should it be necessary.
Contributing author: Jeff Davies, M.A. EEO/AA Consultant at Biddle Consulting Group, Inc.