2M00 Analyzing the Selection Process

a. Identifying and Mapping the Employment Process. The CO must determine how the contractor’s selection process for the job or opportunity under examination works, including identifying all of the steps and decision points involved. The employment process usually consists of decisions and actions an employee or applicant must take such as applying for the job, as well as qualification points such as passing a test or interview in order to move to the next stage of the process. Once the CO identifies the employer’s process, he or she may find it helpful to “map” or create a diagram representing the process and the points or steps in the process where the contractor can eliminate an employee or applicant from further consideration. In addition to mapping the process, the CO will need to gather information to understand when the process was instituted, why, by whom, whether there have been changes in the process and, if so, when and why.

b. Identifying Pass/Fail Points. If an individual must successfully navigate a certain step(s) in the process before proceeding on to the next, that step(s) is a “pass/fail point.” Ideally, the CO can examine each pass/fail point to see if the contractor is disproportionately screening out members of any specific group. However, if the contractor does not eliminate individuals at separate steps throughout the process, but instead weighs their performance at each step to formulate a final decision, then the CO evaluates the process as a whole. To do this, the CO must ask the contractor to explain the importance and weight of each step in the selection decision. If an analysis of the entire selection process shows that the contractor disproportionately screens out members of a specific group, the CO must attempt to determine the particular criteria that are the source of the problem.

c. Lack of Data: If the contractor failed to maintain adequate data as required by 41 CFR Part 60-3, the CO must cite the contractor for this violation. If the contractor’s lack of data hinders the CO’s ability to determine the step or criterion that disproportionately eliminated members of the nonfavored group, then the CO must conduct the analysis based upon the “bottom line effect” or the existence of an adverse impact due to the total selection process. However, under such circumstances, the CO must obtain as much information as possible about the process and make every effort to identify the step or criterion causing the problem. Below is a very basic example of the concept.

Example. A selection process consists of: 1) completing an application; 2) having a high school diploma; 3) taking a written test; and 4) participating in an interview during which the contractor accepts or rejects the applicant. In this example, applicants must demonstrate they have a high school diploma before taking the test, and must pass the test before participating in an interview. For each applicant, the CO must determine the applicant’s race, ethnicity and sex, whether the applicant had a high school diploma, the test results, and the results of the interview including the interviewer’s notes, if any. This information is used when determining the number of members within each group eliminated at each step of the process. The CO must conduct an adverse impact analysis at each stage of the selection process to determine whether adverse impact occurs.

If the contractor’s application process, or any part of the process, involves Internet applicants, the CO must obtain applicant data and information regarding the basic criteria used in determining applicants for the job.164 The CO must discuss with his or her supervisor the appropriate analyses for Internet applicants.

d. Relevant Pools. The relevant pool for each step of the contractor’s selection process includes the individuals who made it to that step in the selection process compared with those who made it through to the next step. In the example in (c), above, if the CO is analyzing disparity in hiring rates, he or she would first consider the pool of all applicants to determine whether the high school diploma requirement is the source of the adverse impact. If the high school diploma requirement does not fully explain the disparity in hiring, the CO would next determine whether the test disproportionately eliminated members of the nonfavored group. Finally, the relevant pool for evaluating the effect of the interview would be all applicants who have a high school diploma and passed the test. Therefore, any difference in their selection rates would be attributable to the interview.