Jurisdictional Thresholds determinations
I am curious as to the language used to determine if your company is a Federal Contractor. For AAP Coverage under EO11246 it states you are if you have 50+ employees + "A" contract of $50,000 or more. Same for Disability Section 503 and VEVRAA. What if you have several contracts that add up to $50,000 or greater (similarly $150,000)? Is this specific to one contract or the dollar amount of the combination of all Federal contracts?
Part of the OFCCP compliance is notifying our vendors and suppliers of our AAP status and letting them know they must comply as well with it.
Do you know what the best way is to determine who truly should be notified? We have tons of vendors and suppliers some of which are like Amazon, Staples, etc. that I don’t think this would really apply to.
What's the best way to make a distinction?
Historical Data for different time period than your employee snapshot data
I was always taught that you run an AAP by selecting the snapshot date and then pulling the historical actions for 1 year back from that date. I can find no legal or regulatory item outlining that the snapshot date and historical year must match up this way.
Can a Company have a snapshot date of, say, January 1, 2017 for example and have their historical dates be April 1, 2016 through March 30, 2017 rather than the traditional January 1, 2016 through December 31, 2016. This means the Year End monitoring or Q4 would not be completed until 3 months after the snapshot date. This seems wrong to me but I can find no regulatory reason that would not allow this. I might also add that the narrative date (AKA "Plan Date") is April 1, 2017.
The simple answer to your question is "No." The personnel activity data should have dates that are attached to the snapshot date.
There are multiple reasons for this. The simplest is that it is illogical to have data sets that don't correspond to each other. You're trying to measure outreach activities and potential disparities via your personnel activity data, and you're trying to measure demographic distribution within the current workforce via the snapshot data. There should be a link between what the demographic distribution looks like, the outreach activities that may have produced that distribution, and the possible selection issues preventing your organization from achieving a more equitable distribution.
The second reason the personnel activity data should correspond to the snapshot data is that the itemized listing received when an organization undergoes a compliance review requests "Data on your employment activity (applicants, hires, promotions, and terminations) for the immediately preceding AAP year..." (See item 18.) Thus, if you have a January 1, 2017 AAP, the itemized listing is seeking personnel activity information for the period from January 1, 2016 through December 31, 2016.
Finally, the third reason the personnel activity data should correspond to the snapshot data is that federal contractors and subcontractors are required to update their affirmative action plans on an annual basis. The Executive Order 11246 regulations require an annual review of personnel activity data, while the regulations regarding protected veterans and individuals with disabilities require an annual analysis of a series of data metrics. One could make an argument that these analyses could cover non-corresponding period (there could be an April 1 snapshot in an AAP with personnel activity data for January 1 through December 31), but there would then be problems with analysis of personnel activity being constantly out of sync with the annual update to the snapshot-related reports such as the organizational profile, the comparison of incumbency to availability, and the disability utilization analysis.
The bottom line: you might be able to argue that there is no specific, formal requirement in the regulations that the snapshot data should match the period for the personnel activity data, but OFCCP will have very strong objections to this and you will miss an opportunity to review your data in the same manner in which OFCCP expects to review your data.
Must an applicant apply to all 5 positions if they are exactly the same?
If multiple positions are posted, all the same job, same days worked and same shift, for example we need 5 people to work Monday - Friday 9am-5pm to file. All 5 positions are exactly the same. If an applicant applies to one, can they be assigned to all? It is our understanding under OFCCP we must list all available positions but not all applicants attach their application to all as they believe they have applied it. If we get 10 applicants to the one posting, can they all be attached to all 5 of the same positions?
It doesn't actually make sense to produce five separate postings if you expect to hire individuals for the exact same job where the individuals will start on the same date (or approximately the same date), work the same shift, and so on. Candidates will be confused by this kind of posting, and you will potentially have issues as to who should be considered for an opening if a viable candidate applies to only one of the five openings.
Further, any statistical analyses of applicant and hire data may be problematic. If you truly have five of the exact same openings with the exact same start date, the collective pool of candidates should be compared to each to determine whether outreach activities were successful and whether there were any selection disparities. By posting five separate openings, you may have candidates applying to only one or two of the openings while they are effectively considered for all five openings.
The formal required in the protected veterans regulations (which is where there is the most explicit required to list open positions) is to "list all employment openings." Positions must be listed with the local state employment service office (i.e. the Employment Service Delivery System office) in "any manner and format permitted" by that office that will allow for priority referral of veterans. (See 41 CFR 60-300.5.) Note that the requirement to list open positions does not say that federal contractors and subcontractors must "separately list all employment openings" or "individually list all employment openings." Thus, listing five positions together where the substantive requirements for the job are the same and the start date is the same seems to be acceptable under the federal affirmative action regulations (so long as the local state employment service office allows that kind of listing).
If you have five of the exact same positions with the same (or at least a very close) starting date, it makes sense to list the position once, note there are five openings on the listing, collect information on candidates, and then evaluate the candidates against each other. Ultimately, you will be looking for the effective of outreach efforts on this entire pool and you will be looking for whether there were disparities involving all persons hired vs all candidates.
PLEASE NOTE that this is NOT an encouragement to list dissimilar positions together. It is very important to list different types positions separately. Dissimilar positions should not be listed together because of the importance of analyzing job qualifications against candidates expressing interest in open positions. If you list five openings for "production" or "office" positions, and then try to evaluate candidates against each other, there will be many questions about who was qualified and appropriately selected for the different kinds of jobs that may constitute "production" or "office" positions.
Please also note that if you are listing multiple positions, but these positions will have staggered start dates, you may want to list these positions separately. If there are staggered start dates where, for example, one new employee starts in June and another starts in October, there may be different pools of viable candidates for these jobs because persons who express consider for the June positions may no longer be interested in the October positions.
What is a good strategy to reduce time of position not filled with extensive background checks
How many people can be selected and offered a position based on successfully completing the next phase of the hiring process? Many of our positions must go through a long background check (6-12 months) and then at times people do not make it through that part of the process. If we have for example a position can we select 3 candidates from that one posting and offer the job based on successfully completing the background review? The first candidate that is successful begins in that position. As the others clear they are offered the same job but possibly a different shift time. Is this compliant? Do the other candidates have to formally apply to the other shifts or are they considered internal transfers?
To wait for one person who falls out in month 8 of the background check to start again creates a tremendous deficiency in people actually staffing open positions.
Nothing in the OFCCP regulations limits the amount of offers that a company may make. Of course ensuring the potential outcomes for offers is prudent under other laws and regulations. Careful record keeping and proper dispositioning of the applicants, offers and any changes to the offer would be very helpful should your company be audited. While the process appears facially neutral, it is a good idea to take a look to ensure no particular group is impacted by any changes to the offers. (I don't see anything in your question that would indicate any issues, but not knowing anything more about your process, it is worth mentioning as I expect the OFCCP could take a look at impact in an audit.) It is important to make the regulations fit with your business practices and from the facts in your question, the practice seems compliant.
Adverse Impact Population Pools
When comparing Promotions/Employees, is it appropriate to add selections back into the overall employee pool?
For example, say we have 500 active employees when our data snapshot is taken. During the plan year, 50 employees are selected for promotion. Should the promotions be added into the employee pool? I feel like this would falsely inflate the population by counting those employees twice.
Some federal contractors use "beginning of year" data to establish the pool from which promotions are selected. This avoids having to add the employees back into the denominator. However, if you are only using a single "end of year" snapshot, the successful promotions should be deleted from the job group into which they were promoted, and added to the job groups from which they came. This avoids the artificial inflation problem you identified.