So, the data is a list of employers by establishment. There can be multiple sites for someone (Starbucks, McDonalds). The original data has 113 columns/fields. However, there are only 4 that really matter. They are account number, year, qtr, and county. The task at hand is finding those records where the the account number is the same for multiple counties. The theory that I am being tasked with finding is where the account number was in one county and then moved to another county. Attached is a sample data set.
The desired result would show that there are 3 account numbers (1234567890, 2345678901, and 1234567777) which come up twice with multiple geographies. In the original data, it is possible for the same account number to show up in the same county. The same person owning 12 starbucks in the same county is the best example.
So my...
