Randomising data
I recently needed to randomise some data to keep some information secret. The idea was that it looked roughly similar to the real data, but was sufficiently different to avoid any identifying features. After discussing it with the client, it was agreed that… 1/ ID numbers would be mixed around the people in the list. Therefore, they were all real numbers (that therefore matched the rules governing what made up a legitimate number), but they would be reordered at random amongst the people. 2/ Dates would be set randomly between the minimum and maximum dates available. 3/ Strings would become … Continue reading Randomising data