Documentation

[RoxanneConnelly]

Hello, I wonder if you have any documentation for how to use these files? Many Thanks

[dncnbrn]

Not yet - probably in a few weeks' time. In the meantime: the basic principle is to generate a random variable with uniform distribution [0,1] against each case, and then use the old occupation and then the random number to determine which new occupation it maps into (the 'from' and 'to' columns create ranges which are derived from the share of matches in the overlapping LFS datasets for each occupation).

[RoxanneConnelly]

Many Thanks

[teresa-anselmo]

Hello, thank you very much for pulling these files together. I appreciate a lot of time has passed, but was wondering if you could provide some details around how the soc90 and soc00 matching was created? Many thanks