Understanding Society User Support

Hello there,

I'm working with h_fimnnet_dv, and trying to build a regression model to predict income based on a range of other factors (primarily ns-sec and method of travel to work).

I've read that monthly income is top-coded at +/- £15,000 (https://www.understandingsociety.ac.uk/documentation/mainstage/dataset-documentation/variable/fimnnet_dv).

However, when running a boxplot, I find I've got a load of values above £15,000 p/m (image attached). I'm going to apply my own top coding, in order to reduce the number of variables at this top end.

At the same time, I just wanted to check whether I'm missing something here about how this variable is calculated and the top coding that is applied. I've checked, and I am using UKDS SN6614 (standard EUL).