Understanding Society User Support

Hi,

I'm trying to construct a length of residence variable for several waves from the harmonised BHPS / UKHLS.

From looking at the universe, I think the best way to do this is using the plnowy4, plnowm, mvyr, mvmnth and mvever variables; however, I am a little confused as to which respondents these questions are presented to, and why they aren't available for all waves of the harmonised BHPS/UKHLS. I've looked through the forum but can't find a definitive answer.

As far as I understand:

- I can use plnowy4, plnowm to compute residential length (using date of interview) for those who have been interviewed previously, where they have not lived continuously at a household that hasn't moved, or where the household has moved.
- For new joiners, who have not lived at their address their whole lives I can use mvyr and mvmnth.
- For new joiners who have lived at their address their whole lives, I can use mvever.
- I would obviously carry all of these forward where applicable for unavailable responses in subsequent waves.

My questions:

1. Why are mvyr, mvmnth and mvever not available for waves 2-18 of the BHPS? : e.g. https://www.understandingsociety.ac.uk/documentation/mainstage/dataset-documentation/variable/mvyr. What did the survey do for new joiners in these cases?
2. When the question asks the month / year of moving to the respondent's current address, and where the respondent has not lived continuously at this address, does this apply to the first date when they first moved in, or to the current period of residence?

Would you be able to help with this? Or to point me to a guide that might show better how to approach this? Apologies if I am misunderstanding anything - please say if so.

Many thanks