Understanding Society User Support

Hi,

I have been using the w_callrec datafiles to look at the number of calls that were made to a household, something which is recorded under the variable w_ivtnc. I merged this variable into the w_hhresp datafiles for each wave, but then I noticed something surprising. The number of calls received by a household ranges from, at a maximum, 32, to a minimum of 0. In the dataset documentation for the w_ivtnc variable it is explained that "This indicator assumes a value of zero [if] no calls have been made because the household was withdrawn from the field after issuing". And indeed, in the w_callrec datafiles, the value 0 for this variable is labelled "no call, withdrawn from field". I was assuming that if no calls had been made to a particular household because it was withdrawn from the field, this household would not show up in the relevant w_hhresp file. But this is not the case. There are around two thousand households in each wave which have a record in the _hhresp file (including responses on all variables included in my study) but for whom the _ivtnc variable is 0 - "no call, withdrawn from field". I did a bit of further digging, and it appears that most of these supposedly withdrawn but still present households are located in Northern Ireland. Is there an explanation for this? I had a look if the User Guide mentions anything about this but couldn't find anything (although it is possible I've not looked hard enough!)

Many thanks,

Sanne