Yes point taken about mixed marriage divorces counting, and about comparability with other religions.
With statistics from censuses in general, there are lots of complicating factors to consider. When we see graphs, and tables of numbers, we may naturally tend think that the production of the data was as reliable, accurate and uniform as the reassuringly organised method of its display. Unfortunately there is lots of room for skewed data and misleading comparisons to arise.
1. The format of the census may differ from year to year, including different options, in different orders, and with different prompts. Such apparently small changes can significantly impact results.
2. Religious organisations may encourage identification with their church during a census. For example I read a report that the Mormon church was concerned about the low numbers identifying as Mormon in South American censuses. So they initiatiated a campaign to contact and encourage all inactive Mormons to identify as Mormons in the next census. This no doubt impacts results where such campaigns are conducted. Needless to say not all churches carry out such campaigns, so are we really comparing like with like?
3. Censuses differ hugely from country to country in scope and intent. The UK census is pretty useless at counting JWs because the British census invites all Chrsitan denominations to identify simply as "Christian". Those JWs who wrote "Jehovah's Witness" in the "other" box numbered around 70,000 in 2001. On the one hand that's a lot fewer than the 130,000 publishers JWs claimed in Britain. But on the other hand those JWs had resisted hints in the census form to pick "Christian" as a catch all term. In fact JW s were the biggest "write in" category of "other religion" in the UK. Which may be interpreted as an impressive display of distinctiveness. But ultimately, UK census results are pretty useless for comparing numbers of JWs with other groups or over time.
There's a good discussion of the problem of the "other" category in the UK census here:
It's interesting that the author of that page assumed that the JW "write in" response was "orchestrated". I can see why he would assume it must have been, because it's such a large number, and "Christian" would appear to be an adequate response for JWs to most outsiders. But I don't recall any particular instruction from WT about how to fill in the census. And I think people familiar with JWs and how they think realise that often JWs don't need to be told to act distinctly in such situations. For a significant proportion of JWs it comes naturally to insist on a separate identity, even when being encouraged to identify as "Christian": for about 70,000 of them in the UK in 2001, to be exact.