When life went through a phase when I knew quite a number of people whose marriages were ending in divorce. I commented to a friend that having got used to DH and all his irritating little ways I really couldn't be bothered to swap him for another model that meant that I would have to get used to a different set of irritating little ways.
I have never had reason to think otherwise and as for having a possible substitute waiting in the wings.......................
Hollydaze you ask 'Also - how can they know that 'more than half of all wives' are doing it? Wouldn't they have to actually ask all wives first?'
The answer is no, they do not. Without going into the depths of statistics it is possible to calculate a sample size that, if the sample is randomly selected, you can calculate with what degree of accuracy any sample of that size will produce the more or less the same result and will be representative of the whole population. I still have, from my market research day a little wheel that told me the size sample I needed for a total population I specified and the degree of accuracy I would accept. It is surprising how accurate quite a small sample can be.
Having said this a lot of the polls carried out on subjects like this are not based on properly drawn up samples, the wording of their questionnaires is dubious and the final results, when they get into the hands of journalists working for the popular papers can be misreported and distorted to get a good story and headline.