My main published work on unpaid work used the general social survey, but in doing so I discovered both the strengths and weaknesses of that. It's very rich in detail. You can decide, concerning what Ms. Brown said, whether you're going to count shopping as work or not in how you do the analysis. You can have lots of debate on that. There are also things you can't do with that, and certainly in that work we cross-referenced with the census.
In other international comparison work I'm involved in, and particularly in my teaching work, the unpaid work is very central, using the unpaid work as a factor in an analysis of paid labour market inequality that you can do with the census--so the gender wage gap, or penalties in terms of having children, or whatever. In the analysis of paid wage work you can also use summary unpaid work as a variable. You are not necessarily focusing on the unpaid work, but you are taking it into account in analyzing other aspects of inequality. It's very useful in that way. I have used that.