To add what to Pete wrote, let me first say the nice folks at the Census Bureau seem to be having second thoughts about their low-ball estimate of our numbers. While that number -- 105,000 at-home dads -- is still in this year's press release, they also add that 2 million preschoolers have dad as a primary caregiver. While this stat makes zero sense next to the 105,000 number (and even less sense when you count the 600,000 gradeschoolers that have dad as primary caretaker), it's nice to have it included, to offer a competing view.
(I've discussed before the wide divergence between the Census Bureau's household survey data -- from which the 2 million number comes from -- and the labor market data, where the 105,000 number comes from. Suffice it to say that there hasn't been a big effort on their part to explain that difference.)
These numbers have become my great white whale, and I actually took a chunk of naptime yesterday to pull out the raw data from the SIPP survey that gives us the big numbers to see if I could better estimate the current at-home dad numbers. But I gave up after a struggle ... I'm no demographer. If any of you are, I'd be happy to point you in the right direction.