It occurs to me that it would help if we had better stats. I'm not a stat guy but it's clear that the 70 percent figure for black out of wedlock births is dubious in that it reflects a possible rise in out of wedlock births, as well as a possible decline in "in wedlock" births. We should avoid confusing a rise in the relative number of black kids born out of wedlock with their percentage of black births overall. What we need to know is the number of black women who are having children of out marriage versus their numbers in the population. That seems to me much fairer. I've seen the "out of wedlock births per thousand" bandied about for instance. I'm currently fiddling with the National Center for Health Statistics website to get a fix on that number, and also to get some historical and demographic comparions.
Here is what we know: Despite grave reports about the demise of the black family in the post-Civil Rights era, in fact, the number of black women having children out of wedlock in 1996 was at its lowest point in 40 years. Maddeningly, the Times article I linked to provides no stats for 1956, it simply states it as fact, and for now I'm accepting it, given that i comes from a reputable source. If we have any stat guys looking at this or anyone who wants to join me in trying to navigate the NCHS site, please feel free to help. I should add that in the Times article Stephen Thernstrom says we should be more concerned about the 70 percent figure--but if we're going to do that we need to start critiquing black married couples who only zero, one, two kids as hard as we critique black unmarried couples. Fair is fair, no?
UPDATE: Big, big props to commenter Akali for fishing these two reports which look at the rate of childbearing among unmarried women. The basic conclusion is that the birth rate for unmarried black women is--and has been--declining. In 1970 the birth rate for unmarried black women was 96 per 1,000. In 1980, it was 87.9. In 2005 it was 60.6. There is a huge spike in the late 1980s, but the overal trend is clear--the birth rate for unmarried black women has been declining for almost 40 years.
Something else that should add some context to that 70 percent figure which we all love. The birth rate for married black women has declined way more for married black women than it has for married white women. Also, the birth rate for unmarried women overall is on the increase, but that seems to be being driven by an increase among white and Hispanic women. It's also worth noting that the rate for unmarried black women is still waaayyyy higher than the rate for white women, while lower than the rate for Hispanic women.
I was not a statistics major in college. If anyone wants to debunk these or add context, I'm totally open.
UPDATE 2: I also agree that "born in wedlock" seems to be a very crude stat for measuring the quality of life for a child in the home. I'm esepcially suspicious--as Kathy notes in comments below--of this idea of causality, instead of correality.