7% "unwanted pregancy"
72% wanted or mistimed pregnancy
In 2005, the figures had changed modestly:
11% "unwanted pregnancy"
70% wanted or mistimed pregancy
An unwanted pregancy was defined as follows:
"Right before you became pregnant, did you yourself want to have a baby at any time in the future?" If they said no, the pregnancy was defined as "unwanted." Pregnancies that occurred sooner than the woman wanted were instead classified as "mistimed," according to the federal study.
[T]he National Center for Health Statistics, which surveyed 7,643 U.S. women on that and many other family planning and reproductive health questions in 2002 and early 2003. The surveyed women were between the ages of 15 and 44.
I suspect that these numbers involve less of a change than one might think. I suspect that two social changes drive the unwanted pregancy numbers.
First, there are a significant number of people who now get fertility treatments, have children using those treatments and then after the fact discover that they have gotten pregnant wtihout those treatments. A preganancy like that is likely to be classified as unwanted, as the couple who likely had a couple of kids already and typically are rather advanced in age at the time of conception hadn't wanted to have more children. But, they weren't necessarily terribly opposed to the idea either.
Second, there are a significant number of young women who ten years ago would have assumed that they would have children at some time in the distant future, and hence would have been classified as a "mistimed pregnancy", who now would have personally not been planning to get pregnant at all, now that there are more role models for that kind of life. They didn't want children ever, but much like women in the past who got pregnant at 22 when they planned on getting pregnant at 35, may decide simply to roll with the punches of life and not be really unhappy about getting pregnant, even though the pregnancy was "unwanted".
Of couse, the drop off in abortions could explain about half of the difference in unwanted pregnancies. If it is this could be because "The number of U.S. abortion providers has fallen steadily, from 2,400 in 1992 to 1,800 in 2000.", or it could be because women are less inclined to obtain abortions. Certainly a 25% drop in the number of abortion providers could help explain a 10% drop in the proportion of pregnancies aborted. This, of course, is the controversial question. Are there fewer abortions on a percentage basis because women don't want them, or because they can't get them? The numbers don't tell us.
Indeed, the difference, even with such a large sample size, could contain a significant proportion of statistical fuzziness. The change was a five percentage point change (only people who actually gave birth were questioned, the abortion data is from a separate source). The margin of error in the full sample is +/-1.12 percentage points at the 95% confidence level. But, since only a fraction of the entire sample would actually have gotten pregnant and been able to provide an answer to the key unwanted pregnancy question, the actual margin of error on that question might be significantly greater. In a sample of that size, one would expect only about 10% to be able to give a wanted or unwanted pregnancy answer, and the margin of error for a sample of that size is 3.55 percentage points at the 95% confidence level. The reported change in the percentage of "unwanted preganancies" is large enough that there is almost certainly an increase in the number of "unwanted pregnancies", but whether it is really that larger, or in fact, is either much larger or much smaller, is hard to tell.