NRL News

Digging Deeper into the Data: Analyzing the decrease in the number of abortions, Part 1

by | Feb 5, 2014


By Randall K. O’Bannon, Ph.D. NRL Director of Education & Research

AbortionDeclinePinkreBy now, you’ve seen the major outlines of the latest abortion report from the Guttmacher Institute – 1.06 million abortions, down 13% from 2008 to 2011, with abortion rates and ratios lower than they’ve been in any year since abortion became legal over forty years ago.

While those are the most critical numbers, they weren’t the only ones in the report. We’ll get an even clearer picture of what’s happening with abortion in America if we dig a little deeper over the next couple of days. But to lay the groundwork, let’s make sure we’re clear on the main data points and what they mean.

Understanding Abortion Numbers, Rates, and Ratios

There were, according to Guttmacher, 1,058,490 abortions in 2011. The last time we got this close to a million abortions was in 1975. Guttmacher recorded 1,034,200 two years after Roe.

Understandably, more than a few people get confused when comparing the numbers from Guttmacher and those that come from the U.S. Centers for Disease Control. While useful for other reasons, the CDC totals generally run two or three hundred thousand fewer. Why?

Guttmacher surveys abortionists directly, sending out questionnaires, following up with multiple phone calls, and cross checking their data against info published by state health departments. They admit that there may be some abortions at private practices they miss, but they generally are thought to obtain more complete, reliable estimates than the CDC.

By contrast the CDC relies on state health departments for its figures and several states, including California, the nation’s most populous state, do not report any data to the CDC.

Once a special research affiliate of Planned Parenthood, the nation’s largest abortion chain, Guttmacher is known and trusted by most of those in the industry and is able to get information directly from clinics and abortionists in the states where they operate. This means they do have data from California, New Hampshire, and from many of the smaller offices that the CDC might miss.

It isn’t perfect, and there are fudge factors in Guttmacher’s data. Not every clinic returns the survey. Some don’t return phone calls. There may be an isolated doctor here or there than has ordered and prescribed abortion pills for a couple of patients that doesn’t do it on a regular basis and doesn’t show up on the radar. So, besides looking at state data, which often breaks down numbers by county, Guttmacher also finds out about abortionists who advertise abortions on line, and checks with other abortionists about their local competitors.

The bottom line is that abortionists with any sort of substantial caseload are unlikely to be missed. Generally, Guttmacher knows who the big players are in the business and makes sure they get data from them into their survey.

Every single number represents the loss of life. Drops in the absolute number mean fewer babies dying and that is very important and very encouraging. With context, however, you can appreciate just how significant the progress has been.

This is important, as the raw number, taken in isolation, doesn’t tell us whether abortion is becoming more or less common among American women.

For example, if our population was dropping, abortions would too, even if the rate remained the same. If population increased, the same number of abortions for the following year would actually represent a reduced prevalence.

This is why we look not just at the 1.06 million abortion, but also at the abortion rate. The abortion rate is the number of abortions for every thousand women of reproductive age, ages 15-44. On this score, too, we find very good news.

The abortion rate for 2012 was 16.9. This means that the occurrence of abortion in the population of women aged 15-44 is less frequent than it has been at any time in the last 41 years!

By contrast in 1980, if you chose a thousand women of reproductive age at random, 29.3 would have had an abortion– nearly twice as many. The last time the 2011 abortion rate was anywhere near that low was in 1973, when it was just slightly lower, 16.3.

There are, of course, many different ways to get a lower abortion rate. Anything that reduces the number of pregnancies – abstinence, the use of birth control, some disease or condition that impacts a large portions of the female population rendering them infertile — could, in theory, reduce the number. That itself would reduce the rate even if everything else stayed the same.

To get a better idea as to whether there are real attitudinal and behavioral changes behind these lower numbers, one looks at a different number, the abortion ratio. While the abortion rate measures the general prevalence of abortion in culture, the abortion ratio specifically looks at the likelihood that a woman who is pregnant will abort.

Though calculated somewhat differently by Guttmacher and the CDC, both essentially balance the number of abortions against the number of births. A higher number means more pregnant women are aborting, a lower number means more are giving birth.

According to Guttmacher, there were 21.2 abortions for every 100 pregnancies ending in abortion or live birth in 2012. This is also the lowest ratio since 1973, the first year Roe was in effect. It was 30.4 in 1983 and was as high as 25.1 as recently as 1998.

This is important not just because it means fewer abortions, which we’ve already seen. It also is an indicator that we have fewer abortions not simply because of population shifts or declines, or just because there are fewer pregnancies overall, but because there are real behavioral changes, that pregnant women are more likely to choose life.

And that’s certainly welcome news.

More tomorrow in Part Two about some of the possible factors involved in lowering the abortion numbers rates, and ratios.

Please join those who are following me on Twitter at Send your comments to