This is the first post in what will (hopefully) become a series on the tools used by anti-abortion activists to bend science. My hope is that this information will be used by pro-choice advocates to critically evaluate and counter the massive quantity of bad research produced by anti-abortion interests.

Today's post covers the use of biased participant groups.

Perhaps the most classic example of this trick is a 1994 study that found that 94% of the participants had, at some point, regretted the decision to abort. The big problem here is that all of the participants in this study were women who had made contact with Women Exploited By Abortion (WEBA), Victims of Choice, or Last Harvest Ministries. Furthermore, some of the participants had already been "counseled" by these groups. The only conclusion supported by this research is that 94% of women who contact anti-abortion groups for post-abortion counseling have at some point regretted the decision to abort. Not exactly a shocker.

(Note that support groups exist for mothers struggling with the decision to parent or give up a child for adoption, and one could easily follow a similar procedure to get equally distorted results with these groups.)

Even the author admits that his findings aren't applicable to post-abortive women in general:

Because this is a self-selected sample of the those who had a "bad experience" these findings should not be interpreted as representative of a random sample of all women who have had abortions.

However, when this study is cited by anti-abortion organizations, any mention of the biased sample completely vanishes. As you can see below, these groups are blatantly lying about the implications of the research.

Sozo Life Clinic (Family Life Ministries):

A Survey of Post-Abortive Women Found That:

28% attempted suicide
31% experienced suicidal feelings
60% commented that the decision to abort made their lives worse
94% regretted the decision to abort

Hope for Families:

...what your mother may not realize is that 94% of women who have had an abortion regret their decision.

Twin Falls Pregnancy Crisis Center:

Most women regret abortions. In fact, when Dr. David Reardon, Ph.D., surveyed 260 women who’d had abortions, he found that:

94 percent had regrets about their abortion
28 percent attempted suicide
58 percent lost pleasure from intercourse
63 percent had flashbacks of their abortion

As this example shows, it is extremely important to read about the participants when evaluating research. There are countless ways to manipulate the participant pool to get more favorable results. Off the top of my head, I can think of several:

  • Select women from a conservative church who attend multiple services.
  • Select women who are having to abort a wanted pregnancy.
  • Select impoverished or homeless women.
  • Select women who are currently being treated for depression or anxiety.

Note that there's nothing inherently dishonest about studying abortion outcomes in these groups - the problem lies in extrapolating conclusions about the general population of women from information about smaller, specialized samples. I think it would be very useful to know more about how abortion affects the homeless and conservative church-goers - but it doesn't say anything about women who don't fall into those specific categories.

Finally: when you're quoted statistics that seem wildly inaccurate, demand the source. If someone can't provide any credible information about their data, chances are that something is being skewed - either in the research itself, or in its interpretation. Beyond that, hold a high standard for the information you pass on. A lot of journal articles and research reports are just one Google search away!