Free Republic
Browse · Search
News/Activism
Topics · Post Article

To: StJacques

StJacques, thank you so much for taking the time to look up and explain these statistics. They were very helpful. Yes, thankfully abortion rates are on the decline. The reason why I asked for the stats is that I once attended a conference where a pro-life physician spoke and pointed out the before r v. w, there weren't too many woman in their 20's getting breast cancer and now after r v. w she is seeing a high amount of women getting breast cancer which is usually a disease which older women get and when she does her interview (she's a breast reconstruction surgeon) one of the common points she found out was that, in the women with NO family history of BC, abortion was the only common factor in these young women with breast cancer. Most of these girls had one or more abortions in either high school or college.


28 posted on 06/11/2006 11:48:31 AM PDT by Coleus (Roe v. Wade and Endangered Species Act both passed in 1973, Murder Babies/save trees, birds, algae)
[ Post Reply | Private Reply | To 25 | View Replies ]


To: Coleus
Coleus; First of all; you're welcome and I appreciate the anecdote you related about the breast reconstruction surgeon mentioning the relationship she observed in her own work regarding women with breast cancer who had no family history of the disease who also had abortions later in life. That is a first-hand (primary source) observation of a health-care provider that should carry some weight; at least with everyone besides NOW, NARAL, and Planned Parenthood.

If you would permit me, I would like to make a few more comments about how statistical research can be used to advance our understanding of a possible relationship between abortion and breast cancer.

Since all scientific reasoning is essentially "inferences drawn from repeated observations of phenomena," it is worth noting that there are levels of accuracy in these observations. If a truly reliable study to trace the relationship between abortion and breast cancer were to be conducted it would involve only a modest use of statistical reasoning. I think it would look something like taking one group of women who had abortions in their 20's and tracing their medical history over time, taking a second group of women who did not have abortions and tracing their history over time, and taking a third group whose history of either having or not having abortions was unknown and tracing their history to see if the relationships developed in observing the first two groups held true for the third (the control group). This kind of research involves direct observation of primary sources (the women in the study) and it is always the most reliable because the researchers actually collect the data first-hand, which limits the possibility of errors in what is recorded.

Reasoning by "statistical inference" is not as reliable as the direct observation method described in the preceding paragraph, but on occasion it can be every bit as accurate; especially when the data used is of a high quality. When one attempts to use statistics to explain a relationship there are usually two primary goals in mind; to explain the form and the strength of that relationship as revealed in the data. By the form of the relationship I refer to what we would describe as the "slope of the line" as it would be plotted on a graph with x and y axes. Since you would want to explain incidences of breast cancer (the y value) in terms of the incidences of abortions (the x value) you would expect, if the relationship is "valid" or "meaningful," that in a given population the percentage or number of incidences of breast cancer will rise as the percentage or number of incidences of abortions rise. In other words, you would expect the line to rise from left to right and the closer that slope is to a perfect 45º angle, the more meaningful the relationship becomes. And if the line rises in such a fashion, then you can go on to test its strength by trying to remove those data points from the set of observable data that may contain incidences of other factors known to raise the risk of breast cancer such as family history of the disease, estrogen therapy, contraceptive use, etc. to see if the data points gather more closely to the line you plot on the graph that "averages" (calculates the mean of) their placement. If you can prove that removing other known risk factors from the data causes these data points to gather more closely together (this is revealed by a statistic known as the R2 value) then you have further reinforced the "validity" or "meaningfulness" of the relationship.

So Coleus; to sum it all up, I believe there is enough data already collected by medical statisticians whose content can be analyzed so as to permit reliable statistical inference as to whether the incidence of abortion increases the likelihood of the incidence of breast cancer in women later in life. That statistical inference cannot, and will not, prove that abortion can be a "causative factor" of breast cancer, but it can establish, if the relationships I described in my preceding paragraph are borne out in the analysis, that an abortion earlier in life may be considered a legitimate risk factor for breast cancer later in life, since "risk factors" describe "probabilities" and the probability, at least theoretically, could be clearly shown.

I hope I did not bore you with all of this technical speak, but I noticed that some seem to doubt that the correlation between abortion and breast cancer can be logically demonstrated and, provided that data can be collected along the lines of what I described in my post #25 above, I believe that it can.
29 posted on 06/11/2006 3:13:01 PM PDT by StJacques
[ Post Reply | Private Reply | To 28 | View Replies ]

Free Republic
Browse · Search
News/Activism
Topics · Post Article


FreeRepublic, LLC, PO BOX 9771, FRESNO, CA 93794
FreeRepublic.com is powered by software copyright 2000-2008 John Robinson