Friday, June 18, 2010

Means, Medians, and Modes: Vision Forum Definitely Works the Psychology of Numbers to their Advantage at the Expense of their Followers

... Why I take issue with the misleading nature of the “1 in 60 million” rate of ectopic pregnancies that result in live births as quoted by Vision Forum and Samaritan Ministries. 

Link HERE if you haven't read the previous posts pertaining to Vision Forum, Samaritan Ministries, and their dangerous views on ectopic pregnancy.

I suspect that until I get caught up and do at least a rough chi-square analysis to demonstrate how these calculations work, the probability is actually zero and the actual cases of live births may be more one in some obscene number that’s so big, it has no meaning.  That’s why the probability when calculating these odds is equal to ZERO.

Let’s start with Don Veinot’s blog article which also takes issue with Vision Forum’s position on ectopic pregnancy to demonstrate how such large numbers can become meaningless quickly.

From "The Crux," 17Jun10:

Doug Phillips and Vision Forum take the view that the medical treatment which preserves the life of the mother, abortion, is not an option and cites the case in Canberra, Australia which trumpets the headline Mother Gives Birth to Girl After Nine-Month Ectopic Pregnancy, No Abortion
This may sound persuasive unless you actually read the article which clearly states:
Doctors are calling Durga a miracle baby and say the chances of both mother and baby surviving are a million to one.
Using this as a persuasive argument is like using the true story of Michael Holmes, Skydiving miracle: Man falls two miles (the jump is on video ) to prove that sky diving without a parachute is safe.
Technically, the "million to one" comment is just an expression of speech to convey the rarity of the incidence.  How many of people constitute a million anyway?  How many babies are born in the US?  In the world?  In a year?  In ten years?  In half of a century?  How many people does that add up to be, and how often are these babies born?  That incidence is much much greater than one in one million!

Sometimes numbers get so big that we don't really even understand what they mean.  We can develop ways of understanding very large numbers, and in healthcare, we often create and use an index that adjusts for individual variation.  In monitoring the pumping action of the heart, a person might have a cardiac output of 5 Liters per minute which might be normal for a six foot man of normal weight, but for a woman with a small build who is only five feet tall, a normal output might be only 3.5 Liters or less.  To adjust for these variations, we calculate an index by bouncing the cardiac output number off of body surface area.  Using the cardiac index, I have a better idea of how they are doing without knowing their own personal norm, though I am quite literally still monitoring their cardiac output.  Just as in statistics and in "hemodynamic monitoring," numbers become tools that help us discern reality and understand what is happening to us.  We have a responsibility to properly interpret and use this information properly.  Handling numbers properly should be of an even greater concern for the Christian, especially the Christian pastor or teacher.

Let me explain why these numbers can be misleading.  To arrive at the 1 in 60 million figure, I believe that I understand the physician's comment and where he derived his numbers.  (In a previous post, I gave the Vision Forum author the benefit of the doubt that the physician did state that there were 60-100 such babies that survived an ectopic pregnancy, though I cannot find it documented in the source mentioned.)  If you take all the babies born in the world in approximately 30 years and divide this number by 60, you do come up with a "mean."  But what does this "mean" really MEAN?

Darrell Huff in his excellent book, "How to Lie with Statistics" explains this problem quite well.  The physician has calculated the Arithmetic Mean or what is commonly called an "average."  It is the sum of the values of n divided by n, essentially.  But does this give us the type of information that we need in order to understand the frequency of ectopic births as the physician has done?  NO!  He's given an an arithmetic average.
Looking at the chart above, a picture tells us much more than prose can at this point.  In Huff's book, he describes a person who wants to move to a community, and the prospective person seeking to relocate speaks with someone who seems like a typical representative of the town in question.  But can just one individual give you the kind of information you seek?  If the community representative represents the median, this might prove helpful to you.  He is the true guy in the middle with equal numbers of people below and above him.  But if you likely fall into the category of with the mode (where most people fit), if you pay attention to the man who represents the average or mean calculated by how much money people make divided by people in the community, you will be quite disappointed if you move there.

The figure of 1 in 60 million is a number we can comprehend somewhat.  The city where Vision Forum is headquartered, San Antonio, broke a population of 1 million while I lived there a few years ago.  I can comprehend something of this number of people.  I can then imagine that sixty million is something like sixty cities that are this same size.  Surviving babies from ectopic pregnancy are so rare that I find this framework to be too generous, and the number is as misleading as the man in the figure above that makes 57,000 per year.

As stated in an earlier post, I want to keep investigating the matter to discover just how many documented cases of these births exist, and I will be happy to report and celebrate them.  Though the US still has lousy infant mortality rates in comparison to other countries which is always surprising to me, medicine has made many great strides and advancements while the incidence of ectopic pregnancy rises worldwide.  The statistics may well have changed and I'd be happy to report this here.  But at least the reader here may gain some insight into the problems I find with the way some of these numbers have been handled.

More to come.