People love to throw around numbers and statistics when they try to make a point. Numbers suggest mathematics and impartial scientific study. Mathematics and scientific studies are the domain of smart people. Therefore, what better way to look like you have it all figured out than by merely giving numbers that you found somewhere? All the work is done for you and all of the critical thinking and analysis can be avoided. One problem however is that statistics may do a poor job of proving what is often the wrong point anyway.

Example 1 – The Distortion – 50% of marriages end in divorce!

People use this statistic to push whatever agenda may be aided by this statistic. Whether you are trying to say that the institution of marriage is hopeless, same-sex unions are good, same-sex unions are bad, religion has lost its effect — hey, you can use this for whatever argument you may need. Realize though that this statistic is not exactly correct.

 People look at a particular population and see that there were say, 100,000 weddings in the past year and there were 50,000 divorces in the same time frame. That’s 50%! But it is not exactly as it seems. Many of those 50,000 divorces could be from weddings that happened in prior years, so you are talking about two different pools of people. What if there were 50,000 weddings last year and 100,000 divorces? Does that mean that the divorce rate is 200%? It could be, if you needed to make that particular point for some cause you are pushing for. Another angle that could be argued here is that married people are much more likely to get divorced than non-married people.

Example 2 – The Self Evident Shocker – Studies show that 50% of all doctors in the U.S. graduated medical school in the lower half of their class!

No study is really needed here actually, it has been this way for years in any school. But you could use this to argue that the state of medical education – or statistical analysis – in this country is lackluster.

Example 3 – The Poll Results – 60% of the people think that we should withdraw all or some of the troops from Iraq. Only 16% think we should send more.

This from  a CNN poll conducted by Opinion Research Corporation asking 1,025 adults nationwide. It is not clear how those people would know how many troops should be sent anywhere for anything, especially in light of this:

Pew Research Center for the People & the Press survey conducted by Princeton Survey Research Associates International. Nov. 9-12, 2006. A poll of 1,479 adults nationwide found that 74% think that there is not a “Clear Plan” in Iraq.

I have my own question for these poll participants: How many troops are there in a squad? How many squads are in a platoon? How many people are needed to operate an M1A2 Abrams tank? How many troops do you need to staff a 2-person security post 24 hours a day? In addition, if you are ignorant of what the plan is for Iraq, how can you possibly start making staffing recommendations?

I shudder to realize that there are politicians and “leaders” who base foreign policy – as well as tactics – on the results of these polls. Please tell me that you use the vast resources of knowledge and experience that our military has in order to guide your position.

Yes, math and statistics are often used to further whatever agenda you may have. Sometimes, the numbers are simply wrong. Sometimes, they are correct but misleading. In so many cases though, numbers are simply used to avoid discussing the real issues. Statistics make a great footnote for the “Letter to the Editor” and give the appearance that research was conducted. They turn what could be a thoughtful discussion about a complex issue into over-simplified black-and-white. Further, they can be cleverly employed by those that seek to steer an argument away from one that cannot be won by critical thinking and analysis toward an argument that can be won by often meaningless numbers.

 Decisions should based not only on “facts”, but on the recommendations of those people who understand the context as well.

Don’t be clouded by the authoritative-looking math.