Thursday, August 17, 2006

CFPC: Statistics Don’t Lie…

I have heard it said before that statistics don’t lie, but liars use statistics. An example of this is that one half of all marriages end in divorce. That is true; however, the population used to reach that figure includes individuals who have had three, four, or more divorces. So, one person may contribute five divorces to the overall ratio. This is an unfairly grim forecast to those who are only on a first marriage, especially since every divorce a person has makes them more likely to have another. Therefore, if you are on your first marriage your odds of staying together are much better than fifty percent.

Another example of how statistics can stick you is correlations. A correlation is simply two factors that tend to occur together, or in tandem. Number crunchers can look at these factors together in a population and can find the probability that these factors would co-vary. Correlations can then be used to make predictions…. Or can they? Correlation does NOT equal causation! An example of this is exercise and income. The more exercise you get, the higher your income is. Does being fit CAUSE a higher income? Of course not! The third factor that influences BOTH is higher education. The more you know, the more likely you are to stay fit, and to get a better paying job. Recently a study based on correlations made headline news and it really hit close to home. After looking into it a bit further, I thought the evidence needed some more scrutiny.

Harvard recently conducted a quasi-experimental study (i.e. looked at self-selected groups) in which they found that there was a correlation between heart attack and light to moderate coffee consumption, about two to three cups per day. This flew in the face of other research that has suggested that coffee has some health benefits. In addition, the study indicated that if you were a heavy coffee drinker you were less likely to have a heart attack. That sounds counterintuitive. So coffee is bad for me, but the more I drink the better off I’ll be? Does that make sense? Is the same true for French fries? Let’s add some logic to the equation.

Apparently the individuals who drank coffee moderately also had at least three other factors that put them at high risk for a heart attack. Smoking, French fries, and a sedentary lifestyle are all examples of such factors proven to increase risk of heart attack. So the more of those three factors I indulge in, the higher my chances are of having a heart attack. Aha! That makes more sense. Since the study was done in Costa Rica, it is not surprising that most of the individuals drank some amount of coffee. So what about the heavy coffee drinkers? Shouldn’t they have a higher risk of heart attack if coffee is truly bad for the heart? Good question.

Shortly before this study was released Coffee Pacifica, Inc. (OTCBB:CFPC) announced that they were expanding their wholesale roasted coffee business. They also executed a letter of intent to acquire a coffee roasting company in Denver, Colorado, pending certain requirements are met. It may seem to some that Harvard’s study, which made national news, could halt any expansion for CFPC. The study could be a roadblock; I would say it is more of a bump in the road. In a study based on correlations, your opinion is as good as mine. I’ll drink to that.

No comments: