Most of the information we get in the mainstream media is based in scientific studies.

And scientific studies are expressed in statistics. This article is a brief introduction to this way of providing information.

Statistics are nothing grand – they are just calculating the odds.

For instance if you smoke the odds go up that you will get lung cancer. It doesn’t mean that you will. It just means that the odds go up that you will. We all know people who have smoked lots all their lives and never got long cancer, one possibility is that the odds didn’t come up for them.

Once it is understood that scientific findings are like this it makes things much more understandable. Thing like, “How come if smoking causes cancer, there are all these people who have smoked lots and never get it.” The problem is with that little word “causes”. It gives the impression that this leads inevitably to that. In understanding scientific studies, if you want to use the word ’cause’ at all, then it should be understood as meaning “the odds go up”.

There are then two ways in which the odds are used.

  • The first is: how likely the finding is to be right, and not just a fluke. There are two measures commonly used – one in twenty and one in a hundred. These are often expressed as 0.05 and 0.01.  These mean that there is a one in twenty (0.05) or a one in one hundred (0.01) chance that something occurred by chance.  So when it is said in a study that, “this is significant at the 0.01 level of significance” this means: the chance is only one in a hundred that this finding is a fluke (due to chance).  When it is said that a finding is significant at the 0.05 level then this means that the chance is only one in twenty that this study is a fluke.
  • There is a second way statistics are used. This is to express how much more likely something is to happen. This is much trickier. For instance you may be told that doing something increases your chances of getting a cancer by 50% (ie. makes it more likely that you will get cancer if you do it). Should you worry. Well, maybe and maybe not. It all depends on what the 50% is an increase of. If the chances were a billion to one to start with, then the news that the chances are now five hundred million to one may not bother you. That’s why this use of statistics is trickier – you need to know what is being multiplied: what the odds were to start with.

I hope these couple of points make it easier to understand how health is reported.

Leave a Reply

You can use these tags: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>