Re: Ask Questions Here!
"There are lies, **** lies and statistics."
Originally Posted by odfrank
You have to be a bit canny to know just what a statistic is telling you. You have to look up the study, see how the data was collected, study how it was processed. You have to critique the study to discover what weaknesses it has, and to form a view of just what it represents, and why.
To do that well you have to be able to make such studies yourself, and then be prepared to treat the issue seriously and give it quite a bit of time. Those are skills usually learned during doctoral studies. It involves what is to most people (me) some fairly tricky maths, and a lot of thinking about how different approaches to collecting and processing data will supply different conclusions, and what that means for the study. If you've never done that (I haven't though I have peeked at the process) I think its fair to say that you often _should_ be confused and bewildered by the effort to discover just what a statistic is telling you. What that means is: you've got something right. There isn't a simple answer to most question about a statistical statement.
The bald, unstudied answer is, as others have said, yes this study suggests that treated vs untreated came out about even.
Quite what you can take from that is tougher - though to me its a heartening start. But, to be accurate, there isn't a simple answer to your question.
You can't conclude from that 50:50 ish result that if you stop treating you will have a 50% chance of success.
And you can't either process any deeper maths that tells you what you're chances of success are.
You can however maximise your chances - and get them very good indeed - by studying the mechanisms involved and going about things in a well planned way. Or you can just get lucky.
PS If my (untreated data) had been included it would tend to slew the results downwards - I took about two thirds losses last winter. But that was deliberate - I was allowing the winter to sort the men from boys in a population largely composed of swarms of varying origins. There may well have been others doing the same thing.
Last edited by mike bispham; 07-24-2013 at 02:10 AM.
The race isn't always to the swift, nor the fight to the strong, but that's the way to bet