Return to Website

Number Watch Web Forum

This forum is about wrong numbers in science, politics and the media. It respects good science and good English.

Number Watch Web Forum
Start a New Topic 
Author
Comment
Andrew Wakefield

Although I completely agree with Wakefield's demise, I am not convinced that the scientists who have feel that science is being shunned recognize that they need to check the mirror.

I worry that epidemiology will be what finds us the "higgs" boson. Scientists will be huddled around the petabytes or zettabytes of data created by the LHC and the indicator that it was found will be a RR of 1.05...

I hope our great scientists aren't so foolish, but the amount of data being collected by the LHC is very large.

Re: Andrew Wakefield

Large amounts of data can be useful if they are reliable. A good example is sequencing of entire species genomes such as humans. The error rate for DNA sequencing is of the order of 1 in 10,000, which can be drastically reduced by repeated sequencing, and detection of polymorphisms from the large number of donors involved. Still, given the size of the human genome a lot of the data will be wrong.

With the LHC, if it is built correctly, you probably only need to see a rare and unexpected event twice to have a high index of suspicion that it really did happen. With epidemiology, there is greater underlying certainty about what is being measured. Even a test for a disease might only have a 90% probability of being right. This is not a problem if you know about it and adjust the sample size accordingly.

Re: Andrew Wakefield

I agree, but I remember hearing that 1 run of the LHC will create enough data to fill up enough cds to go from here to the moon (or something of that order).

I assume that it is all high quality data, I have just seen shimmerings of our beloved friend epidemiology in unexpected places of late. Weasel words creaping into harder sciences.

The 90% accuracy on disease identifications is scary when applied to large populations with suggestions that we scan everyone for everything. If you were to apply those tests to everyone, would it be almost a certainty that everyone would have something?

Re: Andrew Wakefield

I'm more worried by the sudden disappearance of our generous host's commentary about Wakefield.

When JB said he was legally exposed by merely commentating, he may have just tempted the Dark Lord of Libel.

I fear a new round of bending at the begging bowl may be called for at the delight of He Who Clearly Cannot Be Named.

Re: Andrew Wakefield

I have replace the comment. I hope the disappearance was not as sinister as it might seem.

Re: Andrew Wakefield

That the Dark Lord of Libel has not paid a visit is truly wonderful news. Rescue party called off.