Re: [tlug] How much of radiations measured in Central Tokyo?

["Stephen J. Turnbull" <stephen@example.com>]

On Wed, Mar 16, 2011 at 12:46 AM, Josh Glover <jmglov@example.com> wrote:

> An excerpt:
>
> "The study, published in the Lancet, examined Danish male jet cockpit
> crew flying more than 5,000 hours. The researchers estimated that such
> crew members receive up to nine mSv a year. Out of 3,877 crew, 169
> developed cancer, compared to 153 in a similar-sized sample of
> non-pilots."
>
> I'm no statistician, but I wonder if the sample size if large enough
> for the result to be significant.

Assuming exactly the same size sample of non-pilots, it's not large enough
to be significant at levels acceptable to economists.  Cheating by assuming
that 153/3877 is the population proportion for non-pilots, the z-value is 1.32,
with a significance level of .093 (economists like to see .05 or lower).  (The
correct test for difference in means would have an even higher significance
level.)

Still, the odds that there really is a difference should be about 10:1.