Chronology Current Month Current Thread Current Date
[Year List] [Month List (current year)] [Date Index] [Thread Index] [Thread Prev] [Thread Next] [Date Prev] [Date Next]

Re: counts in an interval



At 10:35 AM 7/10/00 -0400, Ludwik Kowalski wrote:
I wish I could honestly say that "it was a typo".

Big hint: Dimensional analysis can save you from "typos" of this sort. rt
has different dimensions from r or t.

According to Poisson formula [g(m,t) = (rt)^m exp(-rt) / m!]
with r*t=11.8*0.2=2.36 one finds:
...
This is reasonable but far from being satisfactory.

The value rt=2.36 gives a chisquare per degree of freedom of about 35. If
my spreadhseet is to be believed, it would be better to use rt=2.77 which
gives a chsqpdf of about 8.6 -- not very nice but a heck of a lot better
than 35.

counting experiments are the ways to explore
the random nature of radioactive transformations

I agree with that statement in principle. A similar thought was expressed
at 11:25 AM 7/9/00 -0700, when Bernard G. Cleyet & Nancy Ann Seese wrote:
this experiment was necessary to assist in the verification of various
theories about decay

****** however ******

I consider such statements to be quite a departure from the previous
statement (06:16 PM 7/8/00 -0400)
MY POINT WAS THAT GEIGER COUNTS ARE HIGHLY RELIABLE SOURCES OF CLEAN DATA.

IMHO, if you start by assuming/stipulating that the Geiger counter is a
highly reliable source of clean data, then you cannot possibly be in a
position to verify the randomicity of various theories....


If you want to study randomness per se, you're better off using the rand()
function rather than a Geiger counter. If you want to study the randomness
of the Geiger counts, then you'd better not start by assuming they are
clean and random.

Put away your flamethrowers, folks: I'm not saying all Geiger counters are
useless or all experiments are useless. I'm questioning whether one
particular setup is suitable for achieving its stated purpose.