[lit-ideas] Re: Calling all women and someone who understands statistics
- From: "Andreas Ramos" <andreas@xxxxxxxxxxx>
- To: <lit-ideas@xxxxxxxxxxxxx>
- Date: Thu, 6 Oct 2005 17:02:34 -0700
The larger the sample as a percentage of the population, the better.
The percentage (or ratio) of sample to population is not relevant.
400 events is sufficient to draw conclusions with a 5% margin of error, regardless of the
size of the population.
In fact, you might not ever know the size of the population. If you sit by the side of the
road and count the color of cars, as you reach 400 cars, you've got a pretty good sampling,
even tho' you may not know (or never know) the total number of cars. The same hold true with
samples of mosquitos, etc.
BTW, I've noticed that marketing people are immune to numbers. I explain this over and over,
and it's like talking about Kant's Categorical Imperative to a dog. The next day, they've
forgotten everything. Six sales in a morning and they're think it's a pattern.
yrs,
andreas
www.andreas.com
------------------------------------------------------------------
To change your Lit-Ideas settings (subscribe/unsub, vacation on/off,
digest on/off), visit www.andreas.com/faq-lit-ideas.html
Other related posts:
