Re: Statisically Insignificant


Richard Ulrich wrote:
On 30 Mar 2006 08:30:44 -0800, "z" <gzuckier@xxxxxxxxxxxxxx> wrote:

[snip, material that this post is ignoring.]

I suppose you could ask, given X lottery tickets total, how many would
you have to buy to have your probability of winning be statistically
signficant, i.e. a less than alpha chance, where alpha = .05 or
whatever. Large number.

z,
This is a restatement of the mis-use of "significant" that
was cited by the Original Poster.

I think I could hear someone say, somewhat reasonably
and idiomatically, "How many tickets do I have to buy
in order to have a significant chance of winning?"
- In this case, 'significant' means *big*.

And you could say to that person,
"What do you mean by significant? 5%? 10%? 75%,
like the Australian syndicate achieved when they
tried to buy up one of every number in a U.S. lottery
that had 'rolled over' to reach a few hundred million?"

As I posted before, there is no connection to
statistical-significance. It seems that 'statistical' is used
as a generic intensifier, either in humor or by mistake.
It no longer relates to statistics.


--
Rich Ulrich, wpilib@xxxxxxxxx
http://www.pitt.edu/~wpilib/index.html

Yeah, but...
I may be befuddled for various reasons but using statistically
significant in the shorthand meaning "significantly different from
zero", and given that the chances of winning are equal to the
proportion of total tickets which you buy, then you could just see
where the CI for proportions for x/total no longer includes zero. No?
Of course, this doesn't address the question of return over investment
at all.

.

Quantcast