Date:         Thu, 2 Jan 1997 00:35:03 GMT
Reply-To:     Jay Weedon <j_weedon@ESCAPE.COM>
Sender:       "SAS(r) Discussion" <SAS-L@UGA.CC.UGA.EDU>
From:         Jay Weedon <j_weedon@ESCAPE.COM>
Organization: Zip News
Subject:      Re: NON PARAMETRICS
Content-Type: text/plain; charset=us-ascii

Mike Rayle <rayle@OPUS.OCA.UDAYTON.EDU> wrote:
 
>We are trying to compute a test of significiance to tell whether
>there is a significant number of successes in a sample.  There were
>12 subjects in the sample.  Four out of the twelve were successful.
>
>Does anyone know of a way to say that 4 out of twelve is
>statistically significantly different from 0 using either
>parametric or nonparametric statistics.
 
I'm not sure that your thinking isn't confused here. If you mean,
"Test the null hypothesis that the population parameter is 0%
success", then clearly H0 has to be rejected, since it doesn't admit
the possibility of there EVER being a success!
 
If you mean something else, please elaborate.
 
Jay Weedon.
 
 
"Just a wee doachandoris,
 Just a wee one an' a'.
 Just a wee doachandoris,
 Afore ye gang awa'!
 There's a wee wifey waitin'
 In a wee bundaben -
 If ye can say tis a braw bricht moonlicht nicht,
 It's a'richt, d'ye ken?"