Re: [R] pvalue calculate

2012-07-22 Thread Peter Ehlers
On 2012-07-22 13:09, Henrik Singmann wrote: Hi Mary, I think the good old t-test is what you want: Maybe, but calculating p-values with absolutely no consideration of assumptions is pure folly. It may well be that Mary has some assumptions in mind, but the way the question was posed does not i

Re: [R] pvalue calculate

2012-07-22 Thread Duncan Murdoch
On 12-07-22 3:37 PM, Mary Kindall wrote: I have a value a=300 observation (x) = sample(1:50) How to find a p-value from this. I need to show that "a" is different fom mean(x). Thanks This question doesn't really make sense. sample(1:50) gives you the same sample as 1:50 does, just in a dif

Re: [R] pvalue calculate

2012-07-22 Thread arun
HI, Probably ?pnorm x1<-mean(x) x1 [1] 25.5 > pnorm(25.5,mean=300) [1] 0 A.K. - Original Message - From: Mary Kindall To: r-help@r-project.org Cc: Sent: Sunday, July 22, 2012 3:37 PM Subject: [R] pvalue calculate I have a value a=300 observation (x) = sample(1:50) How t

Re: [R] pvalue calculate

2012-07-22 Thread Henrik Singmann
Hi Mary, I think the good old t-test is what you want: x <- sample(1:50) t.test(x, mu = 300) gives: One Sample t-test data: x t = -133.2, df = 49, p-value < 0.00022 alternative hypothesis: true mean is not equal to 300 95 percent confidence interval: 21.36 29.64 sample

[R] pvalue calculate

2012-07-22 Thread Mary Kindall
I have a value a=300 observation (x) = sample(1:50) How to find a p-value from this. I need to show that "a" is different fom mean(x). Thanks -- - Mary Kindall Yorktown Heights, NY USA [[alternative HTML version deleted]] __ R-he