On 10/30/06, Richard Johnston [EMAIL PROTECTED] wrote:
The confidence interval calculation in prop.test appears to be
incorrect when alternative=greater. The upper limit is always set
to 1.000. Am I missing something?
Like the t.test function the prop.test function returns a one-sided
The confidence interval calculation in prop.test appears to be
incorrect when alternative=greater. The upper limit is always set
to 1.000. Am I missing something?
total=c(250,250)
success=c(55,31)
prop.test(success,total,alternative=greater,correct=TRUE)
2-sample test for