# [R] prop.test() - need algorithm or reference

Ronggui Huang ronggui.huang at gmail.com
Thu Aug 13 17:01:07 CEST 2009

```The help page of prop.test gives you three references. Isn't it enough?
For example,

Newcombe R.G. (1998) Two-Sided Confidence Intervals for the Single
Proportion: Comparison of Seven Methods. _Statistics in Medicine_
*17*, 857-872.

Newcombe R.G. (1998) Interval Estimation for the Difference
Between Independent Proportions: Comparison of Eleven Methods.
_Statistics in Medicine_ *17*, 873-890.

2009/8/13  <gheine at mathnmaps.com>:
>
> Preparing a paper for a medical journal.
>
> Using the prop.test() function in R (v2.4.0)
>
> to compare two groups' response to data like the following.
>
> A sample of 100 individuals from Population I, 18 with positive readings
>
> from a certain test,
>
>  vs.
>
> A sample of 148 individuals from Population II, 61 with positive readings.
>
>
>
> Results look like this:
>
>
>
> R version 2.4.0 Patched (2006-11-25 r39997)
>
> ......
>
>> prop.test(c(18,61),c(100,148))
>
>
>
>        2-sample test for equality of proportions with continuity
>
> correction
>
>
>
>        data:  c(18, 61) out of c(100, 148)
>
>        X-squared = 13.7676, df = 1, p-value = 0.0002069
>
>        alternative hypothesis: two.sided
>
>        95 percent confidence interval:
>
>         -0.3498963 -0.1144280
>
>         sample estimates:
>
>            prop 1    prop 2
>
>            0.1800000 0.4121622
>
>
>
>
>
> Presumably the p-value measures that the likelihood
>
> that the two populations have the same proportion of
>
> response.  My question is this.  The reviewer of the
>
> paper has asked for a reference on the algorithm used
>
> to compute the p-value. The R Reference Manual is not
>
> clear on this.  Is this a standard algorithm that can
>
> be quoted by name (e.g., "Two-sample T Test")?  I
>
> do note that the manual quotes a 1927 article by E.B.
>
> Wilson.  Is the method of computation explained there?
>
>
>
> Thank you for any assistance you can provide.
>
>
>
> George Heine
>
> gheine at mathnmaps.com
>
> ______________________________________________
> R-help at r-project.org mailing list
> https://stat.ethz.ch/mailman/listinfo/r-help
> and provide commented, minimal, self-contained, reproducible code.
>

--
HUANG Ronggui, Wincent
PhD Candidate
Dept of Public and Social Administration
City University of Hong Kong

```