binomial.all.test: Perform a two-tailed sign test on p values between 0.00 and...

In roblanf/pcurver: Tools to analyse distributions of P-values from the published literature

Description

This function tests the null hypothesis that there are an equal number of p values in two bins: 0.0<=p<0.025, and 0.025<p<=0.05. Note that p values of exactly 0.025 are excluded, because they do not fall in either bin. The test uses a two-tailed sign test. A significant result indicates that you can reject the null hypothesis of an equal number of p values in each bin. Significantly more p values in the smaller bin is expected for collections of p values that contain evidential value. Significantly More p values in the larger bin is consistent with intensive p hacking or publication bias.

Usage

binomial.all.test(p)

Arguments

p	a vector of p values between 0.0 and 0.05 (inclusive)

Value

a list giving the number of p values in each bin, and the p value of the two-tailed sign test.

See Also

binomial.bias.test, binomial.sns.test

Examples

p <- c(0.00001, 0.024, 0.002, 0.045, 0.00003, 0.021, 0.0001, 0.0000948, 0.0000002)
binomial.all.test(p)

roblanf/pcurver documentation built on May 27, 2019, 11:58 a.m.