tMeanTest: t-test for means and standard errors, assuming unpooled...

In jimtyhurst/codesamplerr: Scratchpad of sample code to demonstrate functions in various R packages

Description

t-test for means and standard errors, assuming unpooled variance.

Usage

tMeanTest(mean.x, se.x, mean.y, se.y, alternative = "two.sided")

Arguments

mean.x	Mean of sample of random variable X.
se.x	Standard error of sample of random variable X.
mean.y	Mean of sample of random variable Y.
se.y	Standard error of sample of random variable Y.
alternative	a character string specifying the alternative hypothesis, must be one of "two.sided" (default), "greater", or "less".

Details

Assumptions: The two samples are drawn from normal distributions. The two samples do not need to have the same variance.

Value

List of the 't' statistic and boolean 'isSignificant', which is calculated at the 95% confidence level, i.e. within 1.96 standard deviations of the mean.

Note

Standard error = (standard deviation ^ 2) / N, i.e. the square of the standard deviation divided by the size of the sample.

jimtyhurst/codesamplerr documentation built on Aug. 13, 2021, 8:45 a.m.