Cprop.test: Test for proportions of one or two samples

Description

Performs hypothesis testing and confidence interval for a proportion or difference of two proportions. The values of the samples necessary to perform the function are the number of successes and the number of trails.

Usage

Cprop.test(ex, nx, ey = NULL, ny = NULL, p.null = 0.5,
           alternative = c("two.sided", "less", "greater"), conf.level = 0.95,
           ...)

Arguments

numeric value that represents the number of successes of the first sample (see Details).

numerical value representing the total number of trails of the first sample.

(optional) numerical value representing the number of success of the second sample (see Details).

(optional) numerical value representing the total number of trails of the second sample.

p.null

numeric value that represents the value of the population proportion or the difference between the two population proportions, depending on whether there are one or two samples (see Details).

alternative

a character string specifying the alternative hypothesis, must be one of "two.sided" (default), "greater" or "less". You can specify just the initial letter.

conf.level

confidence level of the interval.

…

further arguments to be passed to or from methods.

Value

A list with class "htest" containing the following components:

statistic

the value of the test statistic.

parameter

number of trails and value of the population proportion or the difference in population proportions.

p.value

the p-value for the test.

conf.int

a confidence interval for the proportion or for the difference in proportions, appropriate to the specified alternative hypothesis.

estimate

a value with the sample proportions.

null.value

the value of the null hypothesis.

alternative

a character string describing the alternative.

method

a character string indicating the method used, and whether Yates' continuity correction was applied.

data.name

a character string giving the names of the data.

Details

So that the contrast can be made must be fulfilled that at least 1 hit. That is, in the case of a sample ex must be greater than or equal to 1 and in the case of two samples, ex or ey must be greater than or equal to 1.

Furthermore, for the case of a sample value p.null must be strictly positive.

Examples

Run this code

# NOT RUN {
## Proportion for a sample
Cprop.test(1,6) # 1 success in 6 attempts

#### With a data set: proportion of cars not manufactured in US
data(cars93) #data set provided with the package
exitos<-sum(cars93$USA == "nonUS")
total<-length(cars93$USA)
Cprop.test(ex=exitos, nx=total)

## Difference of proportions
Cprop.test(1,6,3,15)
  # Sample 1: 1 success in 6 attempts
  # Sample 2: 3 success in 15 attempts

#### With a data set: difference of proportions of cars not manufactured in US 
#### between manual and automatic
exitosx<-sum(cars93$USA == "nonUS" & cars93$Manual == "Yes" )
totalx<-sum(cars93$Manual == "Yes")
exitosy<-sum(cars93$USA == "nonUS" & cars93$Manual == "No" )
totaly<-sum(cars93$Manual == "No")
Cprop.test(ex=exitosx, nx=totalx,ey=exitosy, ny=totaly)
# }