Learn R Programming

vcmeta (version 1.5.0)

se.biphi: Computes the standard error for a biserial-phi correlation

Description

Computes an estimate of a biserial-phi correlation and its standard error using the frequency counts from a 2 x 2 contingency table where one variable is naturally dichotomous and the other variable is artifically dichotomous. The biserial-phi correlation approximates a point-biserial correlation between the naturally dichotomous variable and the unobserved quantitative variable that was measured on a dichotomous scale. A biserial-phi correlation biserial-phi estimate and the standard error from this function can be used as input in the meta.ave.cor.gen function in a meta-analysis where a point-biserial correlation has been obtained in some studies and a biserial-phi correlation has been obtained in other studies.

Usage

se.biphi(f1, f2, n1, n2)

Value

Returns a 1-row matrix. The columns are:

  • Estimate - estimated biserial-phi correlation

  • SE - standard error

Arguments

f1

number of participants in group 1 who have the attribute

f2

number of participants in group 2 who have the attribute

n1

sample size for group 1

n2

sample size for group 2

Examples

Run this code
se.biphi(34, 22, 50, 50)

# Should return:
#               Estimate        SE 
# Biserial-phi:  0.27539 0.1074594


Run the code above in your browser using DataLab