psych (version 2.1.3)

fisherz: Transformations of r, d, and t including Fisher r to z and z to r and confidence intervals

Description

Convert a correlation to a z or t, or d, or chi or covariance matrix or z to r using the Fisher transformation or find the confidence intervals for a specified correlation. r2d converts a correlation to an effect size (Cohen's d) and d2r converts a d into an r. g2r converts Hedge's g to a correlation. t2r converts a t test to r, r2t converts a correlation to a t-test value. chi2r converts a chi square to r, r2chi converts it back. r2c and cor2cov convert a correlation matrix to a covariance matrix. d2t and t2d convert cohen's d into a t and a t into a cohen d. See cohen.d for other conversions.

Usage

fisherz(rho)
fisherz2r(z)
r.con(rho,n,p=.95,twotailed=TRUE)
r2t(rho,n)
t2r(t,df)
g2r(g,df,n)
chi2r(chi2,n)
r2chi(rho,n)
r2c(rho,sigma)
cor2cov(rho,sigma)

Arguments

rho

a Pearson r

z

A Fisher z

n

Sample size for confidence intervals

df

degrees of freedom for t, or g

p

Confidence interval

twotailed

Treat p as twotailed p

g

An effect size (Hedge's g)

t

A student's t value

chi2

A chi square

sigma

a vector of standard deviations to be used to convert a correlation matrix to a covariance matrix

Value

z

value corresponding to r (fisherz)

r

r corresponding to z (fisherz2r)

r.con

lower and upper p confidence intervals (r.con)

t

t with n-2 df (r2t)

r

r corresponding to effect size d or d corresponding to r.

r2c

r2c is the reverse of the cor2con function of base R. It just converts a correlation matrix to the corresponding covariance matrix given a vector of standard deviations.

Examples

# NOT RUN {
 n <- 30
 r <- seq(0,.9,.1)
 d <- r2d(r)
 rc <- matrix(r.con(r,n),ncol=2)
 t <- r*sqrt(n-2)/sqrt(1-r^2)
 p <- (1-pt(t,n-2))*2
 r1 <- t2r(t,(n-2))
 r2 <- d2r(d)
 chi <- r2chi(r,n)
 r3 <- chi2r(chi,n)
 r.rc <- data.frame(r=r,z=fisherz(r),lower=rc[,1],upper=rc[,2],t=t,p=p,d=d,
     chi2=chi,d2r=r2,t2r=r1,chi2r=r3)
 round(r.rc,2)


# }