cohend {iCAMP}

Cohen's d effect size

Description

This function is to calculate the popular effect size index Cohen's d.

Usage

cohend(treat, control, paired = FALSE)

Arguments

treat

a numeric vector. treatment group.

control

a numberic vector. control group.

paired

logic. Whether the samples in treatment and control groups are paired. default is FALSE.

Details

This function computes the value of Cohen's d statistics (Cohen 1988). The effect size magnitude is performed using the thresholds proposed by Cohen (1992), i.e. |d|<0.2 "negligible", 0.2<=|d|<0.5 "small", 0.5<=|d|<0.8 "medium", |d|>=0.8 "large". The variance of the d is calculate using the conversion formula reportead at page 238 of Cooper et al. (2009): ((n1+n2)/(n1*n2) + .5*d^2/df) * ((n1+n2)/df) Its square root is output as standard deviation of d.

Value

A list of values will be returned

d

Cohen's d value, (mean(treat)-mean(control))/sd

sd

standard deviation of d

magnitude

a qualitative assessment of the magnitude of effect size

paired

whether the samples are paired

Note

version 1: 2016.2.12

Author(s)

Daliang Ning

References

Cohen, J. (1988). Statistical power analysis for the behavioral sciences (2nd ed.). New York:Academic Press

Cohen, J. (1992). A power primer. Psychological Bulletin, 112, 155-159.

The Handbook of Research Synthesis and Meta-Analysis (Cooper, Hedges, & Valentine, 2009)

Examples

x=c(1,5,8)
y=c(2,6,10)
cohend(x,y)
cohend(x,y,paired=TRUE)