power.indep.cor: Power analysis for two independent correlations

Description Usage Arguments Value Author(s) References Examples

View source: R/power.indep.cor.R

Description

This function can be used to conduct power analyses for the comparison of independent correlation coefficients, as tested via Fisher's z-test. It can (1) determine the power for detecting a significant difference between 2 correlations given N ('post hoc'), and (2) determine the required N to achieve a desired power ('a-priori').

Usage

1
2
power.indep.cor(r1, r2, n1 = NULL, n2 = NULL, power = NULL,
  sig.level = 0.05, alternative = "two.sided")

Arguments

r1

Value of first Pearson correlation coefficient, must be passed.

r2

Value of second correlation coefficient, must be passed.

n1

Sample size of group 1; pass only for 'post hoc' analysis.

n2

Sample size of group 2; pass only for 'post hoc' analysis.

power

The statistical power; pass only for 'a-priori' analysis.

sig.level

The employed alpha level, defaults to 0.05.

alternative

Must be 'greater', 'less', or 'two.sided'. 'greater' (r1 > r2) or 'less' (r1 < r2) will result in one sided tests (to be used if a directional hypothesis exists). Default test is two.sided.

Value

A list that contains all passed and computed values.

r1

passed correlation coefficient r1

r2

passed correlation coefficient r2

q

The effect size. Represents the difference between the z-standardized values of r1 and r2

n1

sample size in group 1, either passed ('post hoc'), or computed ('a-priori')

n2

sample size in group 1, either passed ('post hoc'), or computed ('a-priori')

power

Power to detect a significant difference between the two correlation coefficients r1 and r2, either passed ('a-priori') or computed ('post hoc')

sig.level

The significance level that was employed for the power analysis

hypothesis

Passed argument 'alternative'

Author(s)

Martin Papenberg [email protected]

References

Fisher RA. Statistical Methods for Research Workers. Edinburgh, Scotland: Oliver and Boyd; 1925. Available: http://psychclassics.yorku.ca.

Cohen, J. (1969). Statistical power analysis for the behavioural sciences. New York: Academic Press.

Examples

1
2
3
power.indep.cor(r1 = 0.4, r2 = 0.3, power = 0.8)

power.indep.cor(r1 = 0.4, r2 = 0.3, n1 = 450, n2 = 450, alternative = "greater")

m-Py/cower documentation built on May 21, 2017, 4:41 p.m.