Power Calculation, Two Correlations, Z-Test

Power calculation for two independent correlations via Z-Test with option for charts of power vs. sample size (assuming equal sample sizes), sample 1 size, sample 2 size, population one rho value, and alpha level.

General

Element Name Description
Sample One Size Sample size of group one for the power calculation (acceptable range 2 to 150000).
Sample Two Size Sample size of group two for the power calculation (acceptable range 2 to 150000).
Rho1 Population number one correlation coefficient.
Rho2 Population number two correlation coefficient.
Alpha Level Alpha level (Type I error rate for Z-statistic) (acceptable range 0.0001 to 0.25).
Hypothesis type Hypothesis type selection:

 Two-tailed (Rho1=Rho2, population one rho = population two rho);

 One-tailed (Rho1<=Rho2, population one rho <= population two rho);

 One-tailed (Rho1>=Rho2, population one rho >= population two rho)
Computational Method Computational method selection:

 Fisher Z Refined, employ an approximation based on the Fisher transformation, but using refined series approximation formulae for the mean and variance;

 Fisher Z Crude, use the traditional (somewhat cruder) approximation assuming the mean of the Fisher transform of r is the Fisher transform of r and the variance of the Fisher transform of r is 1/(N-3).
Detail of computed results reported Output level of detail selection:

 Power Calculation produces Power Calculation Spreadsheet;

 Pwr Calc plus Charts option produces, in addition, Power vs. Sample Size (assuming equal sample sizes), vs. Sample Size 1, vs. Sample Size 2, vs. Population One Rho, and vs. Alpha Level charts.

Chart Parameters

Element Name Description
Number of Steps. (For charts only) Number of individual steps for which to calculate power between the starting and ending parameter values for Power Charts (Min 4, Max 100).
Starting Sample Size. (For charts only) Initial sample size for calculation of all Power vs. Sample Size charts. (Power Vs N chart assumes individual sample sizes are equal.) (Acceptable range 2 to 140,000).
Ending Sample Size. (For charts only) Final sample size for calculation of all Power vs. Sample Size charts. (Power Vs N chart assumes individual sample sizes are equal.) (Acceptable range 5 to 150,000).
Starting Correlation (Rho). (For charts only) Initial population one correlation value for calculation of Power vs. Population One Rho chart (acceptable range -0.9999 to 0.9999).
Ending Correlation (Rho). (For charts only) Final population one correlation value for calculation of Power vs. Population One Rho chart (acceptable range -0.9999 to 0.9999).
Starting Alpha Level. (For charts only) Initial alpha level for calculation of Power vs. Alpha Level chart (acceptable range 0.0001 to 0.2499).
Ending Alpha Level. (For charts only) Final alpha level for calculation of Power vs. Alpha Level chart (acceptable range 0.0002 to 0.25).