Calculating Required Sample Size

Select the Quick
tab of the Two
Correlations: Sample Size Calculation Results dialog box to access
options to interactively view the results of sample size calculations
and generate plots of required sample size as a function of quantities
that affect sample size in practice.

X-Axis Graphing Parameters. The entries in the boxes under X-Axis Graphing Parameters establish the start value, end value, and number of steps on the X-axis for the graphs specified under Sample Size Charts.

Start Rho1. In the Start Rho1 box, enter the starting value of rho1 used in the N vs. Rho1 sample size chart.

End Rho1. In the End Rho1 box, enter the final value of rho1 used in the N vs. Rho1 sample size chart.

Start Alpha. In the Start Alpha box, enter the starting value of alpha used in the N vs. Alpha sample size chart.

End Alpha. In the End Alpha box, enter the final value of alpha used in the N vs. Alpha sample size chart.

Start Power. In the Start
Power box, enter the starting
value of power goal used in the N vs. Power

End Power. In the End
Power box, enter the final
value of power goal used in the N vs. Power

No. of Steps. In the No. of Steps box, enter the number of steps, or intervals, to plot on the X-axis in the charts specified in the Sample Size Charts group box.

Computational Algorithm. The Power Analysis module allows you to choose from two algorithms:

Fisher Z Refined. Select the Fisher Z Refined option button to employ an approximation based on the Fisher transformation, but using refined series approximation formulae for the mean and variance (Fouladi, 1991).

Fisher Z Crude. Select the Fisher Z Crude option button to use the traditional (somewhat crude) approximation, based on the Fisher transformation, assuming that the mean of the Fisher transform of r is the Fisher transform of r, and that the variance of the Fisher transform of r is 1/(N - 3).

N to Vary. Use the options under N to Vary to specify which sample size (if any) to vary when computing sample sizes and plotting charts. STATISTICA allows you to compute and plot sample size in two types of situations: (1) where sample sizes are assumed to be equal, and (2) where one sample size is varied while the other is held constant.

Equal N. Choose the Equal N option button if sample sizes are assumed to be equal.

Vary N1. Choose the Vary N1 option button to vary the degrees of freedom for the first sample, while holding degrees of freedom constant (at N2=100) for the second sample.

Vary N2. Choose the Vary N2 option button to vary the degrees of freedom for the second sample, while holding degrees of freedom constant (at N1=100) for the first sample.

Sample Size Charts.
Use the buttons under Sample
Size Charts to produce charts of required N as a function of power
goal, population correlation, or type I error rate.

N vs. Rho1. Click the N vs. Rho1 button to produce a plot of sample size (N) versus the group 1 population correlation. The population correlation for the second group is held constant at the current fixed parameter value.

N vs. Alpha. Click the N vs. Alpha button to produce a plot of sample size (N) vs. type I error rate.

N vs. Power. Click the N vs. Power button to produce a plot of sample size (N) versus power goal.