------------------- help for pwrdemo2 -------------------
Draw Curves Illustrating Power ------------------------------
pwrdemo2 draws curves showing the distribution of the null hypothesis, and the alternative hypothesis, showing the rejection region and the area representing power.
By default, when you run pwrdemo2, a dialogue box is brought forward with buttons allowing you to vary the sample size (N), the difference between the mean under the null hypothesis and the mean under the alternative hypothesis (diff) and alpha.
There are also buttons that allow you to show movies which show the impact of N, diff and alpha. You can vary the speed of the movie by altering the movie delay drop down.
pwrdemo2 , n(#) diff(#) alpha(#) nodialog
The options n(#) diff(#) alpha(#) allow you to set the sample size, the mean difference between Ho and Ha, and alpha. If you don't include any options, the default values are...
n(25) - sample size of 25 diff(.25) - the mean difference between Ho and Ha is .25 alpha(0.05) - alpha is set to 0.05 (two tailed)
If you include the nodialog option, the display of the dialogue box is suppressed.
. pwrdemo2 Starts the program and brings up the dialogue box with the default values for n, diff and alpha.
. pwrdemo2 , n(100) Uses a sample size of 100, and the default difference of .25 and alpha of 0.0 > 5 .
. pwrdemo2 , n(100) diff(.15) alpha(0.01) Uses a sample size of 100, a difference of .15 between Ho and Ha, and alpha of 0.01
More Advanced Syntax
pwrdemo2 , n(#) diff(#) alpha(#) t1( ) t2( ) l1( ) <other grap > h options>.
You can use the t1 option to specify the t1title in the graph, the t2 opt > ion to specify the t2title in the graph, and l1 option to specify the l1title in the graph. See help graph for more information on these title options. You can include other graph options to control how the graph looks.
More Advanced Syntax Examples
. pwrdemo2 , n(100) diff(.15) alpha(0.01) t1("My Power Graph") The t1title in the graph will be "My Power Graph"
Statistical Consulting Group Institute for Digital Research and Education, UCLA firstname.lastname@example.org