A shiny app to visualise how the effect size and number of observations affect the power of a t-test.
The first panel show the expected distribution of two a control and treatment group with a specified difference in means and standard deviation. The uncertainty of the means can also be shown.
The second panel shows, in the upper pane, the distribution of t-values under the null hypothesis. The red portion of the curve show values that exceed the 95 These represent false positives (type 1 errors). The lower pane shows the distribution of t-values under the alternative hypothesis. The red portion of the curve show values that exceed the 95 These are true positives. The area under the curve in the red region is the power of the test.
t_test_power_app()if (FALSE) { # \dontrun{
t_test_power_app()
} # }