Here, we will use the power and sample size tools to find the number of samples needed to observe a difference in the means between two populations.
We will not need to open a dataset for this recipe. Using a standard deviation that has been set to 1, we will discover the number of samples required to observe a 1, 2, or 3 standard deviation difference between two population means.
The following steps will help us find the sample size required to detect differences of 1, 2, or 3 standard deviations between the means of two samples:
The differences are stated in terms of the value of the standard deviation from the results shown in the following screenshot:
To observe a 1 standard deviation difference between the population means of two samples, we would need 17 samples to have an 80 percent chance of observing this difference or 23 samples to have a 90 percent chance of observing the difference. This sample size is for both groups of data.
The options for this test allow us to specify a one-sided test or change the significance level.
3.147.71.94