Now Playing: Tests for One Poisson Rate (2:16)
The Poisson distribution is typically used to fit count data, such as the number of defects on an item, or the number of individuals contracting a disease in a month. The Poisson distribution is characterized by a single parameter lambda, which represents the mean count.
The Tests for One Poisson Rate procedure in PASS is used to calculate sample size or power for testing whether lambda is less than or greater than a specified value.
In this video, we will demonstrate how to use PASS to determine the sample size needed to achieve the desired power for various values of lambda, when testing against a lambda of 3.6.
First, the Tests for One Poisson Rate procedure is opened in PASS.
We wish to solve for sample size for a ‘greater than’ hypothesis test.
We set Power to 0.9 and Alpha to 0.025.
The Null Rate is set to 3.6.
We examine a range of alternative rates from 4 to 10.
The Calculate button is pressed to generate the results.
The numeric report gives the sample size needed for each of the assumed alternative lambda rates. Much higher sample sizes are needed when the assumed lambda rate is close to 3.6.
The sample size curve gives a visual representation of the needed sample sizes. The needed sample size goes down dramatically until the assumed lambda rate is around 5, at which point it levels off.