Choose Language Hide Translation Bar
Highlighted
pickerins
Level I

Proportions Testing

There is a calculator for proportions testing but is there a quick way to check a proportions hypothesis.

 

Say I want to calculate the 95% Confidence interval for 75 of 100 parts passing as well as compare that to a typical 80%  FPY value is there a way to do that easily?

 

Currently only solution is to create a sheet of 1s and 0s to get the CI and then change the values to nominal to test probabilities. is there something more streamlined?

1 ACCEPTED SOLUTION

Accepted Solutions
Highlighted
KarenC
Super User

Re: Proportions Testing

There are a couple of short cuts...

1. Create a table 2 columns. Column 1 = nominal and has a 0 and 1. Column 2 = continuous and contains your counts of 0s and 1s (25 and 75 for your example). Then use distribution with column 1 = Y and column 2 = Freq. From the distribution red triangle window you can then get CI and run tests of probabilities.

2. Go to Help>Sample Data and then in the teaching resources (bottom right side) under the calculators is an option for hypothesis tests for one proportion. You can use summary data in that calculator to run your tests.

View solution in original post

2 REPLIES 2
Highlighted
KarenC
Super User

Re: Proportions Testing

There are a couple of short cuts...

1. Create a table 2 columns. Column 1 = nominal and has a 0 and 1. Column 2 = continuous and contains your counts of 0s and 1s (25 and 75 for your example). Then use distribution with column 1 = Y and column 2 = Freq. From the distribution red triangle window you can then get CI and run tests of probabilities.

2. Go to Help>Sample Data and then in the teaching resources (bottom right side) under the calculators is an option for hypothesis tests for one proportion. You can use summary data in that calculator to run your tests.

View solution in original post

Highlighted
pickerins
Level I

Re: Proportions Testing

Thank you Karen! Using the Frequency Option made it much simpler! Thank you for the quick response.

Article Labels

    There are no labels assigned to this post.