Question 83:Dear Jeff Sauro,
We found your website very interesting, and were hoping that you could provide some insight into a question we have concerning a potential application of the Adjusted Wald method. We would like to use this method with sequential sampling to improve % coverage for the desired confidence interval. In our case, we are not measuring human factors but outcomes in biological experiments where the result is either a + or – effect.
We would like to perform an initial experiment with a small sample size of approximately 30. From this experiment, we would like to determine the proportion coverage for a 95% confidence interval, then perform a second experiment based on the results of the first. Is there an approach to determine the number of additional samples required for the second experiment in order to attain the desired coverage, and can the two experiments be combined? We would like to take this approach in order to economize on the number of samples used for the experiments.
Your response would be greatly appreciated. Thank you in advance.