What inspired this wish list request?
The motivation for this request comes from the requirement from regulatory agencies like FDA to obtain confidence intervals using the Clopper-Pearson (C-P) "exact" method or other methods like the Score (Wilson) Method with continuity correction that could properly handle small sample sizes. For example, it is common to calculate CI for subgroups in a study, some with small sample sizes where using such calculation methods become particularly important. That said, the C-P method is considered (too) conservative so it would make sense to apply less conservative yet acceptable calculation methods like the Score method with continuity correction or other similar methods.
The request for having CI by C-P echos similar requests already on the wish list.
Clopper Pearson Confidence Interval for One Proportion as an Option in Distribution Platform
Confidence Intervals for SMALL sample sizes using Clopper Pearson Beta Quantile
Confidence interval for proportions in Contingency Analysis
What is the improvement you would like to see?
I wish this platform would have suggest several options for CI calculations to choose from or maybe apply several methods and show all results in a table.
Why is this idea important?
Currently, the Distribution platform offers only the Score method with no continuity correction, which is less acceptable in some areas. This leaves no other choice but making manual calculations when wanting to have CI using other methods.
The distribution analysis serves a basic step for many other analyses, which makes these additions even more important.