What inspired this wish list request? I have a distribution graph of a categorical variable with three levels. The difference in distribution was significant, but two of the levels were so close together that I can't tell visually if their error bars are overlapping or not (see the picture below). Having a numerical value for the error bars would allow me to see if these two levels are different or not.
What is the improvement you would like to see? It would be nice to either have a label on the graph for the error bar ranges or a summary table listing their values.
Why is this idea important? Having this will allow me to know if similar values are statistically different or not.
1 Comment