There are 50 values and the square root of 50 is about 7.
The range of values is 53, so if we used 7 bins each bin would have a span of either 7 or 8.
This is an awkward number and the square root rule errs on the low side, so I decided on a bin width of 5. This gives more bins than I would normally prefer for 50 values, but it turns out to be a good choice because of the outliers that are some way above the majority of the values.
Notice that I've shown the midpoint value underneath each bin 'bins' rather than the range of values. Thus the first bin includes all values from 12.5 to 17.5 with a midpoint of 15:
If you decided on a different number of bins, that's fine.
The main thing is that the histogram reveals patterns in the data that suggest how you can improve your process. In this case the histogram shows that the typical wait is around half an hour but a few patients waited over an hour. Is there a valid reason for these long waiting times, or does it point to a system problem?
