I need to make a histogram of employees in firms across the country. My goal is to examine the extreme observations and there is of course a lot in the beginning and very few in the end so I can't figure out how to make one histogram give a clear vision of both how large the extremes are and how many there are in the absolute beginning.

My best idea so far has been to just divide the dataset into two, but that surely can't be the best way. What are your ideas?

I have SAS 9.2.


If there is a big gap between the bulk of the data and the extremes I would suggest using a uniform bin size based on the giving an appropriate shape to the first part. You may then have many empty bins until you get to the outliers and then they will show up in the last bin or last few bins. If the scale makes ot difficult to fit on one graph do a cut out of a range of numbers where no data exists and then show the "outlier" bin(s).

  • $\begingroup$ okay. Thanks. But it has to be like bins of 50 to get it clear in the beginning and since it goes to 25.000 they are really just lines. I reckon i'll stick to the cutting strategy then. $\endgroup$ – Henrik May 20 '12 at 20:38

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service, privacy policy and cookie policy

Not the answer you're looking for? Browse other questions tagged or ask your own question.