Frequencies for first 10,000 rows of region variable in nhamcsed2003 dataset : geographic | region | Freq. Percent Cum. ------------+----------------------------------- northeast | 2,002 20.02 20.02 midwest | 2,496 24.96 44.98 south | 3,291 32.91 77.89 west | 2,211 22.11 100.00 ------------+----------------------------------- Total | 10,000 100.00 by Jean Roth , jroth@nber.org , 28 Mar 2016