Flowers, Just to add another perspective, it seems to me that if you are trying to correlate flow data with any kind of biochemical or bulk analysis (eg the total fluorescence of cells in suspension as measured on a fluorimeter or the concentration of a soluble component ) --- then the only valid correlation by flow would be with the arithmetic mean (NOT the geometric mean and NOT the median). This is because the median and the geometric mean both "underestimate" the contribution from outliers whereas a bulk assay evaluates everything in proportion to its actual value. However, I do agree that the median is probably the most robust flow number --- just because it does not respond sensitively to outliers....... Alice Alice L. Givan Englert Cell Analysis Laboratory of the Norris Cotton Cancer Center Dartmouth Medical School Lebanon, New Hampshire NH 03756 tel 603-650-7661 fax 603-650-6130 givan@dartmouth.edu
This archive was generated by hypermail 2b29 : Wed Apr 03 2002 - 11:57:30 EST