SOME THOUGHTS ON 'SOME THOUGHTS ON RANDOM ERRORS, SYSTEMATIC ERRORS, AND BIASES' BY JOHN L. JAECH

The "thoughts" in this essay arose from my being asked to comment on a paper by John L. Jaech, "Some Thoughts on Random Errors, Systematic Errors, and Biases," prepared for the Journal of the Institute of Nuclear Materials Management. Tempted as I was to broaden the points raised to encompass a wider spectrum of literature and practice, I decided that the focus of my thoughts would be sharper by limiting them to the symbols and concepts appearing in Jaech's paper and its references. The opinions expressed herein are my own; they do not necessarily reflect those of the management or staff of the Los Alamos Scientific Laboratory or those individuals credited in Section XI with helping me clarify my thinking. My primary concern is that certain statistical practices and definitions in the nuclear materials industry do not match up with statistical practices and definitions encountered in other industries. Such mismatches inhibit inter-industrial communication, and clarification within our own industry is difficult to achieve. Innovative statistical methods are necessary and must be encouraged, but their advocates should be certain of their internal and external consistency and that the language in which they are presented does not conflict with "established" terminology.