Accuracy vs. Precision

Back when your blogger was a freshman physics major, slightly over a half-century ago, the lab had a textbook by the late Philip R. Bevington, Data Reduction and Error Analysis for the Physical Sciences. I still own my copy, one of the only textbooks I saved from those days. It cost a cool $5.95 back then. Now, Amazon has the third edition; you can rent a copy until January for $20.61, or buy a used copy for—eek!—$93.24.

Anyway, right on page three, there was a brief discussion of "Accuracy vs. precision". It begins:

There is considerable confusion among students as to the meaning of and difference between the terms accuracy and precision. To add to the confusion, Webster defines them equally. In scientific investigation, however, they are assigned distinctly different meanings which must be kept separate.

And there follows three longish paragraphs describing the difference. I remember reading that section. And then rereading it a couple more times.

Ah, if only Prof Bevington had this xkcd cartoon to illustrate:

[Precision vs Accuracy]

Mouseover: "'Barack Obama is much less likely than the average cat to jump in and out of cardboard boxes for fun' is low precision, but I'm not sure about the accuracy."