"Copied citations give impact factors a boost," by Tom Clarke. Nature, 22 May 2003, page373.
Two electrical engineers at UCLA have used a mathematical model to predict thenumber of papers that are cited at least 500 times in a database of high-energyphysics papers. Their model predicted 40 of the 24,000 papers in the database,whereas the actual number was 44. The researchers claim that the modelbolsters their contention that authors often don't read the papers that theycite but instead copy from other papers' reference lists, so that the number ofcitations given for a paper is more a function of chance than of the paper'simportance. A bibliometrics expert disagrees with that conclusion but calls themodel "fascinating, unorthodox and inspiring."
--- Mike Breen