Page 128 - Becoming Metric Wise
P. 128

119
                                                    Publication and Citation Analysis

              The second one assigns credits as whole numbers or allows for fractions,
              in which case the sum of all fractional credits is one. Table 5.1 illustrates
              this two-fold distinction.
                 Harmonic counting (i.e., giving credit according to a harmonic
              progression) was originally suggested by Hodge and Greenberg (1981).
              This proposal went virtually unnoticed until it was re-proposed by
              Hagen (2008). In this proposal, the author ranked at the i-th place in the
              author by-line receives a credit equal to
                                             1
                                              i
                                         1   1        1                   (5.1)
                                     1 1   1   1 ::: 1
                                         2   3        n
                 Hagen (2010) proposed to adapt this formula in cases where it is
              shown that the last author played a special role, e.g., it is the correspond-
              ing author. In that case the first and the last author could receive an equal
              credit. We note that any method, such as this, that uses the order in
              which authors are placed in the byline is only valid if this order really
              represents the importance of the contribution, with a possible exception
              for the last author. Complete-normalized counting and harmonic count-
              ing are examples of fractional counting: the sum of credits of all coauthors
              always equals one.
                 Giving only the first author a full credit (first author count) is an easy
              method. When, however, alphabetical order is applied, this method makes
              no sense (unless, maybe, for very large units, as when comparing the
              contributions of countries). Lindsey (1980) notes that the straight count
              procedure should be considered a sampling strategy. As such, it should be
              examined in terms of representativeness. On author level this method
              presupposes that the set of papers on which a scientist’s name occurs first
              (including solo-authored papers) is a representative sample of all that
              scientist’s papers. In this context we also mention that Zuckerman (1968)

              Table 5.1 Counting methods
                                    Each contributor      Some contributors
                                    receives some credit  may not receive any
                                                          credit
              Credits are natural   Complete or normal    First author count;
                numbers               count                 Major contribution
                                                            count
              Credits may be        Complete-normalized   Fractionalized major
                fractions, summing    fractional counting;  contribution count
                to one                Harmonic counting
   123   124   125   126   127   128   129   130   131   132   133