Mathematicians have learned to compare studies from different areas of science

Italian mathematicians have shown that the current method of evaluating scientific articles are equally correct for work performed in various fields. Article scientists, which they describe methods of comparison, accepted for publication in the journal Proceedings of the National Academy of Sciences. Briefly the main provisions of the leading portal Nature News.

The scientific community judge the relevance and weight "of an article on the number of references to it in other articles. However, this figure is very much different in various sciences. For example, among the works published in 1999, article, which cited 100 times to 50 times more likely to have been devoted to the study of biology of development than the development of aerospace technology.

Despite this substantial difference in the results, traces the distribution of the number of references to articles equal in all disciplines. Mathematics is not a curve on the basis of "clean" of citation, but on the basis of normalized data. The number of references to each article divided by the average number of links to an article on the research in this area. This normalization gives a true "share" section. For example, the same article on the development of aerospace technology, which cited 20 times, "is" an article on the biology of development, which cited 100 times.

Normalization of citation can also get the same distribution curves number of links to articles published in different years. That is, scientists can directly compare the "weight" of articles, publication of which has more than one year.

According to an expert in the field of statistics, quoting from Boston University Sidney Redner (Sidney Redner), the work of Italian mathematicians is a "good first step in the development of methodology for creating user-friendly tool for comparing citation (work) in various fields." Redner, however, added that a test comparison is unlikely to be applicable to human subjects.

Despite the convenience of the new test will probably not all scientists accept it. In June the group of mathematicians from three major institutions has produced a report that substantiated the lack of evaluation of scientific work on the basis of their citation.