Sir

The letter by Michael J. Larkin about research productivity among young UK scientists highlights a ‘bean counter’ mentality all too common in the academic community, apparently on both sides of the Atlantic (Nature 397, 467; 1999).

While compiling an analysis of the productivity and quality of research of a randomly selected group of US academic researchers funded by a large government agency (n = 60), we noted an almost complete lack of association between the number of years of experience in research and productivity as measured by the mean number of papers published per year (R2 = 0.103; see Fig. 1).

Figure 1
figure 1

There is no strong relationship between experience in biomedical research (as measured by years of publication) and productivity (as judged by the number of peer-reviewed papers published per year).

Full size image

This finding appears somewhat at odds with that of Larkin who showed that, within his cohort, mean numbers of papers increased steadily over the course of several years. This may reflect a difference in the publication patterns of UK versus US academic researchers at different career stages. Even so, Larkin's evaluative criterion of a requirement for ‘excellence’ in the early career publications of the UK researchers in his study was not further defined.

Our assessment, using a quantifiable metric, clearly shows that the quality of research bears no relationship to the number of years of experience a researcher has accumulated in scientific publication (R2 = 0.0007). Put another way, we believe our data suggest that excellence in scientific research is manifested from the beginning of one's research career, and that publication rate (at least in the United States) does not change dramatically over the course of one's career. Therefore, perhaps more attention should be paid to where one's papers are published and less to how many papers are published.