Productivity and mobility in academic research: evidence from mathematicians

Abstract

Using an exhaustive database on academic publications in mathematics all over the world, we study the patterns of productivity by mathematicians over the period 1984–2006. We uncover some surprising facts, such as the weakness of age related decline in productivity and the relative symmetry of international movements, rejecting the presumption of a massive “brain drain” towards the US. We also analyze the determinants of success by top US departments. In conformity with recent studies in other fields, we find that selection effects are much stronger than local interaction effects: the best departments are most successful in hiring the most promising mathematicians, but not necessarily at stimulating positive externalities among them. Finally we analyze the impact of career choices by mathematicians: mobility almost always pays, but early specialization does not.

Appendices

Appendix 1: additional tables

See Tables 8, 9, 10, 11, 12,13,14,15,16,17, 18, 19, 20, 21, 22, 23 and 24 .

Appendix 2: results based on the Impact factor

The data presented in this section are similar to those obtained in other parts of the paper. However, the basic indicators are based on the IF rather than on MCQ. More precisely, the weight attributed to each article is equal to the number of its pages times the IF of the journal where it is published, rather than the square of the MCQ.

Table 19 is the analog of Table 8.

Table 8 Fixed effects of major departments

Table 9 Fixed effects of a selection of major departments, with and without local journals removed

Table 10 Effect of individual and department variables on authors’ impacts and number of articles

Table 11 The determinants of mathematicians’ scientific output, without/with fixed effects, US only

Table 12 Comparative characteristics of articles in different fields

Table 13 Proportions (in %) of international collaborations over time

Table 14 Transition matrix of mathematicians: locations of the mathematicians who have started in a given country

Table 15 Mean impact of mathematicians depending on country of first location and current country

Table 16 Number of “permanent” moves from one country to another

Table 17 Mean impact over lifetime of mathematicians moving from one country to another

Table 18 Size and share of top authors in top departments

Table 19 Fixed effects of major departments, by IF

Similarly, we have the analog of Table 9 in Table 20 with the impact of authors based on the IF.

Table 20 Fixed effects of a selection of major departments, with some journals removed, by IF

We now consider the factors playing a part in a mathematician’s scientific productivity. Table 21 is the analog of Table 10 based on the IF rather than the MCQ.

Table 21 Impact, effect of various variables, based on IF

The analog of Table 11 with the impact of authors based on the IF is in Table 22 . Finally, Table 23 is the analog of Table 7 based on the IF.

Table 22 The determinants of mathematicians’ scientific output, without/with fixed effects, US only

Table 23 Effect of individual variables on mathematician’s output, by IF

Appendix 3: more on the data

Table 24 contain the list of journals used here. With each journal we list the total number of pages published in the sample period, the number of articles, the 2007 M.C.Q., the mean number of pages by article, and the mean number of authors by article. The code used for each journal (in the first column) should make it easy, for those who are familiar with the mathematical literature, to identify each journal.

Table 24 Journals in the database, A–Z