Letter to CRUS regarding Bibliometric Evaluation

[ français | deutsch | english ]

Swiss Association for Research in Information Technology (SARIT)
The President
Prof Dr. Daniel Thalmann
EPFL-VRlab
1015 Lausanne

To the
President of the Conference of Chancellors
of the Swiss Universities (CRUS)
Professor Hans Weder
Post-office box 607
3000 Bern 9

Lausanne, 16.01.2008

CRUS — Project for the analysis of the services of research of Swiss universities based on bibliometric criteria

Mr. President,

We learned that CRUS started a project aiming at collecting bibliometric data in order to analyze the performance of research of the Swiss universities (Project "Bibliométrie 2007").

This project gives rise to the greatest reservations for the undersigned professors of information technologies of the Swiss universities, represented by SARIT [1], for several reasons:

1. The measurement of scientific performance through a bibliometric analysis causes behavior prejudicial to Information Technologies and science itself (by supporting quantity instead of quality, the publication of incremental results with little significance, increasingly frequent appearance of joint authors whose contributions are negligible, and the easiness of crossed citations, to quote only some well-known problems). We also refer in this respect to the article published recently, "Stop the numbers game" of David Parnas [2], one of the most famous researchers in Computer Science.

2. Bibliometric investigations are clearly not very reliable; see the article "For the evaluation of data processing by means of bibliometric analyses" of Friedemann Mattern [3]. In particular, the "frequency of citation" does not account for the quality of the researchers, because

We can add to these criticisms the distortions by the absence of distinction between external and personal quotations, the "trusts or business of citation" and the large lists of authors. The same applies to "impact factor", which, in spite of relevant and repeated warnings (for example, Amin and Mabe 2000 [4]), is still used wrongly today as the key criterion for the evaluation of a person.

3. Measurements of impact using citation indexes of the ISI in the field of Information Technologies are often erroneous. This is because the culture of publication in Information Technologies is very different from that in other sciences. Whereas in other scientific fields, the publications in prestigious journals are more recognized than papers presented in conferences, this is not always true in Information Technologies: the rate of acceptance of contributions in top conferences, even in some workshops, is often lower than in top journals. As ISI quotation indexes refer only to contributions in scientific journals, and not in conference proceedings, their use as factors for evaluation lead to false conclusions.

In addition, some journals in information technologies, which have an excellent reputation, are unfortunately not covered by these indices because of their relatively young age.

The signatories do not oppose an evaluation of the quality of research for the academic staff in Information Technologies. We ask, however, that such be based on criteria of evaluation scientifically recognized and legitimate. Given the current state of knowledge, only an evaluation by experts (peer review) seems to us appropriate to satisfy these requirements ----- certainly not a purely bibliometric analysis. Such an analysis, even carefully carried out, could be at most used to refine the evaluation of an institution or possibly of a researcher in Information Technologies. The blind application of automated bibliometric measurement covering all scientific disciplines in an undifferentiated way would prove seriously harmful for science and information technologies.

To quote Albert Einstein: "Not everything that counts can be counted, and not everything that can be counted counts.".





Prof. Daniel Thalmann
President SARIT
As well as the professors of Information Technologies listed below.


1. All Professors of Computer Science at Swiss Universities are normally members of SARIT.

2. David Parnas (2007). "Stop the numbers game", Communications of the ACM 50, 11: 19-21.

3. Friedemann Mattern (2002). "Zur Evaluation der Informatik mittels bibliometrischer Analyse", Informatik-Spektrum 25, 1: 22-32. Friedemann Mattern (2002). "Bibliometric Evaluation of Computer Science — Problems and Pitfalls" http://www.vs.inf.ethz.ch/publ/slides/Mattern-Bibliometry-SARIT06.pdf

4. Mayur Amin, Michael Mabe (2000). Impact Factors: Use and Abuse. Perspectives in Publishing, No. 1 (October 2000). Elsevier Science. 1-6.


Signatures

Name / Nom Institution
Prof. Andrea Back U. St. Gallen
Prof. François Bavaud U. Lausanne
Prof. Abraham Bernstein U. Zürich
Prof. Hanspeter Bieri U. Bern
Prof. Torsten Braun U. Bern
Prof. Didier Buchs U. Genève
Prof. Joachim Buhmann ETH Zürich
Prof. Horst Bunke U. Bern
Prof. George Candea EPF Lausanne
Prof. André Csillaghy Fachhochschule Nordwestschweiz
Prof. Jacques Duparc U. Lausanne
Prof. Boi Faltings EPF Lausanne
Prof. Pascal Felber U. Neuchâtel
Prof. Pascal Fua EPF Lausanne
Dr. Patrick Furrer Euresearch
Prof. Harald Gall U. Zürich
Prof. Walter Gander ETH Zürich
Prof. Benoît Garbinato U. Lausanne
Prof. Solange Ghernaouti-Hélie U. Lausanne
Prof. Eduard Glatz Fachhochschule Ostschweiz
Prof. Martin Glinz U. Zürich
Prof. Gaston Gonnet ETH Zürich
Prof. François Grize U. Lausanne
Prof. Markus Gross ETH Zürich
Prof. Thomas R. Gross ETH Zürich
Prof. Matthias Grossglauser EPF Lausanne
Prof. Dominik Gruntz Fachhochschule Nordwestschweiz
Prof. Jürg Gutknecht ETH Zürich
Prof. Rolf Haenni Berner Fachhochschule
Prof. Jürgen Harms U. Genève
Prof. Pius Hättenschwiler U. Fribourg
Prof. Matthias Hauswirth U. Lugano
Prof. Michael Hess U. Zürich
Prof. Lorenz Hilty EMPA
Prof. Hans Hinterberger ETH Zürich
Prof. Béat Hirsbrunner U. Fribourg
Prof. Juraj Hromkovic ETH Zürich
Prof. Jean-Pierre Hubaux EPF Lausanne
Prof. Paolo Ienne EPF Lausanne
Prof. Rolf Ingold U. Fribourg
Prof. Gerhard Jäger U. Bern
Prof. Mehdi Jazayeri U. Lugano
Prof. Rolf Jufer Berner Fachhochschule
Prof. Gerhard F. Knolmayer U. Bern
Prof. Jürg Kohlas U. Fribourg
Prof. Dimitri Konstantas U. Genève
Prof. Peter Kropf U. Neuchâtel
Prof. Martin Kropp Fachhochschule Nordwestschweiz
Prof. Viktor Kuncak EPF Lausanne
Prof. Friedemann Mattern ETH Zürich
Dr. Jacques Menu U. Genève
Prof. Hansjürg Mey Prof. em. U. Bern
Prof. Bernard Moret EPF Lausanne
Prof. Peter Müller ETH Zürich
Prof. Thomas Myrach U. Bern
Prof. Oscar Nierstrasz U. Bern
Prof. Jürg Nievergelt Prof. em. ETH Zürich
Prof. Martin Odersky EPF Lausanne
Prof. Renato Pajarola U. Zürich
Prof. Christine Parent U. Lausanne
Prof. Cesare Pautasso U. Lugano
Prof. Christian Pellegrini U. Genève
Prof. Claude Petitpierre EPF Lausanne
Prof. Rolf Pfeifer U. Zürich
Prof. Yves Pigneur U. Lausanne
Prof. Bernhard Plattner ETH Zürich
Prof. Marc Pollefeys ETH Zürich
Prof. Thierry Pun U. Genève
Prof. Lutz Richter Prof. em. U. Zürich
Prof. José D. P. Rolim U. Genève
Prof. Timothy Roscoe ETH Zürich
Prof. Ivo Sbalzarini ETH Zürich
Prof. Helmut Schauer U. Zürich
Prof. André Schiper EPF Lausanne
Prof. Heiko Schuldt U. Basel
Prof. Gerhard Schwabe U. Zürich
Prof. Stefano Spaccapietra EPF Lausanne
Prof. Burkhard Stiller U. Zürich
Prof. Kilian Stoffel U. Neuchâtel
Prof. Thomas Strahm U. Bern
Prof. Peter Stucki Prof. em. U. Zürich
Prof. Sabine Süsstrunk EPF Lausanne
Prof. Nesime Tatbul ETH Zürich
Prof. Stephanie Teufel U. Fribourg
Prof. Daniel Thalmann EPF Lausanne
Prof. Marco Tomassini U. Lausanne
Prof. Christian Tschudin U. Basel
Prof. Ulrich Ultes-Nitsche U. Fribourg
Prof. Serge Vaudenay EPF Lausanne
Prof. Thomas Vetter U. Basel
Prof. Roger Wattenhofer ETH Zürich
Prof. Alain Wegmann EPF Lausanne
Prof. Eric Wehrli U. Genève
Prof. Emo Welzl ETH Zürich
Prof. Peter Widmayer ETH Zürich
Prof. Carl August Zehnder Prof. em. ETH Zürich

[ français | deutsch | english ]