Page 265 - Becoming Metric Wise

257
Research Evaluation

Units under scrutiny will adapt their behavior and hence the use
of indicators changes the system through the incentives they establish.
In this context one refers sometimes to Goodhart’s Law (Goodhart,
1975). which states that that when a feature is picked as an indicator
it ceases to function as an indicator because people start to game it.
These effects should be anticipated. One way to do this is by using a
suite of indicators and not a single one.
10. Scrutinize indicators regularly and update them.
In order to perform acceptable research evaluation exercises bib-
liometricians should work on definitions, reproducibility, validation
and compatibility (Gla ¨nzel, 1996) of existing and newly developed
indicators. Such indicators should not be a hinder to the develop-
ment of science worldwide but a help to better performance.

8.3 UNIVERSITY EVALUATION
8.3.1 A University’s Mission
Evaluation of universities or its subunits should first and above all take the
mission of the university into account. Nowadays universities have three
major roles to play: research, education and community services i.e., play-
ing a socio-economic role. All three contribute to the definition of
the mission of a university and distinguish it from other institutions.
The third role includes contributing to the economic development and
to the social and intellectual welfare of the region to which the university
belongs. The university’s economic role has led to the idea of a triple
helix (Etzkowitz & Leydesdorff, 1995, 2000) of universities, industry and
governments. From a more humanistic point of view the third mission
includes reaching out to the society by actively contributing to the public
debate on matters of importance to the society at large. Although all these
aspects are expected from the university as an institute, it would be inhu-
man to expect that each member of the university plays these three roles
to the full.

8.3.2 The Ideal University

Related to the mission of a university and the existence of university
rankings, discussed further on, we already note that university evaluation
exercises and rankings are based at least partly on indicators. These indica-
tors and the weights accorded to them should correctly reflect the