Well defined criteria are critical for making complex decisions. Yet a revi
ew of the literature shows that criteria selection has not received major a
ttention. This paper presents a framework for guiding practitioners in deve
loping and structuring criteria for decision making on public-sector resear
ch. First, an initial list of criteria is generated based on objectives, av
ailable resources, context, and clients' needs. Second, the relevancy of th
e criteria is judged with respect to the research alternatives to be evalua
ted. Third, the availability of data to apply the criteria is evaluated. Th
e principle is to narrow down a broad list of criteria to the point where t
he research alternatives can be characterised by a minimum number of criter
ia. The procedure is illustrated with the case of Chile's agricultural biot
echnology programme. The paper concludes with a discussion on the benefits
of the approach and the issues to be addressed in future applications. (C)
2001 Elsevier Science B.V. All rights reserved.