In some Organisation for Economic Co-operation and Development (OECD) countries, the interactions of the economic actors are in transition (Inzelt, 2004). A triple helix model of these interactions is emerging, based on a spiral pattern of relations among three metaactors in a society: Industry, University and Government (Etzkowitz & Klofsten, 2005; Marques, Caraca, & Diz, 2006). None of these three meta-actors prevails on the others; each influences its own trajectory of actions while every trajectory is influenced by the others (Leydesdorff & Meyer, 2006). Their roles and interests are "intimately intertwined in a complex combination of financial, intellectual, personal and legal relationships" (Campbell, Koski, & Blumenthal, 2004a, p. 4).
Never has the university had such a place. It has been motivated by external pressures such as the emergence of the knowledge economy within which the university, as a knowledge-producing and disseminating institution, plays a larger role in industrial innovation (Meyer, Sinilainen, & Utecht, 2003; Cooke, 2005; Landry; Amara, & Saihi, 2005); and the steady decline of public funding of research, which increases the competition for public funds (Etzkowitz & Brisolla, 1999). In this context, the university is experiencing its second academic revolution (Jacob, 2000), which leads to the emergence of a third role, beyond teaching and research: the entrepreneurial role (Etzkowitz et al., 2000). The entrepreneurial university is a "key instrument of technology innovation" (Degroof & Roberts, 2004, p. 327).
Two trends in the scientific literature have focused on this new academic environment (Marques et al., 2006). The first is focused on the study of the university's three roles, particularly the horizontal relations among the three meta-actors (Gunasekara, 2005). The second is focused on the institution's internal environment (Kirkland, 2005). In this trend, two orientations were considered: the first is an institutional perspective focused on structure, as implicit and explicit rules of play that define specific constraints and opportunities for actors (Kleinman, 1998). The new academic environment has increased the need for the institutional management of public higher education organisations (Hemlin, 2006b). The university faces formidable legitimacy challenges (Gumport, 2000). Attaining financial freedom will increasingly depend upon the university's willingness and capacity to collaborate with industries and other organisations (Landry, Traore, & Godin, 1996). Consequently, research and entrepreneurial activities, which were until recently seen as quasi-completely individualistic activities, have been increasingly considered as an organisational objective that should be managed accordingly (Fisher & Atkinson-Grosjean, 2002).
There is a growing interest in the way researchers assume their responsibility in terms of research, entrepreneurial and teaching activities (Kreber, 2000; Porter & Umbach, 2001). Consequently, the second orientation is an agent-centered analysis of faculty members' activities--particularly those of researchers, as operators in a highly manipulable environment and subjected to minimal constraint (Landry, Amara, & Rherrad, 2006). However, this orientation seems to correspond only to ivory tower institutions (Etzkowitz & Klofsten, 2005). In an increasing number of other institutions, this orientation does not take into account the growing pressures to which researchers are increasingly subjected (Meyer, Du Plessis, Tukeva, & Utecht, 2005), and prevents them from self-managing in a "completely free environment" (Laperche, 2002).
Each orientation gives only a partial view of the university's complex mechanisms. A more comprehensive understanding of the internal academic environment reveals the need for integration of the different orientations of the literature in a unique conceptual framework (Audretsch & Lehmannb, 2005) that encompasses all the determinants of the researcher's activities and the relation between them. As the literature is disparate, concentrated on different levels of analysis, a complete development of this framework should allow both the mapping and assessment of the existing intellectual territory. A systematic review appears to be the more useful method to manage the diversity of knowledge on a specific academic inquiry (Tranfield, Denyer, & Smart, 2003). Understanding the determinants of the researcher's role has numerous implications both in identifying gaps that can be filled by future research, and in terms of university management (Hemlin, 2006a).
The article begins with a review of the protocols used and the reasoning behind them. A map is then introduced by considering six trends that characterize the field. Researcher activities and the relations between them follow, and a conceptual framework and synthesis of the evidence on the researcher's activities are then presented. The article concludes with an identification of gaps and implications for future research and policy.
A systematic review is a methodological process that identifies, evaluates and analyzes research evidence to synthesize and map it (Kitchenham, 2004; Staples & Niazi, 2007). The systematic review is a defined, methodical way of identifying, assessing, and analyzing published primary studies to investigate a specific research question (Staples & Niazi, 2007). It is based on a rigorous, transparent and reproducible process allowing development of the most complete view of the literature for researchers and policy-makers (Tranfield et al., 2003; Kitchenham, 2004). Undertaking a systematic review is increasingly regarded as a fundamental scientific activity, and its frequency in management is growing (Tranfield et al., 2003).
The basic steps of a systematic review include: 1) identifying the need for a review, 2) developing a research protocol (formulating an explicit research question, fixing inclusion and exclusion criteria), 3) identifying relevant studies, 4) selecting the studies according to the inclusion and exclusion criteria, 5) assessing the quality of retained studies, and 6) summarizing and synthesizing study results (Kitchenham, 2004; Staples & Niazi, 2007; Tranfield et al., 2003).
Inclusion and Exclusion Criteria
This systematic review sets out to answer the following three questions: 1) what are the activities related to the researcher's roles, as well as their conceptual and operational definitions? 2) what are the institutional determinants of these activities? And 3) what are the other determinants of these activities?
Articles were selected if they considered one of the researcher's roles as the primary concept and if they answered one of the three questions. Preliminary research identified other inclusion and exclusion criteria. First, the new environment of universities described seems to specifically characterize the OECD countries (Lach & Schankerman, 2003). In these countries, there is a growing trend of links between science and education policies on the one hand, and economic policies on the other (Laperche, 2002; Pilbeam, 2006). Moreover, according to Teodorescu (2001), a common structure of the determinants of the researcher's role in OECD countries would be applied with difficulty to other countries. This leads to the conclusion that the systematic review should consider only articles based on OECD countries. Furthermore, according to Meyer et al. (2005), Lee and Rhoads (2004), and Baldini (2006), in the mid-1990s and particularly since 1995, there has been a shift toward rapid expansion of university commercialization. This leads to the conclusion that the systematic review should consider only documents studying post-1995 situations. As the systematic review process started in 2007, the review was limited to documents that had been published or available by the end of 2006.
Published, peer-reviewed papers and research reports were considered. Books, dissertations and book reviews were excluded, due to time and resource limitations.
Strategy of Localization
The strategy of paper localization included two subsequent phases. The first phase contained three steps: 1) a systematic computerized search within multidisciplinary (ISI Web of Science) and specialized databases (Academic Search Premier, ERIC, CBCA Complete, Current Contents, Francis, Education abstracts), as advised by an expert librarian; 2) a web electronic search using Google and Google Scholar; and 3) sorting documents extracted from the retrieval system according to explicit inclusion and exclusion criteria.
At the end of this phase, 5,463 documents were identified and reviewed, based on the inclusion and exclusion criteria. After initial sorting based on the documents' titles and abstracts, 5,129 documents that did not meet the inclusion and exclusion criteria were excluded. A thorough reading of the full text of the remaining 334 documents excluded an additional 250. Hence, 84 documents survived the double sorting to be included in the first phase of the systematic review. Eighty-eight percent of these came from three electronic databases: ISI Web of Science, Academic Search Premier, and ERIC.
The second phase also included three steps: 1) the most important journals were selected according to the results of the first phase (Higher Education, Research in Higher Education, Research Policy, Review of Higher Education, Scientometrics), and a manual search made within volumes published between January 1995 and December 2006; 2) the most prominent authors of the literature selected in the first phase were contacted by email (of 12 authors selected, four responded); and 3) 14 documents were subsequently added. All the identified documents were run through EndNote software to identify and eliminate duplicates. Ninety-eight articles were ultimately selected.
General Characteristics of the Literature
The trend of...